Slashdot Mirror
The End of Unix?
Unix may never become big on the desktop, but that battle is still being fought and it probably won't be over for a few years. However we shouldn't forget where the large strength of Unix lies: the network. Unix -runs- the Internet. I don't think any other Operating System can say that. The Internet started on Unix, the Internet was built on Unix and unless something better comes along (and that implies that we don't have "better" yet) the Internet will die running Unix.
Of course Unix, like any other modern OS, must change over time to accommodate new technologies and methodologies, but I see Unix being more able to adapt in todays fast changing Information Technology world than other operating systems based on monolithic kernels.
What do you think? Am I missing something? Is there a Unix killer in the works that I might have missed?
It's called GREED.
Click the little banners and make them come up.
Click the little banners and make them come up.
Click the little banners and make them come up.
Well, click the fucking banner ads you sheep!
Dudes: I thought in 2035 or something like that the number of milliseconds or something in the kernel hits maxlimit and the date freaks out/cycles... reborn? or maybe just apply patch to fix?
-Too lazy to register.
The funny thing is, that's true for Windows 3, but not DOS. Maybe that particular version (2.0) is dead, but later versions are still in use, and they're only superficially different. DOS outlived Windows 3. It will outlive Windows 9x and NT too.
In case you haven't noticed, Windows NT has eaten a good chunk of Unix's traditional market share. Not to say if that's a good thing, just a true thing.
Commercial Unix vendors are only still in business because they've moved up market and started competing with mainframes. They're far more interested in moving hardware than moving the "Unix Philosophy". Good for profitibility, bad for building a large user base. Thank goodness for Linux -- it's the only thing that gives unix any mindshare nowdays.
I really had an interesting talk with one of my professors a couple of days ago and pretty much found that all the major universities are using Windows type development models for their CS programs. Essentially I was faced with a rather unpleaseant concept. Basically I could be forced into buying a new machine just to do standard coding.
Some universities have been using MS development tools since at least '90 when I was a CS major. It really sucked to have to try to run VB on my 486 by the time I was done with it. But most people here are right. Unix mainframes and even the Mac will never go away. They each do something that neither a PC or Microsoft will never be able to handle.
Unix as a file/application server
Mac for graphics
and mainframes for data IO
Now if we can just get linux onto my grandmothers desktop I'll be happy.
Les
Oh, you've got your MCSE. So does my mom. --My boss to a guy we were interviewing
Basically, UNIX is like that really great knife you have that has lased years and years, with only 6 new blades and 7 new handles - i.e. It continues being called UNIX as it evolves into somthing that's markedly different from what it was before. I predict we'll some day be using an *ix that has Plan 9's process private namespaces, eros-os's capabilty security, Amiga (and the Amiga rip-off called Be) high-speed message passing and data streaming. And we'll still say it's a UNIX variant.
We certainly will. Even as Unix has had its guts replaced multiple times, it has always maintained implementational simplicity. You can teach bright students the internals of a Unix system inside a semester. You can't do that with any other mainstream OS in the market; they have too many layers of cruft. So sure, the Unix of the future won't look (miteimasu?) like the Unix of today, but you'll still be able to fit your head around it.
> I think that Unix will be replaced, if only
> on the corporated desktop. Thin clients will
> replace them due to their stability, security,
> and cost. The system admins and architects to
> whom I've spoken say that it's the "wave of
> the future." They say that they won't need to
> manage thousands of desktops, just a few servers
What do you think is going to run on those clients? Why not a unix flavor with everything but swap mounted from network unix servers?
If something breaks, you just bring out another machine, slap it on their desk and you're done.
Therefore any OS that wishes to supplant Unix will need to supply the same kind of protection for users' information.
Well, yes but no... any such OS will definitely need to address the issue, of course (which I believe was your point here). It won't necessarily have to use the same old UNIX permissions setup (user, group, other, chmod/chown, etc.), and that may be one of the things that makes it a better OS. I agree that there's nothing on the immediate horizon, but things like this often seem to sneak up.
Capability-based systems (access control lists for each resource, and all that) still have some theoretical advantages that could make future OSes more secure -- or so I'm told, anyway (I'm not a security or capability list expert). It definitely seems that there are "traditional" things like this in UNIX that don't have to be done the way they are, and once someone comes along with an OS with capability-based security, migrating processes, a voice interface, automatic buzzwording of your press releases, or what have you it might be better than (and fundamentally incompatible with) UNIX.
I've never heard of this kernel (Adam Mark's PassArch Kernel) Does anyone have more information?
'round these parts, Taco Bell is paying $9.50/hour to start. That ain't chump change.
(gotta go drop my chalupa now)
Just because Apache's market share continues to climb does not make the OS's market share climb.
One reason people used to believe that Unix would kill off the mainframes is that a few years ago mainframes were hideously over-priced. When IBM cut the price (and the margin) and moved to more standard air cooled hardware, the 'frame business picked up.
Another issue was Y2K. Some people chose to dump their legacy systems and start over with Unix client-server systems. To do this properly, they had get started by at least 1996-7. Those that didn't start replacement in time ended up fixing their mainframe apps, which turned into expanding their mainframe apps, which turned into buying more mainframes.
Oh, so now its the STOCK PRICE which is important. Why did dumbfuck.com's stock QUADRUPLE in price the first day it was released?
Must mean its better than Microsoft or Redhat.
BTW, having a market share has nothing to do with the retail price of your product.
This is nothing more than what any reasonably abstract system can handle. Mach and Be are not special in this respect. You only think it's remarkable because Microsoft makes such crap OSes. They lowered everyone's expectations.
It doesn't follow that Unix, as a whole family, will die. People said stuff like that about mainframes. Did they go away? Noooooooooo MVS and CMS are still going strong. MVS TSO is not a very friendly interface, even to command line commandos who use Linux. But MVS is very robust. We crow about uptimes of 3 or 4 months( or more ). MVS system uptimes are typically measured in years. Some big websites are using MVS muscle. It ain't purdy but it's a real workhorse. I doubt it's going away anytime soon and it's much older that unix, et. al.. So why should I think unix is going away?
The obituaries for all other OSes will more than likely be typed up on a Unix box in ed. ITYM, cat > filename or better, cat > /dev/rdsk/c0t0d0s0, and do your own filesystem "stuff". That's how Real Programmers do it (*grunt* *grunt*)
I agree.
UNIX has always survived and will continue to do so. Even in the world of distributed processing (scattering tasks across many machines) UNIX will continue to thrive. There are very few operating systems that have been tested as much and that is what really counts.
These comments are dedicated to Gary Kildall. Without him who knows where we'd be today.
Just upgrade from that old 32-bit box to a 64-bit machine. Surely it'll be feasible by then. It's almost feasible NOW.
Yep, 2038 will finally end the 80386's reign of terror. See you in 38 years.
In unix, you can't know if running a program as root is a security problem without seeing the source to the program. An OS like eros uses a key system so that a program doesn't need to have unlimited access privilidges just to say, run in full screen.
Considering the core of any unix system is a few hundred well-understood system calls, I believe that the future lies in unix/posix compatibility layers over some more advanced operating system (EROS for example). What fun to be able to easily compile and run old code and also explore new and better operating system concepts all in the same environment!
Anonymous Coward asks: "Paper has lived well beyond the era in which it was born (the era of the papyrus reed) and has survived and thrived in the era of the printing press, but now, many people believe that soon we will see the transition from that era into the age of the microcomputer. In that case, has the Slashdot community at large ever considered what the future is? Will paper finally die off, will it adapt as it did before, or will paper find a way to remain the same trustworthy system it always has been? And if paper will come to an end, what does the Slashdot community feel will be its succesor? " (Read More)
Bell Labs has been and gone where this contributor is just now thinking about going. Methinks contributor has not kept abreast of OS developments such as Plan 9.
I dunno, I once had a pointy haired boss that was an embittered old JCL programmer. Talked about it all the time even though we were in the PC group. Of course, his trust of Microsoft was almost as great as his trust of IBM.
I'm not so antiM$zealous as to say it's risky, I just wonder why such a thing - a peripheral - needs an OS of that sort at all.
I more or less disagree. The thing I like about multiuser with Unix is that running as an ordinary user it's nearly impossible for me to wreck my setup, especially by accident. Winduhs has no such protections, I always had to be very careful.
I agree that for most people only two users is necessary, but fewer than that (or some equivalent) is asking for trouble.
Unix isn't 'expanding' to the Mac.
It's buried deep inside the Mac.
I just got through a CS degree at the University of Oregon a few years ago and the CS computer lab is primarily SparcStations running Solaris with a few Windows boxes there for those that need them. Almost every coding assignment I turned in had the requirement that it compile on Solaris. Also, the U of O offers several X programming classes, but I don't recall ever seeing a similar offering for the MS Windows environment. My impression was that most professors preferred Unix to Windows. Of course, when I was there they were teaching C++ in the 200 level courses. I understand that they have now moved on to java, so OS may not be of as much as a concern anymore.
Er... 32-bit archs can happily do 64-bit math, just with roughly twice the no.of instructions - I suspect all that will happen is that on 32-bit archs, the counter will be defined to be longword:longword (i.e. 64-bit), with the counter increment patched to carry the one in the appropriate place. Heck, it worked when you were doing 16-bit math on the 8-bit C64...
Filesystem arithmetic on most sensible 32-bit platforms is already done in 64-bit - even the Amiga NSD (New Style Device) API is 64-bit. Is it so inconcievable that the same arithmetic used for the (non-trivial) 64-bit file system on 32-bit architecture problem may be applied to a (trivial) counter increment?
Sure, some user space apps will need patching (you might get away with a simple recompilation in many cases), so it might be a good idea to start doing this NOW...
If this were a post by a user it would be moderated as Flamebait
mythus@mailandnews.com
not anonymous, just lazy
"Have you ever seen windows crash in safe-mode? "
Yes.
Hey I heard a few months ago thay existd a Plan 9 II (comp.os.plan9). Also plan9 sucessor was call brazil which became the Inferno (comp.os.inferno) software ran on top of winNT/9* and .... yes my brothers and sisters LINUX. But alas that was about a year and a half ago. The news groups are almost dead. I have noticed that our european cousin's are hotter than us in terms of foolin' round with plan 9. By the way I suggest last year to the comp.os.plan9 group to go open source, they didn't buy it. May if enough of the /. folk could drop hints at comp.os.plan9 ATT/BELL/LUCENT would get it. Lastly, plan 9 + source + what ever licensing can be had for about ~400 bucks from lucent (at least the last time I looked). c:\ ----------------------- from the Matrix: (Subway station) Agent Smith: Do you hear that, Mr. Anderson? That is the sound of inevitability. That is the sound of your death. Goodbye, Mr. Anderson. Neo: My name is Neo.
FreeBSD is not a real operating system by any definition of the term, so it is correct and wholesome that Slashdot should shun stories about it. Furthermore, the slobbering zealotry of the average FreeBSD user serves as an added incentive not to post information about it: if you post FreeBSD information here, you get a larger audience of FreeBSD users. This, as I have earlier demonstrated, would result in a much rowdier Slashdot. Therefore, for all of the reasons here, I am compelled to voice my support for the suppression of BSD-related stories. I hope this policy is continued far into the future.
I wish you a pleasant afternoon!
This is total garbage. Although many 1st and 2nd year courses may be run out of labs w/ NT, all the markers, profs, TA's etc. I've talked to use unix machines when they are doing anything of value.
e. g. Spring is long gone ('96?)
And Fortran 95 has full OO, on a par with C++, anway, and also parallelises better (think HPF).
except Sabrina! :)
Are you thinking of George Santayana, Mr. Anonymous Malaprop?
Some day even Linux might be POSIX compliant.
Of course, that implies that there will be a version frozen long enough for compliance testing to be completed on it.
Right now Linux is POSIX-wannabe.
There isn't even a third-party API, like Interix for Windows NT/2000, to make Linux POSIX compliant.
BTW- Windows NT with Interix is POSIX compliant. To the letter of the law.
What I was saying earlier about Linux being wannabe.... hmm...
I think it's this guy's creative way of saying, "They're all full of shit." No links need apply.
This is a totally pointless post. Did the Y2K bug not teach us anything? I work over at Unisys...we have clients with systems from ages ago. Most big important systems are very rarely replaced.
If it ain't broke, don't fix it. The only thing that would kill Unix is if it stopped working. In other words, Unix will die when it stops being a viable way to solve computing problems. I don't see this happening anytime soon. Even it does, these would have to be new computing problems. It would still be good at the current and old computing problems and these problems are still going to need to be solved. Some very important problems have not changed in years. Thus, "old" Unix is still good at solving these problems.
This is why mainframes hung around. They are still damn good at things that still need to be done in the same way they were always done. Our mainframe business is still very profitable.
When it comes to computing, most people hate change. They take time to learn enough about the system to do their job. Once they reach this point they do not want to learn something else. The only real reason most people upgrade (new software and hardware) is a fear of lack of support. Often the fix for v1 is to go to v2. My aunt has a 286 dos box with no hard drive. She has been using it to write papers ever since she got it, when it was state of the art. It is still perfectly good at word processing. Just recently she considered upgrading to a portable system. When started talking about sound and color she asked, "Sound? Color? Why do I need that? I just need to write papers." This is question that should be asked of every new feature that we are forced to take: "Why do I need that?"
In the business world of big computing, there often isn't an answer for this little question of why.
I agree that any institution would be foolish to limit itself to the products of just one company, but I would also have to point out that most community colleges, such as the one that I am attending, offer classes for any of the popular operating systems including UNIX, LINUX, Novell, and all of the Microsoft stuff too.
goto machine a, Do work. Click Save State.
goto machine b, click retrieve, it comes back
Welcome to ftp!
Hamish
> The bulk of the Usenet traffic these days is purely Binary attachments. The most popular NNTP clients on
:-)
> Windows machines strip off and throw away all 'text' content, keeping only the attachments.
> NewsBin, Pluckitt, etc.
Not the "proper" newsreaders like slrn or XNews
I doubt even (gasp) Outl**k does it.
I would seriously suggest you take a look at http://www.mozart-oz.org( I was introduced to this stuff by a friend who has done network programming for over 20 years). Mozart is 1) considerably faster than Java 2) has better features for distributed programming At this point this is a research project that doesn't have a lot of industry acceptence, but from the reaction I'm getting there is a hell of a lot of potential here. randall_burns@hotmail.com RJB
WinNT's only multi-user problem is that the default security settings are designed to let you run Windows 95 applications, and not to provide a secure environment.
Windows 2000 supposedly fixes this, but under "user" level security nothing runs, not even most MS applications. "Power User" has historically been ridden with holes. Until people actually start designing apps for NT this is going to be a problem. (NT4 SP6 proved that Microsoft isn't even doing QA for user-level security.)
The bottom line is that MS doesn't need to become Unix to pfovide user-level security. They just need to put the smack down on all their Win95 hacks and provide a better implementation of what they've already got.
Those who do not study history are condemned to repeat it, to paraphrase George Sayonara. Computer users (and admins!) familiar with only one OS tend to think of it as the end-all and be-all.
In answer to your question, look up IBM's Transaction Processing Facility. Source code has been available since the early 70's at least. Reliable? There are some instances that have been running continuously across several generations of hardware - an uptime of decades. Will it replace 'Unix'? No, but it, and other mainframe OS's, will remain important.
Whoa, I just had a bad flashback. I'm always amazed that MVS is still around. I got to work on an old IBM360/67 running OS/VS1. The interactive part was initially CP67 (I believe), then CMS. Ahhh, the days of editing code over a 300 baud phone line.
CP67 had a great line editor. If you wanted to go down 10 lines, you would type, d 10. No wait! D is an abreviation for delete. And JCL? Knowledge of JCL meant job security.;-)
//
UNIX will die. Some kids in a NYC loft apartment (garages are now overrated), will puke out the new language in the next couple of years that will change everything.
Think beyond the command line. Think three dimensional. Information creation and delivery will and must change. The information load we carry and distribute is getting a little to heavy. Youth is king, and youth believes that typing letters one by one is insanity.
Except for speed, not much has changed in the way our thoughts and ideas have been delivered since the Renaissance. Bring down your flaying arms to your soon to be archaic keyboard, exit the coffee shop and get busy! Stop thinking, we have been thinking for too long . . . You are youth, act and create it.
Let's see:
Windows -- moving their userbase to a multiuser (timeshare) system.
Macintosh -- moving their userbase to a multiuser (timeshare) system.
Linux, xBSD -- multiuser (timeshare) system, gaining in popularity.
Be -- Single user is good enough for us!
Here's the clue: They aren't encouraging (or allowing, in many cases) "Hackers" to add all kinds of kludges and DLL hell. Nope.
Funny thing is... in the town I live in we have 'info screens' all over the place. Small embeded machines with a touch screen that can give maps and various other info about the place. Usually about half of those are displaying a bsod at any given time.
I have a machine on which windows is reliable.. It has about nothing installed except for IE 5 and some media stuff (DVD, MediaPlayer etc). This machine runs win98 and has never(!) crashed. It has been running win98 since almost as long as win98 exists.. So yes... it is possible... but there are a lot of systems that are much better suited since they offer this reliability without putting completely insane demands on what code you run on them, how many bugs that code has (it will have bugs statistically, no matter how hard you work on removing them) and without being absurdly selective in what hardware I can use it on reliably.
I believe Unix will just refuse to die... there are too many people who have the tools to keep it going and who want to have it go on.
Run in an embedded setting, where the hardware is completely captured, specified, and validated, it's not that risky a proposition. Not that it matters in this discussion. Too many zealouts in these parts for any semblance of reason.
Depends if you count the elite high schoolers that duel-boot with Windows 95.
Linux has certainly outsold commercial Unix by orders of magnitude. However, every commercial Unix sale is in use somewhere, where a huge chunk of Linux sales never go into 'production'.
Banyan Vines was years ahead of it's time. Too bad the market chose (uck) NetWare.
OpenVMS is not dead.....
just sleeping.
The desktop PC market is no longer it's focus - thats why from now on it'll be free. It's hard to compete against Windows, MacOS, and Linux/KDE/Gnome on price unless you define yourself as a niche product. Be did just that - it's now a "network media device OS" (for serious revenue generating purposes) and as a desktop OS it's free. I wonder what % of desktops in Be's head office run BeOS ;-)
...
...
But heh - with, perl, X, python, bash, POSIX, GCC, and the latest improvements in the networking API with version 5.0 guess what Be becomes more and more like with every release? Can you say Unix?
It's got more C++ and more threads and the GUI can't detach like X but other than that
(actually high end Irix and Solaris is pretty darn threaded too just not the GUI)
Unix is the Borg. It will adapt.
Welcome to the collective. You have been assimilated.
Seven of Nine, tertiary adjunct to unimatrix 01
Unix was well designed, but is the design better than other early 70s minicomputer systems? To this day folks will tell you that VMS was, is, and always will be better.
We shouldn't fool ourselves. Unix is popular because AT+T couldn't sell it, so they licenced it to thousands of companies, and it caught on with users. Before the advent of the PC, Unix was the cheapest way to get an operating system on a computer. Now with Linux/xBSD, it's even cheaper. What we see here is the triumph of rational economics, nothing more.
Unix IS dying. Contrary to popular slashdot hysteria, the unix marketshare is shrinking and is being replaced with Windows NT.
Sorry kids.
Your right -- NT was designed to be a "NetWare Killer" and a "OS/2 Killer". The "Unix Killer" moniker was only adopted after the first foes were vanquished so quickly.
I think that Unix will be replaced, if only on the corporated desktop. Thin clients will replace them due to their stability, security, and cost. The system admins and architects to whom I've spoken say that it's the "wave of the future." They say that they won't need to manage thousands of desktops, just a few servers and the alread-existing network. That said, I don't see anything major replacing Unix in the server market. Unix has its place, much like mainframes, Macs, and Windows boxes all have their places (mission-critical apps, multimedia apps, and games, respectively). Someday, something may replace its ubiquity, but I think it will always have a place.
just like the death of rock, punk and pot smoking (drug free america) -- yeah right!
I would have to say microkernel based systems like BeOS and OS X (Mach). They have excellent support for multiprocessing, and adding extenstions and kernel services are pretty easy. BeOS is the only OS I know of that you can download a driver, download an 1 file, put it in some directory, add an interface w/ that driver, configure TCP/IP for the device, and restart the networking subsystem-- all in under 2 minutes, without restarting the computer once. And booting up takes less than 10 seconds on my box, so its some amazing stuff...
Hey yo, this is 1HunditWattWarlock holdin' it down up in DGL studios...west side. I have to say my homey got ahold a some 'bad boulder' if ya know what I mean- A dude's got to be Fraggle Rockhead to even be ponderin' on the demise of Unix! Damnnnn! We use some ruffass behind the scenes, Xeon shitz wit Gigz 'o RAM, and Mylex RAID actions on some FreeBSD up in dis hood...shit! Those servers are keepin' it real for our sample procurement needs. We got all a motown and more stashed up in them hard discs! Brother, I say to thee- getchyo head out yo ass! My, my, when these 'my first internet experience was through AOL' people grow up some, they sure contemplate some deep shit, don't they? Demise of Unix- yeah foo, and when that day comes, you'll know, because that will be the day you wake up every morning to a bowl of hot grits down your pants. Thank you. 100
The distributed OS is IMHO the way to go. Look at the efforts of the SETI@Home Project; distributed processing cycles to a multitude of computers. I believe this is only the beginning. Lucent's experimental Plan 9 OS got the ball rolling, but, according to Linus' comments on the OS, they ended up stumbling backwards through poor implementation. Also look at how most OS work is shifting away from SMP and more toward various clustering implementations. Although distributed OSes may or may not be "the" next big thing, it is definately a step in the right direction.
Beeing a Windows user I have to confess that Unix has loads of things right, mainly its flexibility and robustness. I had some hopes for Windows 2000. This OS is alright but it can't stand one second against a well configured Unix machine. Windows 2000 is fast, look it has the world highest score for the TPC-C benchmark. http://www.tpc.org/new_result/ttperf.idc But when things go wrong with Windows, then things get worse and worse because everything is related to everything else. There are still much better developing tools for Windows but the situation is clearly changing now. But I think the question should be anyway widened. Fearing about the death of Unix is like fearing about the death of Windows. I think and I do really hope that the computers will go to another step and that this step will be totally different. That maybe can sounds stupid but I said many times to my friends that we all developers will end programing with cubes, spheres and cones. Why? First because object oriented is the only way to build rapidly value-added software and because having to each time recreate the wheel is a really dodgy path to design high quality software. Now I don't know when the computing industry will reach this level, maybe if the quantum cpus see the light then we will have time to code *really* interesting stuff.
It is the original "modern" OS: all of the others borrow concepts from it. All it needs to go on is better useability, better hardware support, better applications, games and... better commercials :) All of these issues are being worked on today... I'm not worried.
I'd say more, but others have already done about as good a job as I could.
actually, i think v2os is along the lines of "coding a completely new OS from scratch".. they're breaking alot of conventions, you should check it out (it was on /. a while back)
Well the Usenet problem in particular, and many network bandwidth problems comes in general comes from that that it takes hard work by humans to generate Signal, something that scales linearly (or slower) with the number of folks online. Noise on the other requires no effort and thus scales exponentially as the network grows. The prospects for the S/N ratios are not happy The one bright spot is to imagine how much time email address harvesters are already spending trying harvest each others spam.
Unix is not about the desktop, or user interface. It's about being able to run your business - be it erp, oltp, whatever - reliably.
Businesses around the world have been relying on Unix for ages, and will, without doubt, keep doing so.
"Mission Critical" spells "Unix"
Was this guy trolling? There are TONS more Solaris, HP/UX, AIX installations out there than Linux OR any of the free BSD's combined!
Man, you guys are TWISTED.
The valid reason is that every system needs a development environment, so it's easier to accept a desktop OS and get good development tools than to use a customized embedded OS with second-rate tools. I don't agree with this reasoning but some engineers see it that way. Think of it like this: the resources taken by the OS are a smaller and smaller part of the cost; the NRE cost amortized per machine sold becomes comparable for something like a specialized oscilloscope.
The bad reason: the boss likes to be associated with a market dominating company. The fact that they're associate the way a chump is associated with a con artist doesn't get through to them ("I use Microsoft on all my desktops -- it would be great in our product too!"). More insidiously, the boss can own stock in Microsoft (lots of people do -- all that market cap belongs to somebody).
Some people said USENET would die. Some people said Linux will never compete seriously with Microsoft. Now some people say Unix will die.... Some people are really fucking stupid.
Actually, Banyan is around. I work for them. We don't make Vines anymore, we now do other integration and web application design. Did you guys know Vines was based off of *nix? www.banyan.com
Unix survived the paradigm shift of networking just fine. In fact it has became the dominant OS for networking servers.
Unix assimilated the GUI transition. I won't claim it's got the best GUI but at least it has one.
Unix accommodated the Free Software / Open Source movement. Imagine where we would be today if RMS had decided to work on a VMS-clone or an ITS-clone. Someplace completely different for sure! But thanks to an accident of history, most Free Software is based on Unix as its native, primary platform.
I don't remember any real time scheduling facilities back on those PDP-11's but QNX and Lynx have good market share in those markets.
Maybe Microsoft can kill Unix; maybe it can't (I personally think it can't). But I laugh at the idea that Unix will die because of paradigm changes. Unix has already assimilated more paradigm changes than any other OS in history.
Dude, your comment just plain fucken RULES!!! I can't express my deepest appreciation for such lucid, intelligent verbage!! Glad we have your grey-matter on this planet!
Millenium will be in no way any beter gaming platform than W98SE is right now, it just the beter DirectX. But do not fear, Microsoft will hype it anyway...
That's the same with Windows, you have CE for that. People please keep PC OS's out of our embedded devices, keep it simple please!
Real programmers do it anywhere- clean or dirty
It is only healthy to be mindful of the fact that all things have a life cycle, and UNIX's will eventually end. As a die-hard UNIX programmer, I recognize that such an ending would make my skill worthless, although presumably its successor will be such a joy as to calm any fears. Besides, surely the better ideas in UNIX will live on, long after its death. It is too early now to state "yes, xxx will definately replace UNIX" or "xxx will definately replace the GUI."
Have a look at www.eros-os.org
It's not so much the capabilities of Macs vs. PCs which make Macs so very critical in the graphic design market, but rather the fact that that's what people have used since 1985 or so. Graphic designers are going to want boxes which they can be creative on, and if they know Macs best, they're going to want another Mac so they can keep working on their pictures/ads/layouts/whatever without learning another operating system. Apple realized this and catered hardcore to the 2d design market with seamless Postscript integration, color matching technologies, et cetera. Nowadays, there are so many conveniences under MacOS for graphic artists that even the newcomers (who learned their computer apps post-1995) see real advantages to running Macs and not Windows. So they throw a fit if their bosses try to put a deal-of-the-week Compaq on their desk.
This is where Linux will falter, in the same way that Windows faltered -- i.e. against the surrounding technological environment.
Linux is great, but is not inherently built as a distributed multitasking secure load sharing micro kernel based operating system. Taunenbaum(sp?) and Linus have argued about this, about the lack of theoretical underpinings to Linux.
Unix will have a place, but it is not a "new age" operating system any more.
You're one of those people who really _get it_.
You're confusing easy with primitive.
I wonder if there's something in human nature that causes people to go around,
pointing at things that are quite usable, functional, and/or even popular
and screeching "Obsolete! Obsolete! Throw it away!"
Is it just ordinary cluelessness? Or is it insecurity about being seen as not "with-it" technology-wise? Afraid of being seen as clinging to "old techology"?
I've had quite enough of people telling me that one thing or another is
the "wave of the future". Especially when most of the time, they're really saying "you're not using what *I* think you should be using".
Remember 3DO. Remember the Merced. Remeber the i860. Remember bubble memory.
"pervasive multithreading" is such a marketing joke. Just because the OS design forces you to use threads where they're not needed doesn't mean they'll do anything other than sleep most of the time. The processor intensive part is just as hard to break up into threads on BeOS as it is on any other OS that supports threads. Linux 2.2 utilizes both CPUs on this SMP box about as well as BeOS. (not very well, unless I'm running two processes, or running a process that explicitly split the intensive part up).
you do know that your 100% object oriented kernel is written in PURE C, right?
Hell, you can't even use the scheduler latency arguement anymore, since 2.2+patch or 2.3 both match or beat beos there.
Dear Sir. Your .sig is a complete trolling flamebait. WTF is wrong with you! Or was that jpeg a self-portrait?
I am just wondering why Debian GNU/Hurd is not mentioned once (in the Comment I am replying to at least the GNU Hurd is mentioned) in over 400 comments on the end of Unix and a possible successor. If the Hurd is really that unimportant (less than .25% of the comments) and nobody really knows about and/or writes about it, well, then I guess we will stick with Unix/Linux for some time to come if we like it or not. Other than that I must admit, that I don't see any other alternative for large servers, internet, networks... to Unix. If there is, please post :-) I just happen to be in for choices and even though there are tons of flavours of Unix around it is still Unix. Peace!
Unix has it's weaknesses in today's "distributed" environment.
Guess what? It had them 10 years ago too. NFS was developed. NIS was developed. automounter. ssh. The list goes on. UNIX will continue to adapt.
"Don't assume anything" is perhaps what Thompson and Ritchie had in mind when designing the UNIX kernel, and the result was the flexible system that we still love today.
Add in the reletively recent development of modules and you have a system that can include 30 years worth of development without becoming bloated. Wonderful!
I am just wondering why there is just one comment about the GNU Hurd in almost 400 comments on the end of Unix and a possible successor. If that GNU project is really that unimportant (0.25% of the comments) I guess we will have to stay with Unix for a while wether we like it or not. This is not meant to be critical of Unix/Linux, I am just in for choices and even though there are tons of flavours of Unix it is still Unix. I didn't know about eros before I read the comment I am replying to and I think it deserves the score for sure. When I read about the Hurd for the first time I thought it was the coolest thing in the world and hoped there would be more people like me. Maybe next year ;-)
Peace!
Quack!
Empirical evidence, I think, speaks for itself. 1 year in the running and no sign of 'death' yet. I don't believe any other poster on Slashdot can claim that distinction. Signal 11 speaks on any subject, even those he knows nothing about, appears on multiple user forums, and is gaining ground even today in the karma whore market. I'm betting it will continue to survive for another 20 years. Infact, as soon as I raise enough capital to get a gun, I'm going to do just that: aim at Signal 11.
This would be like asking "Is *nix the operating system of the future for everyone?" on www.welovewindows.com ... I can't honestly say that the /. community isn't the slightest bit biased against windows. However, as a windows advocate I can't disagree with the idea that unix will be around for a long time, but it's going to need an overhaul if it's going to survive because many of the new generation doesn't know how to use unix. I think we (as people knowlegeable about unix) are a dying breed, and unless something is done to stop this, unix might be in some trouble.
- To add new hardware requires too much involvement of the user. It may require recompiling the kernel, creating an initrd image, etc. An operating system should ease the administrative problems of modifying the architecture underneath the operating system.
- Device drivers should not be able to crash the kernel. Usually faulty device drivers from third-parties are the reason a kernel will panic, or why MS Windows crashes. Have you ever seen windows crash in safe-mode?
- Computers need to be "smart." By this I mean an operating system needs to ease the administrative tasks without compromising speed and reliability.
The best option for a system like this is a microkernel design that can work through failures. So that if one part fails, it does not affect the kernel.Actually, I think only fucking morons like you would say "So far, so good!" if you jumped from the 30th story building.
"Empirical ev..."
IT DOESN'T MATTER WHAT YOUR OPINION IS!!!
... or at least thats the number of times Unix has been killed by the media since the 1970's.
I thought the "will Unix die" question to be along the same lines as: Will algebra die? For the population ecologists out there, Unix has the advantage of age: it's been around longer, so its collective intelligence is greater than Microslop. Put that in pipe, and smoke it. Please do not entertain any more Microsoft related topics if you really want to stay on the bleeding edge of computing. I dare you to moderate this down!
I thought the "will Unix die" question to be along the same lines as: Will algebra die? For the population ecologists out there, Unix has the advantage of age: it's been around longer, so its collective intelligence is greater than Microslop. Put that in your pipe and smoke it. Inhalation is optional. Please do not entertain any more Microsoft related topics if you really want to stay on the bleeding edge of computing. I dare you to moderate this down!
well, i totally agree with you, even in the BSD section there's no word about 4.0 release... but you know you can create an account and filter what you want! you can filter all Linux story for example, it's easy!
What makes you think you can raise enough capital with a Taco Bell salary?
The wheel has lived well beyond the era in which it was born...but do you see any major upgrades on the horizon? UNIX is an evolution, it won't die anytime soon because it is fundamental to so many business systems, yes it may have it's unfriendly side, but it is a far more robust and architecture independent design. And why reinvent the wheel?
Linux bootstraped itself into existance by promising UNIX/POSIX compatibility. But despite its spectacular growth, being a unix-clone will only get it so far. the next stage is general user acceptance, and unfortunately the unix-baggage holding it back. Expect the unix command-line utilities (awk,sed,troff,grep,find,etc) to either be tossed out outright (on WebPADs and embedded devices), or marginalized (on servers and desktops). In a few years, linux won't even be recognizable as a UNIX system to most people. people whose only 'UNIX' experience is with linux will be lost trying to administer/use legacy UNIXes.
>Nothing short of an asteriod collision will prevent unix for surviving.
Even then, much of the military network will still be active after our species demise...
Until BeOS becomes a multi-user system, it can never replace Unix (or even WinNT), except as a single user workstation.
Not that BeOS is bad in any way, just that UNIX-like equals multi-user, at the very least.
Fortran is not dead! Many people in science and engineering still use it. It might not be as hip as java or even C but a lot of people still use it to get work done. It is easy to program for, there are tons of math libraries, millions (maybe even billions) of bug free code out there that is still in use, and Fortran 90 handles matrices as easily as matlab does. It may only have a niche but it will never die!
FreeBSD not an operating system? Go back under your bridge...
Just like Fortran, there are too many useful programs and libraries written for it. For many people it will always be the operating system of choice, so it will continue to be widely used until they all die of old age...
That would be true even if Linux and *BSD hadn't come along, but free Unix OS's and the open-source movement associated with them have given Unix a new lease on life. There's a whole new generation of Unix users, programs, and libraries, right at the moment computers are making the jump to pervasive consumer usage. I bet it'll take 10-15 years before an OS comes along with enough advantages to displace Unix from a significant portion of the market.
If someone comes up with an OS that's inherently more secure than Unix, with low overhead, hooks to enable emulation of Unix and Dos/Windows applications, and blatantly superior user interface capabilities, it might replace Unix. In 10 years.
Absolutely. And the future is here:
desktop.com
MyWebOS
I suspect there are others. Anyone?
Needless to say, this causes some code to grok, but at least SGI allows one the option to compile in 32 bit mode.
You do not know annoying until /. crashes your doze workstation :-(
/. loads with 300 something select boxes it just rolls over and dies and it took my system with it.. nuff said.
rofl im meta moderating and when
yeah i use BeOS at home and im loving it :) (though netpositive could use some help!) to install the (RO) linux fs drivers for beos, i unzipped the file, dropped the unzipped files whre the readme said to and then mounted my linux drive :) no fuss, no muss.
i still use linux at werk (i have windows and linux at werk, beos and linux at home) to get things done though.
btw, does anyone know of any good AIM clients for linux? im using the java AIM which totally sux (requirements: must be able to get thru a SOCKS4 proxy).
"There is no spoon"-Neo, The Matrix
"SPOOOOOOOOON!"-The Tick, The Tick
It was kscd or something like that (the kde cd player which is installed by default). Even a loop with kill -9 wouldn't help, and when I was rebooting it was ignoring the SIGTERM, so my filesystem couldn't be cleanly unmounted, so it run fsck upon restarting...
Even worst, the process was restarted automatically when I started KDE, I got rid of this annoyance by renaming the binary.
I think that it has something to do with a buggy drivers (which? soundcard, SCSI driver, CDROM ? I don't know), and I don't blame Linux for this, but it was quite annoying, right now I'm using the gnome CD audio player and it works all right...
My guess is that Windows will eventually collapse under its own weight. I remember when Macs were SOOO much easier to use than DOS machines and we still weren't buying Macs. I'm an old COBOL banger and I installed Linux at home and I use it. It is not only simple, it uses a LOT less memory.
Unix is fast and stable. Is NT? I don't think so. NT is a bloated pig and Unix is now easier to install and maintain. Do the math.
If you aren't part of the solution, there is good money to be made prolonging the problem
If you mean distributed computing or spreading a task among multiple computers that are linked together, you probably mean Beowulf.
If you want my respect, give it first...
If you don't want my respect, expect mine before you give it.
Firstpost!
:wq
screw eros..capabilities are coming to linux, journaling is coming to linux.
Here's a cool scenerio:
goto machine a, Do work. Click Save State.
goto machine b, click retrieve, it comes back
does eros do this? does eros strieve to do this? no, it has to be done on an application programming level. If I get my OS in rom, my files on a server, and my application wrapped in corba beans saving my data (and session) to the server, why do I care if my OS supports any of eros's features. if eros actually wanted to integrate at the application programming level (ie: with gnome), then all this would be useful. You do not require a different OS, period. The _only_ need for a new OS (ie: kernel) is when the hardware is radically different. We don't have that situation occuring right now, and not for quite some time.
So, pick your OS, have fun with it, speed, reliability and hardware support (and open source;) are the only requirements of it. everything else can be added.
NT Baby!
About the only technology that stands any chance at all is Java/Java Platform from Sun microsystems. The Java Platform is an operating system in every real sense of the word. The current problems with the Java Platform that are roadblocks to this happening are: a) Java's performance (its a bit, NOT a lot slower, then again C is a bit slower than assembler- but who cares?) b) its semi-proprietary The main drivers that are making it happen are: a) Java distributes real well b) hacking Java is a lot easier, quicker than hacking C/C++ that is mostly used for Unix c) Java is more reliable (although a lot of JVMs might make you doubt that!) d) Java interworks with lots of things- its a survivor Of course in many current situations it is run on top of another operating system, but there is no necessity for that. Java doesn't need Unix to be there at all. The converse may not be true for long... Still, prediction is always difficult. Especially about the future. (Thanks Woody)
-WolfWithoutAClause
"Gravity is only a theory, not a fact!"Unix is sort of like Christianity. It's an idea whose time has come and gone. Sure, it was useful for a while, but we're to the point now where it has served its purpose and now should be gently handed its hat and shown the door, because it is no longer needed. My guess is that within 20 years, Unix (and Christianity) will have virtually no adherents except for the rabid fundamentalist types who rave on street corners and who you constantly worry about blowing up buildings or buses. The rest of the world will have switched operating systems.
.. I envision a point in the future where BeOS stops just being an OS for the fatties and queers and becomes powerful and useful enough for the rest of us. Heck, who knows? In 20 years time we might be having the BeOS/Hurd flamewars instead of the Windows/Linux flamewars!
Now, switched to what? That's a good question. Windows seems to be attractive at the current time, but let's face it: The type of slobbering invalids attracted to Windows are generally not the people that you would expect to pass their genes onto a new generation. By and large, these people will have electrocuted themselves in the shower or killed themselves in freak blender accidents before they have the opportunity to reproduce.
So Windows and Unix are both out. So where are we at, then? Well, unless I miss my guess, I forsee a future where both BeOS and Hurd have huge followings. This, of course, is notwithstanding the fact that these operating systems are currently being used only by sociopaths, homosexuals, and extremely fat people. That's okay, though
At any rate, we can agree on one thing: We live in interesting times. Let's see how this all pans out.
> multi-user/group kludgework.
Brother, I would like to introduce you to the concept of 'the enterprise'. Or even 'the internet'. Your grasp of what is important in large-scale computing seems to be relevant to a thing called 'ms-dos'. Here in the real world, we like the idea of 'sharing' and 'interoperability'. 'Privilege systems' and 'user rights' are nice things, too, but I guess you're perfect enough to run everything as 'root', or as we like to call it, 'firehose-mode'.
Okay, enough cuteness. Moral: you're an idiot.
But that's okay; most of us are idiots. Your particular fuck-up is to assume that everyone use computers just like you. Which is a stupid thing to believe, about computers or any human endeavor.
I'll let you re-think that post, and point out more of your logical flaws later.
ciao!!
The Amiga OS is hardly dead, it's still being actively developed, in two main streams:
1. a next-gen distributed architecture based on the Tao VM (think of it as a language-agnostic generalised VM a bit like Java, but that can run on real hardware too).
2. the "classic" Amiga OS was extended to PowerPC with WarpOS (no relation to OS/2) microkernel. This allowed the user community to use more modern hardware, such as G3 accelerators, and 3D gfx cards.
The Amiga OS design, in the form of AROS - the "Amiga Research OS", which recently received blessing from Amiga itself, also lives on.
For more amiga info, go to www.amiga.org
Lots of places run life support equipment on Windows. I won't my company while I'm under a defacto NDA.
Tektronix sells Logic Analyzers with Windows 95 embedded into them. That's a pretty "out there" environment where the OS takes a beating and survives.
Here's the clue: They aren't encouraging (or allowing, in many cases) "Hackers" to add all kinds of kludges and DLL hell. Nope.
Run in an embedded setting, where the hardware is completely captured, specified, and validated, it's not that risky a proposition.
Not that it matters in this discussion. Too many zealouts in these parts for any semblance of reason.
No, the 8.3 backwards compat is NOT a good idea, because wildcards will accidently match the 8.3 versions on NT and delete far more files than you intended. VERY BAD.
Well, on the moderated groups, the S/N ratio is usually very good. e.g., compare soc.culture.japan with soc.culture.japan.moderated.
I download and archive select Usenet groups, and... umm... a lot of them are a total loss.
Well this was true from day one. Nothing has changed. More crap sure, but more content too.
The Linux groups have some of the highest traffic, but also the largest volume of clueless noise.
We're all clueless about something. And that's what the Linux groups are for. To ask questions. Not to post braggings about how you're tunneling IP packets over email to get around the worst firewalls. And they're good at this too. When I have a question, I'll search the groups, failing that, I'll search on deja.com to search the stuff that expired off my server, failing *that*, I'll post. And usually get the answer I need within a few minutes to days. And that thread gets snarfed by deja and other news archivers to help someone else later on down the line.
The real action has shifted to private forums like /. and listservers.
Yeah, but the problem of everyone migrating from a common forum (USENET) to Slashdot and freshmeat and kernel.org, and redhat, the mailing lists, etc., is the fragmentation. You've lost the world audience and the single searchable source for answers. I can't know that my question was answered on some obscure mailing list I never knew I could subscribe to. Or worse, you can only discuss whatever the oficially sanction topic(s) of the day are. (see /.)
The bulk of the Usenet traffic these days is purely Binary attachments. The most popular NNTP clients on Windows machines strip off and throw away all 'text' content, keeping only the attachments. NewsBin, Pluckitt, etc.
And these programs are good at doing graphics. When I want to download pr0n or sailor moon images, I'll use GUI news readers, otherwise, it's good ol' trn for the bulk of my news reading. I usually hang out on rec.arts.anime.misc. Most non-binary groups with a lot of traffic, excepting the flame or controversial topic groups, have maintained a decent S/N ratio for the last 10 years and more.
IMO, USENET is better than ever because there are more people using it. Most of the noise lands in the abandoned groups, or the flame groups anyway, so who cares?
>The Internet started on UNIX... Excuse me, but the Internet (ARPAnet back then) started about the same time as UNIX did, and it was a long time before there was a UNIX connected to the Internet. The earliest systems on the Internet were TENEX, OS/MVT, an SDS Sigma 7, a PDP-1, Multics, etc. Bell Labs had only just dropped out of the Multics project, and Ken Thompson was just starting development of UNIX. UNIX wasn't on the ARPAnet until the Univ. of Illinois did a UNIX implementation, several years later.
I would love to see some of the best coders and operating systems people put together a new OS from scratch using the latest techniques. Ideally this would create an ultra stable and very modular system.
Its already been done. Check out either BeOS or QNX. Both are top notch, extremely stable (especially QNX), blindingly fast, and don't carry all of the extra baggage/bloat that legacy OS's have.
Someone once said something like: "I'm not sure what the dominant programming language will look like in the year 2000, but I'm sure they'll call it Fortran." Bad bet, but the right idea. How much does today's Unix resemble the Unix of even 20 years ago?
In fact, maybe it's time to start thinking about this now, while the Y2K hype is still with us?
I agree with the prediction that things will end up moving more towards centralized computer resources
It's not resources that will be centralized, it's data, and it's not centralization you want, it's ubiquitous access such that it all seems as-one personal database. I don't want 7 copies of my address book and manual synchronization, I want one master copy, and cached copies as needed. I want access to all my data no matter where I am and what machine I'm using.
Ooh, a sarcasm detector. Oh, that's a real useful invention.
But, do you want to have your data at home? Or at some crappy internet providers' computer?
Both, for redundancy, accessibility, and speed reasons.
Ooh, a sarcasm detector. Oh, that's a real useful invention.
Ironic that this Ask Slashdot comes up, right after I submit a post in regards to web based office suites, and web oses. MyWebOS.com, Desktop.com have been mentioned. Workspot.com gives you a xwindows window in your browsers. Essentially it creats you an account on there linux box and through there modified VNC java viewer you can use linux from any machine with net access. MyFreeDesktop.com I believe is another one. Halfbrain.com i think is another web based app company that is working on a kick ass spread sheet program. The evhead weblog has some links today in regards to web based applications like that, www.evhead.com. Posted direct from SXSW interactive. My personal favorite is ThinkFree.com's Thinkfree Office. It gives you a remote desktop using java, and it has an excellent Wordpreccesor, Spreadsheet, Presentation program. They rival Microsoft Word and Corel Wordperfect. In fact, they are almost exact clones of the products in the Microsoft Office Suite. They even can deal with the formats of office. Its must excellent, you get 20 free megs of workspace , and you can type and edit your documents anywhere with a java enabled browser. It probably sounds like I work for them, I dont, I only wish I did. :)
Jeff Knox
>I have had limited time to read these posts but I think many people
>sell MACH short. Microkernel seems to be the way things are going.
>Modularity in Linux is very popular and necessary to support all the
>devices that it does and future ones (The kernel can only be
>reasonably big and still be worth while). Microkernels get away from
>this monolithic design. I admit that they have yet to prove themselves
>agains UNIX but I suggest we turn our eyes to the GNU HURD project as
>I think they will have something to reckon with in a few years.
Sheesh. Weren't you mircokernel supporters saying this nearly *10* years ago?
The creators of Unix decided to try developing the successor to Unix, and they called it Plan 9.
Dumb yourself down, run Windows.
Immagine for a moment: someone invents a very powerful palm-top device that lets you word-process, surf the web, POP your mail, and all the other neat things a productivity PC does now. It's not based on current chipsets, etc, etc. This person may say "this concept will someday replace the computer."
We all know what I just described _is_ a computer. What the inventor really meant was "this will replace the computer as we know it today." Now apply this to Unix. Very few people are going to go out and create a new OS without building in someway upon UNIX -- even if it's only just the concept of Unix. Unix has been around for 30 years, and no matter where technology progresses, I have no doubt that we will be able to trace it's roots back to Unix.
Observe: MS-DOS was, IIRC, originally intended to be a 'simple' Unix-like OS for the PC. We all know that it never made it, but the influences are quitte clear. Windows95 evolved from DOS, and implemented more Unix ideas, like long file names and shared libraries (I know, I know, Win3.x had shared libraries too... but it wasn't an OS IMHO). NT is also an offshoot of both Win3.x and Unix ideas - offering network support and user/group-based file permissions.
Yes, each of these took a little different take on Unix, and tried to re-invent the wheel: but the influence of Unix cannot be safely ignored.
Just my $0,03 :P
:: remove the whitespace to e-mail me ::
--
We may not imagine how our lives could be more frustrating and complex—but Congress can. – Cullen Hightower
Anyway, short answer is (with appologies to Mr. Twain):
"Reports of the death of Unix have been exaggerated."
--
If your map and the terrain differ,
trust the terrain.
It happened long long ago.
There once was a patent on the SUID bit.
It, and the other good ideas in Unix are now
available for use by anyone.
Unix is dead, long live Unix.
As a die hard unix advocate of far too many years standing I was surprised to find myself agreeing with you almost entirely. Unix is a beautifully simple and elegant system but you're right. Why is there nothing better. Bill Gates pledged to destroy Unix many years ago. It may well be that he is actually the one who has helped it stay healthy for so long. While all the manufacturers have been pandering to the Windows operating systems which are plainly of no use for real business work, there has been very little development work creating better operating systems for the real stuff! It seems that all the flamers shouting Microsoft has stifled innovation in the computer industry might be right after all.
Of course not. People have been predicting the death of Unix almost as long as they've been predicting the death of mainfraimes. And a year of so ago, IBM released a mainframe (don't remember the model) which became the best selling mainframe ever in initial 6 month sales.
I predict that 20 years from now, we'll still be hearing how Unix is dying and almost extinct, prolly by the same people who will still be saying the same thing about mainframes.
Topher
While I mentally and spiritually agree with probably all the technical/flexibility/etc merits pointed out here, one of the best guarantees of (L)Unix's future is. . .Yup, Micro(in)S(ert word od choice here).
This really isn't well thought out, just sorta popped in my head and I thought I'd post it. But the bottom line is, if something begins evolving that can crush the current popular paradigms in OS's what are the odds that MS will let it survive? Worst case senerio is that MS buys, or otherwise aquires it in it's younger stages and then procedes to screw it up as they do most things. Of course, that's if they don't just kill it flat out due to pride in their best operating system ever, NT (and kin)*puke*.
The end result of this, I believe, is that only General Purpose OS's that are currently entrenched
and have the ability to evolve to meet up and coming issues will be players in the near-moderate future.
BeOS was a niche OS, wasn't a threat. OS-X, well, it's unix (bsd) anyway, but still not a threat. Plan 9, cool but obscure, and still not revolutionary. But if a company (new paradigm Open Source OS's I believe, are exempt from this discussion as MS can't really touch 'em) ever starts to release a ViewPorts 1.0 OS that is a commercial OS with tons of promise, they won't make it.
Anyway, hope this makes a bit of sense, should have thought it out more ; )
G
Yep, when our SUN runs out of fuel...
"In short: just say NO TO DRUGS, and maybe you won't end up like the Hurd people." --Linus Torvalds
but after almost 10 years of using all kind of unix flavors, mostly HPUX, i'm using BeOS, so maybe un*x will never die but will be replaced by some new OS which look like un*x, for end user. Of course for a company running a big webserver or whatever, a true unix like BSD is the best! What make un*x beautiful is its difficulty to configure, i like it! i like to hack every config files on unix, i spent too much hours on HPUx i guess...
--
BeDevId 15453
Download BeOS R5 Lite free!
"Science will win because it works." - Stephen Hawking
I don't think Unix is going to "die". Face it,
in ten years, maybe we won't have Unix anymore,
maybe we'll have XYZ-OS, but I guess, that even that one will still contain many features originating in the Unix age.
Before you lament about the coming of distributed systems, think about - what IS a distributed system?
In a sense, a Unix box alone can already do distributed computing: jobs distributed over multiple processes:
cat
cut -c-4 | \
tr '[A-Z]' '[a-z]' | \
sort | \
uniq -c | \
sort -n
This little "excercise" will extract the first 4 letters of every word in
The principle used, combining multiple programs into performing a function for the user is something that was done very early in Unix. "Modern" GUI-Boxes (Win, Mac,
If you think, that distributed computing is something different, and will only work with multiple machines, then you could still have the same principle working this way over multiple machines (if you want to try this, change the host names for ssh accordingly; and if necessary replace ssh by rsh/remsh):
cat
ssh fenun.icemark.ch cut -c-4 | \
tr '[A-Z]' '[a-z]' | \
ssh varuleon.icemark.ch sort | \
ssh varenorn.icemark.ch uniq -c | \
sort -n
OK, it's not very useful to distribute this example over multiple machines, but it shows that the principle of distributed computing is something, that is not far from what could be done on Unix several years ago - ever since Unix features remsh.
And all that from the comfort of your own shell scripts.
OK, writing a distributed system via Corba or any other mechanism is "a little" different, but these differences don't change too much in terms of what is or isn't distributed computing.
The question that I would REALLY ask, is - will the future bring "distributed SYSTEMS" or "distributed OPERATING SYTEMS"? Distributed systems are implemented on various systems nowadays. Distributed OSs are rare, and I wouldn't really see, whether these are really something required for future applications. A distributed OS will also cause problems -- distributed applications are already hard to code and even harder to debug. A distributed OS might help a little, but it will also cost us in the sense, that debugging what exactly is going on in the system might become a lot harder, since the distribution process would be a "black box" from the perspective of the programmer.
I once wrote a daemon to handle a spec which included "when a file appears in this directory, process it and then delete it". I had a hard time explaining that the daemon could:
Once more unto the breach, dear friends, once more, Or close the wall up with our American dead!
When asked what the programming language of the future will be, a wise programmer once said... I don't know what it will look like but I know it will be called FORTRAN.
// Zarf //
The same holds for Unix. I don't know what the OS of the future will look like, but it will be called Unix.
Same for the internet.
well.. more realistically... **ix, as in Linux, Irix, myix, Jesix, or whatever...
-
[signature]
Also, having a Root account in an OS almost ALWAYS leads to security holes.
Are you being ironic? Are you of the sensible opinion that root should only be allowed to log in on the console? Or are you daft?
Hamish
"Wise men talk because they have something to say; fools, because they have to say something" - Plato
Unix's future depends entirely on its flexibility -- the ability of its core functionality to adapt to the changing face of computing and computers. If Unix becomes irrevocably tied to a specific type of hardware, or a specific type of low-level interface (i.e. libraries, protocol stacks, APIs), then it loses its ability to flow with the river of change.
An example would be Microsoft. At first, MS ignored the Internet, thinking that it was not all that interesting enough to devote resources away from projects like Microsoft Bob. Oops, the Internet turned out to be a VERY big thing, and MS was forced to devote massive resources in catch-up mode. Luckily for MS, they had those massive resources, or the MS today might look remarkably similar to the Amiga of 1996 or so -- desperately hanging on to the "glory days" like a 1960s era hippie still wearing tie-dyed tshirts.
Will Unix be replaced? Not likely, but I say that not from any particular conviction that the Way of Unix is better or a Holy Jihad Against All Things Not GPLed. I tend to think of Unix as a big box o' tools, and as such are more flexible than one big tool that does everything. It's the difference between a workshop and a factory. Given enough tools, a workshop can do anything. A factory, on the other hand, can do what it was designed for very, very well, but needs to be retooled if that thing it does changes. Retooling is expensive, complicated, and fraught with errors. Conversely, if your phillips head screwdriver won't screw those new star head screws, you just go buy a star head driver, and continue on your merry way.
Will Unix change? Almost certainly! A heirarchal filesystem is great for a machine, but normal people (i.e. non-nerds) don't neccessarily think in those terms. They may think in terms of jobs, projects, letters, scans, or big-busted redheads in latex. Arbitrary grouping of data (I think) will be the next Big Thing in user-computer interfaces. Since Unix likes heirarchal file systems, there may be a tool that communicates between the user and OS (like a relational database) grafted onto Unix. Similarly, the distiction between the OS and an application will become blurred -- since most people don't care to differentiate, why should we make them?
Will something else that does the above replace Unix? Not likely. Mostly because the free Unixen available now, and free (speech) software available for it, gives a paradigm-seeking person the ability to start halfway up the ladder. As far as I know, Moore's Law hasn't been revoked yet, and system speed will continue to move innovation away from hardware to software, and as such allow Unixen to graft such additions on without a significant performance hit.
But will it still be called Unix? No. I can say that with conviction, because I have never used Unix. I've used Linux, Open/Net/Free BSD, not Unix. The Great End-User Operating Environment will probably be called SuperHyper Infobahn Facilitating New and Improved Bitchin' Cool Thing, or some other nom de plume that isn't as scary sounding as "Unix". But it will probably be based on Unix, nonetheless. At least, that's my worthless opinion.
Potato chips are a by-yourself food.
I have had limited time to read these posts but I think many people sell MACH short. Microkernel seems to be the way things are going. Modularity in Linux is very popular and necessary to support all the devices that it does and future ones (The kernel can only be reasonably big and still be worth while). Microkernels get away from this monolithic design. I admit that they have yet to prove themselves agains UNIX but I suggest we turn our eyes to the GNU HURD project as I think they will have something to reckon with in a few years.
I really don't understand how this post got on Slashdot... It sounds like something I would read on the Microsoft site, where Unix is defined as the operating system used in the 1960's.
--
Empirical evidence, I think, speaks for itself. 30 years in the running and no sign of 'death' yet. I don't believe any other OS in computing can claim that distinction. Unix is versatile, appears on multiple platforms, and is gaining ground even today in the server market. I'm betting it will continue to survive for another 20 years. Infact, as soon as I raise enough capital to get into the stock market, I'm going to do just that: bet on unix.
really? i find your professor's statement hard to believe. my understanding of the way that people teach computer science at major universities (i attend one of the 5 largest universities in the US, and am enrolled in their top-10 CS program) and certainly my direct experience is that you are taught theory: how to program, how to reason, and how to problem solve in different domains. certainly, there are the low-credit "c++" or "lisp" classes where all you are learning is a language, and the professor can (but hasn't, in my direct experience) specify a compiler/IDE for you to use---but mostly one is asked to submit code that complies to the relevant standards so that any compiler can compile them, and that has certainly always included unix compilers. in fact, the only class in which i was required to stray from unix compilers (from "pc" to "gcc/g++" to "clisp") was the assembly languange & computer architecture class, in which we were forced to use an assembler that only barely ran under the Mac emulator executor and forced me to hit the lab to use an actual Mac.
i hate to call your professor wrong without meeting the guy, but CS is more about how to think than how to implement a solution, and there is no good reason to be platform-specific during the learning process. (there are several bad reasons: TA/professor laziness, microsoft has subsidized all of your labs on campus and there are only windows machines available for development, etc)
of course, i could be misreading your post, and you could mean "Windows type development models" to be something completely different than "development required to be under a Windows OS"...if so i hope you'll correct me =)
Nothing short of an asteriod collision will prevent unix for surviving.
20 years of existance seems to me to be proof that its the most likely operating system to survive the next 20 years.
And how are we defining Unix? To some degree, Windows NT could be seen as a version of Unix. Windows NT is the future OS of MS. Mac OS X is unix based. So is linux! Is there any other widely used operating system in use today that doesn't contain a large number of ideas from Unix?
The internet is helping Unix. Certainly a network/distributed computing system would have Unix in it at some level.
Know what I like about atheists? I've yet to meet one that believes God is on their side.
Thanks for the pointers... however, I guess I should have qualified my (random) comments a bit more... (and read the article more carefully :)
:)
I am fully aware of BeOS and have a grasp of some of the other projects you and others mentioned... Hurd is interesting, and I assume you could use the "capabilities" of Eros to provide similar functionality.
I guess I just want my dream OS now... I guess it's time to start coding
And I still have a sneaky suspicion that current projects are not quite what I want... I don't see any mention of background AI/learning algorithms that automatically troubleshoot and (inobtrusively, intelligently) anticipate my every move. Drool.
Complexity Happens
Some will outright hate it.
But, when you say Unix, do you mean the command-line interface OS designed for use by programmers on 110 baud teletypes and VDT tubes? Or is it the OS with the marvelous scheduler and task manager? Or maybe it is the OS that has adopted the X11 interface as it's defacto standard?
Unix means many things to many people, you may want to clarify, so we can accurately explain why Unix will not die...
Ken
Pining for the fjords, is it? :)
My
I was complaining to a friend the other day that our civilization is still reliant on wheels for transportation - this is *bronze age* technology and we're still totally dependent on the damn things. Why aren't I hovering or jetpacking to work yet? It's the year 2000, for chrissakes!
My friend pointed out that the wheels of today aren't altogether like the ones that graced Assyrian chariots. They're still circles on poles, but advances like pneumatic puncture-proof tires, shock absorbers, independent axles and all the other technologies that surround them have rendered them a very different beast. And while hovercars and personal jets are technologically *possible*, when it gets right down to cost-effectiveness and reliability, nothing yet invented beats wheels.
UNIX is very much the same way. It's old technology, and at its beginnings, there wasn't much to differentiate it from the sledges and log rollers that came before it. But its inherent virtues and inborn flexibility allowed it to grow, to take on and integrate new technologies, and to remain at the base of most of the best technologies out there -- and, in fact, to be the platform on which most innovations take place. It has changed, and it will continue to change, but I'll bet our civilization outgrows wheels before it outgrows UNIX.
--
perl -e '$_="06fde129ae54c1b4c8152374c00";
s/(.)/printf "%c",(10,32,65,67,69,72,
$_="06fde129ae54c1b4c8152374c00"; s/(.)/printf "%c",(10,32,65,67,69,72, (74..76),(78..80),(82..85))[hex $1]/eg;
You can never have authentication portability to untrusted networks, because, by definition, they are untrusted. OTOH, it should be possible to set up (with existing tools) a globally distributed LDAP or NIS+ tree to accomplish the reasonable part of what you're trying to do here.
There's a clunky way that I can think of to handle validation over untrusted networks. It adds complexity but seems fairly robust.
Give each person a portable computer. This computer serves as their access point to systems they come in contact with. It has hardware-assisted strong encryption, and can digitally sign anything (ideally everything) that it sends out.
This tells every system in contact with the user (direct or via an untrusted network) who the user is in an unforgeable manner (the public key distribution problem is left as an exercise to the reader). The untrusted system can't forge the signature, and can't even read the packet if it's encrypted for the intended recipient.
Now, you run into serious problems with this when trying to run applications on the untrusted network as opposed to just using it as a communications link. I'd be interested in hearing about anyone's ideas on letting the untrusted network run software that processes private data without being able to see the data...
You are saying that Unix does not allow one to create a temporary file anywhere other than where TEMP indicates? I wasn't aware of this, how restrictive.
If an application developer ignores procedure, that is hardly an OS problem. How would you react if I said that Linux sucks because I installed an app that keeps all of its files and libraries in /etc?
I have two answers to this. One, UNIX programs often can't create temp files anywhere other than where they're supposed to, or in the user's directory. Everything else is unwriteable by the program.
Windows has a tradition of applications assuming they have the local hard-drive to themselves. Application writers for Windows often write as if this were the case. As you point out, this IS largely an application problem, and some of the most applications that most flagrantly ignore the TEMP variable are Microsoft's own. It's difficult to maintain a convention when the vendor's of the operating system itself don't seem to be able to.
FYI backward compatibility is a good thing. I don't mind if older filesystems (FAT) are supported. I certainly have no problem whatsoever using filenames of any length.
This is largely an OS maturity issue. In the dark ages, UNIX had a 14 character filename problem because directory entries were 16 bytes apiece (2 bytes for i-node #) on older UNIX filesystems. The last time I saw any code that had accomodations for this was about 5 years ago. I assume that 8.3 programs will also eventually disappear.
This is a great advantage. As a programmer, I should not have to worry about whether someone deletes my currently open file from under me. If I were using direct disk writes, then this would be disastrous. Many system processes (on any OS I'm sure) also maintain open files. I certainly wouldn't want to disrupt those.
However, it is sometimes desirable to do this anyway, so Win2K has taken steps so that new DLLs can be installed without having to reboot or close the apps using them.
This demonstrates a lack of understanding on your part. I can delete a file out from under a running process, and the process won't be affected unless it closes the file. I often use this technique when installing new copies of an executable. After being burned by having 'cp' write over my currently executing shell with a new image, and having my shell core-dump for it, I now have a policy of removing executables before I install the new one. All the currently running copies of the old executable continue running just fine after it's removed because it isn't actually removed from the filesystem until the last thing that has it open closes it.
I consider the fact that Windows doesn't do this to be a legacy filesystem design problem that should've been revisited and fixed in NT. It's strongly reminiscent of DOS, which had a very simplistic view of what a filesystem was for. The W2K solution to this problem is a bletcherous wart. I didn't know it did this, and was most disgusted when I learned of it.
Need a Python, C++, Unix, Linux develop
Right -- See Ballmer says Internet is backward, where Microsoft argues for more client control of web content.
Basically, Microsoft sees the industry trending away from client-based solutions (where their revenue stream is), and is scrambling to try to develop more and more client services to stay relevant. AKA, Netscape was right, Windows will eventually be obsolete.
Of course, we may not like the alternative where client control is superceeded by the wishes of content providers (JavaScript, ActiveX, and cookies are already steps in that direction).
--
Business. Numbers. Money. People. Computer World.
Well, the classic problem with a new OS is "Where are the Applications?!?".
RMS, GNU, Linux, Alan Cox are all smart guys. They could have concocted some new thing. But nothing would run on it, and nobody would be that interested in using it. By cloning Unix, they opened the door to 20 years of existing applications. (And thanks to Unix's hardware independance, lots of existing source code.) Unix might not have been the best choice, but it's one that lots of people like.
--
Business. Numbers. Money. People. Computer World.
Unix will never die, and here is why.
Superior Uptime- Name one non-Unix based OS that can beat a unix based OS for number of hours of uptime. Uptime is money, and if your server is offline, you are losing money. And the thing that drives this industry, or any industry for that matter, is the almighty bottom line.
Scalability- You can run Unix on damn near anythi ng out there. Alpha, Sparc, x86, PowerPC, and the great thing about it is, for the most part, that means you can quickly and easily port your files from one platform to another, not to mention embedded systems.
Multiple User Environment- Let me give you an example here. I work for a large, multinational office supply company, and we have a catolog sales division. The way this division is set up as far as hardware goes, is one box running Unix, and about 200 dumb terminals that log into that box. Yes, they bitch that they cannot go online etc, but they are just logging items from the catolog to be delivered, they do not need anything else. And these terminals are a hell of a lot cheaper than a networked PC. Even if you are networking PC's, however, you cannot beat the reliability of a Unix server.
I could keep going on and on and on, but foe all these applications and more, Unix is the best, if not only choice, and unix is so adaptable, it wil continue to rise to future challenges in the computer industry
You say you want a revolution....
I would love to see some of the best coders and operating systems people put together a new OS from scratch using the latest techniques. Ideally this would create an ultra stable and very modular system.
Check out BEos
You just may be impressed, now if only people would develop for it....
You say you want a revolution....
UNIX has been open most of life.
First ATT almost free, toughened up by Bill Joy
& friends at Berkeley. Now the Linux version.
Most of the original computers and hardware companies
in the 70s and 80s have bit the dust, but UNIX lives on.
Unix is like that Chevy V8 engine. Basically the same product since it's creation, but improved over the years. Like, more horsepower. Unix will last forever. It will adopt new architectures. Embedded Unix is finding new markets. Heck, if Win CE finds new markets, surely Unix can.
I have to admit upfront that I am a UN*X junkie.
True UNIX written by AT&T Bell Labs, is probably dead, but it has morphed into many different children UN*Xes which have spun off children, etc There are mainstream UN*Xes like Solaris, HPUX, AIX, IRIX, Linux, & BSD. Then there are some that have fallen to the wayside like SunOS (before the Solaris version), AUX, etc. There are still out there true SysV R4 UNIX running in production enviroments, whereas overall it is dead. The vendor does not even support it. There are children of SysV R4 that are live and kicking today. So, will UNIX die? it depends on how you define the word UNIX. In the original AT&T UNIX, then yes it will die if it had not already. But with the children O/Ses being spun off the UN*Xes then I do not see it going away anytime soon.
Microsoft is doing the same thing (sorta), the release newer and newer O/Ses that are children of their previous O/Ses. Will Windows go away anytime soon, probably not. Windows 3.1 or Windows 95 or a specific version/flavor of it will go away in the not too far away future (year, years, whatever). But overall Windows like UN*X is here to stay. They are both good O/Ses, but....
At this point I piss off all the Windows fans, by saying I will not put anything mission critical on Windows 9x/NT/2000. I do not see it as a stable enough platform. On the other hand, I will run mission critical stuff on 6 year old UN*X hardware and not blink an eye. Or even better, steal the hardware from the Windows 9x/NT/2000 machine and load Linux on it.
That last paragraph is probably a seed for a flame war. Which is not why I said it.
Thanks, that is my $0.02 worth,
Scott
Scott
C{E,F,O,T}O
sboss dot net
email: scott@sboss.net
I am 39.0% slashdot pure
Scott
janitor
sdn website family
email: scott at sboss dot net
The original unix was an implementation of these principles - the first CS project to gain admiration from pure mathematicians
this ring a bell...
possible.
There is the famous comment "Those who do not understand Unix are condemned to re-invint it poorly". This isn't quite as trite as it may first appear... If you don't understand the reasons why Unix is an excellent OS, you won't be able to produce a better implementation.
Who would have thought way back then that a really neat implentation of a grapical user environment would consider it to be a file. X, with its network-transparent nature fits so neatly into the unix paradigm that it could have been designed that way. It wasn't. The crucial point here is that we don't need a new paradigm to cope with the sudden demands, we just use the existing Unix interfaces. This is a hallmark of a good design.
Enough,
Simon
Physicists get Hadrons!
> GNU has likewise changed what was Unix, and,
> despite it's acronymic denial, has become
> Unix.
You mean it doesn't stand for "GNU's Now Unix"?
Or "GNU- New Unix".
Now that is what I call foresight!
--
In the land of the blind, the one-eyed man is kinky.
Distributed systems are already being used. In the group of the division of the company I contract to we have several hundred top of the line UNIX workstations. we also have a very large codebase that needs to be compiled on a regular basis. during the evenings we do distributed builds using all of the clients. It is gorgeous - a 500 CPU build server.
Distributed systems are very real and very useful. If we could not do distributed builds we would be spending upwards of 1 Million to buy a very high end UNIX box to do our builds on.
Now we have a 500 CPU server _and_ all of our developers have top of the line workstations to boot. Great stuff
-- I need more coffee. It's Monday. There is no such thing as enough coffee on a Monday.
I don't see any possibility of Unix passing on.
Ok, some flavors of Unix may outlive their usefullness, but this is the same as say some obscure distro going away, the core of unix and unix like kernals is too usefull, too easily adaptable, too configurable to go away. With the advent of Linux and its dirivative products, Unix and unix variants will be around for a long, long time to come.
Run. I like water. Push My rutabaga.
OSes die for 2 reasons: 1) lost competition with other OSes implementing the same functionality, 2) inability to adapt to new hardware or applications due to the original design choices.
Banyan died because of 1). MS-DOS (aka CP/M) died because of 2)
I think that Unix had a very good skeleton design (just read Maurice Bach!) to face some insurmountable obstacles. (Some) Existing implementations are also of high quality and fare well under Microsoft's weight.
But all modern computing requires programmers. And as programming gets more and more expensive, I could see how some sort of A.I.-based self-programming computers will emerge in a remote (or perhaps not-so-remote) future. This could spell a slow bitter end of all "hard-coded" OSes - not only Unix.
"It attempts to be POSIXish except where POSIX gets in the way"
my other penis is a vagina
I've been running Linux on an old PC just good enough to run X, and doing my computing on my big solaris server for a while. Just like an x-terminal ("Xwrapper -query 'server_name'" in case you're interested. Took me a bit to figue out).
A lot of people do it. It's almost the model for working at Universities. Bunch of x-terminals and a big server. I was talking to an admin in our University Computer Center about it the other day. He says he doesn't even buy x-terminals any more. Too expensive. Just get a cheap PC and run Linux.
Wow, I've been hip all this time and didn't even know it. Fat, thin, shmin, whatever. Do whatever works. That's what Linux/Unix does.
If I had no sense of humor, I would long ago have committed suicide. -Ghandi
That's funny, I thought that Unix was based on a monolithic kernel... silly semantics
... not anything at the semantic level.
A nit, and no one will likely see this, but you mean "syntactic ambiguity"
The homepage with all the details is here
I won't go into the idea of what was thought of before! That's a whole 'nuther can of worms.
Your point is well taken; a lot of the changes are incremental on what came before. That was also kind of what i was getting at as well. Evolution...
But I would argue that there have been some significant changes as well. Look at the journalling filesystems, look at the evolution of distributed computing within the Unix environment. Look at dirstibuted filesystems like tilde, andrew, and nfs. It's all incremental growth, but taken as a whole, it's a "new" Unix.
You can argue it's just new services and daemons. But that's my point. It's the new Unix, same as the old Unix, only it's more capable and has more options. I think it goes well beyond just bells & whistles.
Eloi, Eloi, lema sabachtani?
www.fogbound.net
Yes, but it won't be exactly Unix. Think evolution.
I remember when Unix was a terminal-based OS. X-Windows and the Athena project seemed like a totally new world and a new way of doing things. Sure, it was still Unix, but it wasn't the same Unix.
GNU has likewise changed what was Unix, and, despite it's acronymic denial, has become Unix. But not the Unix from before.
The next Unix will not be today's Unix. But it will be Unix!
Eloi, Eloi, lema sabachtani?
www.fogbound.net
Regarding the issue:
We can build something better now because we know more than the dinosaurs that developed UNIX back in the pre-Cambrian era. (Issue paraphrased from various postings here)
When I was younger (and I've been writing or otherwise producing code since Jr. High, back when dinosaurs roamed the earth and Jr. High schools didn't have computers) I wanted to write or re-write everything myself. I see the younger folks who work for me now with the same attitude. Somwhere along the line I realized that the world doesn't need me to write yet another new library to operate on linked lists.
UNIX is what is it because it has evolved. Great r-wars have been fought in the halls of AT&T, Berkely, Sun Microsystems, The Linux Kernel Mailing list arguing about how things should be done. The good things survived, the cruft falls out. Evolution is a slow process, but it works. As a human, I'm thankful for the Cro-Mangum that figured out how to make a pointy stick. I'm in debt to all the ones that died because they didn't know that it was a Good Idea (TM) to get away from the sabre tooth tiger.
And now I am in debt to all those developers that have come before me, because I can build great things on top of what they have done. We don't have to go write a new OS every time some new hardware comes out anymore. We can port, port, port. Improve, improve, improve.
Starting over, "because we know better now" is a high risk proposition. If a new, primitive (primordial soup style) life form tried to take hold here on earth, it would be overwhelmed by all the advanced competition. This kind of thinking is worse than "not invented here."
I work in the defense industry in the States. I once asked several of my co-workers the following question:
Suppose the US DOD decided they needed to develop their own operating system, so they would not have to rely on Microsoft or Sun or any particular vendor. Suppose the government felt it could have better control over the features that go into an operating system if the government owned it. They could put in all the security features they wanted, they could put in classified encryption stuff right into the OS, they could preload Navy and Air Force theme music, etc.
Suppose the hired a contractor to do this. One of those really big defense contractors. The government could write up all their requirements for an operating system (A project that would take years itself) and have the contractor build it. Could such a project develop something anywhere as good as Linux, or FreeBSD, or (gosh) Windows?
I think not. Because the way the government works, they would drop a truckload (literally) of requirements, and want an OS built to meet all of those requirements all at once. They certainly wouldn't let a system evolve from simpler systems.
My rather long winded point is that UNIX, Linux, etc are successfull because they've started off small and evolved. No one can sit down and develop something as complex as the current state of Linux on one shot. Linux started off small and snowballed. So did the original AT&T UNIX.
Don't fall victim to hubris. Respect your elders. Stand on the shoulders of giants. Build up from what has come before. Wash your drapes every spring, and sweep out the dust-bunnies every fall.
Shit... you use cat? YOU WUSS!
Real geeks pull out the magnets and torx, open the drive in a clean room, and write the inodes with a magnet.
cat? HA!
Give me magnets, or give me death!
"It compiles, SHIP IT!" -Overheard at Microsoft's development lab
Death of UNIX predicted! Film at 11.
All those Apache servers are probably really running NT according to that logic
Heh, well my apache server is running on windows98! none of that NT crap for me!
http://delmoi.dhs.org
ReadThe ReflectionEngine, a cyberpunk style n
Uh, I'm running apache on this box, and it's windows 98, and it works fine.. and its not running with huge numbers of instances, its running with two, but one of 'em has 51 threads. no win32 dosn't have fork, but I doubt that it would be that hard to rewrite for threads (since they did it...)
ReadThe ReflectionEngine, a cyberpunk style n
Mac for graphics
Please, all the major apps have been ported to windows by now, and PC's have always been the way to do 3d on the cheap (as opposed to sgi boxes). You don't need a mac to do graphics if you don't want.
ReadThe ReflectionEngine, a cyberpunk style n
Think about how much IIS sucks ass. Apache is a really elegant system, even on win32. Plus its free.
ReadThe ReflectionEngine, a cyberpunk style n
I dont want to hear any more about policy crap in NT. I just spent all day fixing a policy problem on a NT terminal server. The TS had a hung, non killable program running under the user. The only way to fix it was to reboot the server. NT ....what a piece of crap.
But, do you want to have your data at home? Or at some crappy internet providers' computer?
Sendy
There are a lot of very nice ISPs.
GNU guru and mainframe hacker
I've been thinking about this recently, specifically about the long time that UNIX has been around. I grew up with half a dozen different OS's, from Commodore to DRDOS to Win32. I got to thinking that UNIX is 30 years old the other day and stopped in wonder for a second. It really is an amazing acheivement for the original UNIX architects to have created something so useful for so long.
The article's comment about distributed computing systems is right on though. Unix as we know it, and single processor machines for that matter will die. It won't be much longer (less than 10 years) when many households have more computing power in their refrigerators than you have on your desktop now. All these devices (fridge, washer, dryer, toaster...) will be networked together. The processing power from these devices may be used to drive an AI that runs your household.
UNIX in it's present state is not able to handle multiple processors of different architectures (I may be wrong here, but bear with me please). We'll get to a point where we can run a processor emulator in tandem with a processor's normal tasks, using that emulator to provide clock cycles to a distributed computing function.
Think of the unused clock cycles in downtown Chicago alone. There have to be hundreds of thousands of PC's that are only used 30% of their lifetimes, putting all that power to waste.
But yes, IMHO, Win9x and NT both have their places in our multi-os world =). That place is, of course, in the dumpster (tounge-in-cheek).
Serriously, Win9x is the platform for which a distressing number of games are targeted. I don't know that this is because it's particularly well tailored for this purpose, or if it's the result of driver availablity and that ubiquitous DirectX standard.
-rt
======
Now, I think it would be GOOD to buy FIVE or SIX STUDEBAKERS
and CRUISE for ARTIFICIAL FLAVORING!!
AAAAAAACCCCCCCCCCCCKKKKKKKKKK!!!!!!!!! My G*d, man, we fought the Powers That Be for a very long time to decentralize computing resources! Now we're swinging back to it?
Maybe IBM porting linux to an S390 wasn't such a bad idea after all...welcome back, Big Iron!
James
to add my 2 bits to the 2 million already here,
i vote for BeOS for desktops (audio / video streaming work), and Mac is OK too (the Adobe / DTP box). both could be very good (and user friendly) for "Office use" as well.
as far as servers go, maybe application specific installs (ie. webserver only, or fileserver only, or db only, or any combination) of Linux or FreeBSD should be fine - chosen on install. but i suppose something written from scratch would have performance advantages (?? do Unix's have legacy software/hardware support the same way that Windows does? probably not to the same extent)
This is the biggest wank-fest I've seen in ages -- everyone shouting as loudly as possible: "Not UNIX! UNIX *can't* DIE! It's too flexible! Too modular!"
Okay: is everyone done? Good.
Clearly, UNIX isn't going anywhere *soon*. This is simply inarguable.
But it "can never die"? It's somewhat shocking (and certainly depressing) that it hasn't already died. The OS is over thirty years old. Yeah, UNIX is a durable design. Sure. But the fact that it hasn't been replaced by something better -- in three decades -- is more a sign of businesses' reluctance to change than it is a sign of UNIX's perfection. The computer industry is absurdly conservative. The fact that people still use COBOL -- not to mention C (geez, now I'm gonna get it, aren't I) -- is evidence enough of that.
The fact is: the technology has improved. But we still use the same old tools, because it takes way too long for this stuff to move from academia into the mainstream.
You think UNIX is an ideal OS design? You're fooling yourselves. And don't tell me: "It's better than Windows! It's better than MacOS!" I know it is (well, the new MacOS *is* a UNIX flavor). That isn't the point. Why should an OS even be *susceptible* to buffer overflows? Why should it need a call like setuid(2)?
You ought to be *hoping* that UNIX dies.
By crikey, not that Windows 2000 rubbish, but mighty Millennium, gaming platform extraordinare...
Yeah, I really have to agree with that statement.
I haven't been doing the unix thing too long, but I can remember a time before LVM (on commercial unices, not linux.) In fact that was something we discussed at the last HP operational review I attended in January, whether Unix with LVM was really still Unix, or something else entirely. I personally don't think Unix will ever die, it just gets better every day.
and always has been, and done a fine job so far.
think this guy is missing some point or reading
the wrong magazines
the mainframes also became a helluva lot more userfriendly. although the interface on my as/400 still sucks at least it has a qsh shell (posix compliant too!). and you can always create a quasi unix environment on an s/390 now with the linux kernel.
I, for one, don't like the idea of Network computers, unless I have control over the network itself (ie, my computers). I can just see the TOS for your new network computer.
No hacking files.
No mp3s.
No porn.
No video game roms or emu.
No reverse engineering of software.
No files that explan how to buffer overflow.
No files that bash our company, it's software, it's owners, admins, stockholders, or anything that is owned, employed, and/or controled by us.
All encryption keys will be turned over to us.
These rules may change to keep you under our contol. If these rules are broken, bent, or if you look at your webcam wrong, we have the right to delete all your work, your play, and your life.
Yea, I'm going to give up my desktop to put all my programs, mp3, roms, and text files on some Geocities-like ISP that has complete control over everything that I save and look at. Personaly, I think I'll keep my box and it's files right here.
~Louis Blue
two words: symbolic link.
Why aren't you encrypting your e-mail?
Wouldnt killing X work? How about logging in as root killing your login?
War is necrophilia.
Hum, GNU and Linux are not guys.
In short, the following will occur:
1. Unix will not die, although closed source Unix may. Expect open source mixtures of Unix to survive (Linux, BSD), as well as formerly closed source Unix to transmute into partially open Unix (Sun).
2. Unix GUI shells will change, evolve, die, and change as do the colors of the rainbow. Different GUI implementations, both closed and open source (and some both), will evolve to meet the needs of the desktop, the server, and the appliance. Many religious wars will be fought over these; countless flame wars will spread; much emotion will be garnered. The winners will announce that they knew that they would win and were better - this will only occassionally be true, however. Baby Bills will produce GUI shells - some may survive.
3. The future interfaces of technology will be both totally different and much the same. Someone from 2000 will be surprised at how similar things are in 2020, and yet, how different. Exactly what this will look like is anyone's guess, but prognosticators will be paid ridiculous salaries to be mostly wrong about this.
Will in Seattle
I believe that you can never kill Unix. It will just change to another variant. Can you imagine an internet run by MS-DOS or WinNT. Every DNS server out there could become Windows98. I just don't see a more reliable and adaptable system available. I would not consider myself anti microsoft, but on the server level they cannot compete as the foundation. Unless something really changes.
Jimiz
Two large powers, the US (UNIX) and Russia (Windows) existed with different philosophies about how people, information, and property should be treated.
Russia (Windows) thought that there should be central planning, and that one group of people could do a 5-year plan for the whole country of 300 million + people.
The US (UNIX) by contrast, thought that people should be able to determine there own future and that companies should have to fight it out to meet the needs of the market (the customer).
Russia (Windows) thought it knew best how to plan for all of its people (users), but it could not possibly keep up with the needs of the market, since the needs of the market are so diverse.
The US (UNIX) market is controlled through companies, each one competing for customers (users). Ultimately, the customers (users) get to choose what they want, and the companies that best meet that need end up being more successful than the ones that do not meet that need.
What does this mean for UNIX/Windows? Probably that Windows, despite all of its features, prcing, etc, will NEVER be able to meet the needs of the entire market. UNIX/Linux, however, is backed by several companies all looking to serve various PARTS (not the entire) market. As such, those companies should be better able to serve the needs of their market segment.
Windows has attempted this through NT,95,CE, etc., but they all use a common (more or less) interface. What about users who want a different interface? Microsoft CANNOT meet their needs until it changes it's strategy.
End result? Communism's central control gets beaten by free markets, just as Windows will *eventually* get beaten by UNIX/Linux.
=========
The significant problems we have cannot be solved at the same level of thinking we were at when we created them.
- Albert Einstein
unix will never die... it cant... not unless C dies with it. i feel that unix is simply the natural product of the minds that created C and ultamatly the seed from which the structure of our technology has been assembled.
unix wont die, because the people who choose to use it, due to its practicality or otherwise, will never give up the power that unix gives them. furthermore, they will not give up an OS that fits the way they think and they way they wish to deal with information.
basically, its the same reason that Reverse Polish Notation will never die... it can't, because it works better for the people who use it than anything else would.
>you forgot:
>
>and one for playing games
>
>Read: http://www.microsoft.com/windows95/
2 prominent (in the community) BeOS game houses:
http://www.wcdesign.com
http://www.ngent.com
(though the wcdesign site seems dead right now...)
Sponge
"Hey, I'm still hardware limited. I spend and I spend on hard drive space, but still I can't have all my files on my hard drive at once."
/media or /graphics or /media/graphics or /cartoons or /Scott_Adams? Does you South Park videos go under /South_Park or /media/video or /media/video/quicktime or /media/comedy or /Trey_Parker or /yr1999? Data is just too rich to fit into a one dimensional hierarchy. Data has characteristics in multiple dimensions: content type/categorization (many of them), data format, age, relevance to other object, etc.
Perhaps you need to get that 75GB IBM drive.
"I disagree. My personal files are gigabytes large, and include plenty of small files - interesting pieces of text, Dilbert cartoons, snippets of code - as well as big things like South Park videos. I pay attention to how I file it in my hierarchical file system and I can always find it again because its location just makes sense."
So, do Dilbert cartoons go under
It's 10 PM. Do you know if you're un-American?
And what are symbolic links but a way of putting an object in multiple categorizations. What is better - having a humongous web of symbolic links, or a resource system transparently backended by a database engine that takes care of the numerous characteristics and associations of the data. One dimensional hierarchies only go so far.
It's 10 PM. Do you know if you're un-American?
so this turned into a bit of an essay
why did we decentralise? why did the pc become so popular? bandwidth (and to a lesser extent processing power) - people wanted nicer user interfaces, those could not be delivered over the network available when pc's rolled out.
processing power just keeps going up and it's now cheap and pentiful... hell throw another server into the cluster.
bandwidth is getting cheaper and cheaper so funtional user interfaces are becoming possible to serve over the network.
lets face it, we took a HUGE leap backwards in terms of user interface when we moved to early neo-centralized interfaces (aka the early www and html forms). now as our bandwidth grows and we learn a better balance between fat clients and thin clients things are beginning to pick up. the web still sucks for delivering half way decent user interfaces (in the thin client form) but as bandwidth gets there we'll see improvements.
there are still very important uses for fat clients - imagine playing team fortress classic through a thin client. and then look at what is possible with those fat clients. the classic notion is that we will move towards a matrix like computer system - well right now multi-player games is the closest we've got and they are not possible with thin client technology.
- tim -= remove "-spam-" from address before spamming =-
A lot of people seem to think that in the future, everyone will have a terminal connected to some big iron running somewhere else through the internet. I don'r know about most people, but I, for one, will never give up my desktop PC. I want that CPU right there, next to my monitor, where I can look inside and see the pretty fans going round and round. I want to be able to go to the local grey-market computer parts store and pick up whatever I want and drop it in.
I don't believe that the days where any kind of terminal will be useful for what I do (mostly games at home, I try to keep the work at work) are close at hand. We've been there, and we didn't like it. That's why we're where we are now. You know the tendancy. More processors in the PC (CPU, video, HDD controller, sound (on a side note, I understand that the EMU10-1K that's in the Live! has some pretty serious punch)) instead of offloading everything onto the CPU. There's a reason we don't like WinModems. Why would this not apply on a grander scale? Everyone gets thier own PC, instead of offloading everything onto the big machine behind the windows.
Give me my PC(s) or give me death!
-Wow. I posted.
-godefroi
Karma: Poor (Mostly affected by lame karma-joke sigs)
http://www.eros-os.org/
EROS, (the Extremely Reliable Operating System) is the future.
When we finally get tired of the dismal performance and unreliablity of kludge piled on kludge, UNIX will become an emulation environment running under a genuinely robust, sercure, real-time OS.
-jcr
The only title of honor that a tyrant can grant is "Enemy of the State."
The era of minicomputers may be fading away, but the power of UNIX is hardly limited to that platform. The development of Mach at Carnegie Mellon University brought forth the idea of a device-independent UNIX that can run on any platform. UNIX runs just as well on the desktop as it does on the mainframe.
There's a bright future for UNIX. It continues to be the OS champion for web, file, application, and database servers, whose importance in the era of networked appliances will only increase. Given the current efforts to port UNIX to the PDA and the set-top box, how can you say it's about to die?
One of them is Plurix, which was demonstrated at the CEBIT fair in Hannover a few weeks ago. The main concept is a distributet address space:
When creating distributed, user friendly systems, most of the work is hiding and overlaying the OS concepts with new ones. E. g. using stub objects and transferring the commands and data over the net, where a stub caller accesses the real object. (this way JAVA-RMI and DCOM work). This is overlayed over a local OS, that even builds walls between the applications on the same machine, creating the need for translation and serialization of data, code and requests.
The main reason for this is to protect apps from "bad" apps. Most of the bad apps result from programming errors, very often caused by pointer arithmetics (e. G. off-by-one-errors), Alloc/Free problems and wrong casts. Especially C and C++ are very dangerous in this area.
Most of this could be avoided by using type-safe languages like Oberon or Java with automatic garbage colleciton, and validating the code when linking it into the memory. This way, you can spread one mem-address-space over the network, instead of dividing every machine into several address spaces.
Another project which researches for new security concepts is SPEEDOS.
---
plim-plam-plompudding
Win CE is this. It has the stability of the NT kernel, [note for snide users: NT kernel is stable, but common device drivers are not], and the versatility of NT interfaces (the GUI, COM, etc.) but not the bulk and bloat of: a) massive hardware support and drivers, and b) powerful processes that servers need.
It was originally designed for embedded systems and laptops, but it also makes the perfect terminal for ASP-type systems.
I have a sneaking suspicion that the MS X-Box will in fact be the perfect such terminal, and MS may even serve the games via Internet.
I think you are confusing things with win95 (where all filenames are 8.3, but files have a "long name" associated with them for display purposes.)
In NTFS, there is no such restriction, but you can still name a file 8.3 if you want.
I'll note at this point how wildcards are better on NT in some respects. It is sooo handy to rename a bunch of files by going:
ren Foo.* Bar.*
or even
copy Foo.* Bar.*
but to achieve the same effect on unix, you have to go through a rather complicated procedure of for loops and expr and so on. Someone should port dos copy and dos move to unix..
This kind of FUD gets "5"? An obvious rush of mindwashed unixers moderated this. Let me answer your points:
It has an operating system whose applications, even those written by the OS authors, can ignore the TEMP environment variable and scribble temporary files where-ever they want and fail to operate if they can not do as they wish.
You are saying that Unix does not allow one to create a temporary file anywhere other than where TEMP indicates? I wasn't aware of this, how restrictive. /etc ?
If an application developer ignores procedure, that is hardly an OS problem. How would you react if I said that Linux sucks because I installed an app that keeps all of its files and libraries in
I have seen the future. The future is filled with continued support for legacy drive letters and 8.3 file names with rename.ini kludges during installs.
FYI backward compatibility is a good thing. I don't mind if older filesystems (FAT) are supported. I certainly have no problem whatsoever using filenames of any length.
Drive letters are quite handy: I often map a directory to a drive letter, and then I can handily switch to that directory just by entering the drive letter. It is also very convenient for copying - no typing out long pathnames, just "copy r:*". For example, at work I map S: to my source, R: to an output dir, U: to user files and T: to documentation. It is very useful.
With Win2k's NTFS supporting hard links as well as soft links, even those who prefer that method will have nothing to be worried about.
The future is an operating system where you have to shell out serious dollars to buy third-party utilities to get around security deficiences in the design of the OS. After all, why fix that pesky virus problem when so many anti-virus companies would go under without that revenue stream coming in.
You merely point out that most viruses are written for Windows, and hence most virus victims are Windows users.
It has operating systems whose file systems don't support the concept of being able to delete a file yet have it not actually get deleted until the last remaining process that has it open dies first.
This is a great advantage. As a programmer, I should not have to worry about whether someone deletes my currently open file from under me. If I were using direct disk writes, then this would be disastrous.
Many system processes (on any OS I'm sure) also maintain open files. I certainly wouldn't want to disrupt those.
However, it is sometimes desirable to do this anyway, so Win2K has taken steps so that new DLLs can be installed without having to reboot or close the apps using them.
The only advantage for Unix that you point out is that user security is tighter on Unix. I won't argue with this, since it is true. However, user security on both OSs is being improved constantly, so let us make suggestions and look to the future, rather than denouncing each other.
Really? Even with a sigkill? (kill -s 9)
I have NEVER had an unkillable process in ~4 years on the 11 different servers I have used and/or administered...
I have heard of zombied processes, but one should very very rarely run into one...
Blessed are the pessimists, for they have made backups.
It's the concepts that underly Unix that matter.
..."
"The wheel has lived well beyond the era in which it was born (the era of the Sumerians) and has survived and thrived
You might not like to read this, but what it sounds like you are asking for is BeOS.
It basically is a next generation UNIX, and more and more UNIX apps are being ported over. In fact, Apache 2 now compiles on it.
Many of you may despise BeOS because it is not open-source, but it is definitely a future for UNIX-like OSes. It is clean with no archaic code for backward compatibility, a dream to program for, and offers an excellent GUI and foundations.
I can only see Linux for the desktop heading in the same direction BeOS is going (which includes Internet Appliances).
EC
"...we are moving toward a Web-centric stage and our dear PC will be one of
EverCode
Acutally, BeOS has all the groundwork for a multi-user system built in. If you look at your files in the BeOS shell, you'll find they're owned by "Baron". Eventually, when they get enough time/demand, they will invest the time to finish the job.
Of course, with its mediocre network performance, BeOS wouldn't be too useful as a multi-user platform. That's set to change with the inclusion of the network system into the kernel (as opposed to the net_server). BeOsNetworkingEnvironment (BONE) ... I think... is the name for the new subsystem.
Check out Free Be in a week or two!
Wah!
... because it has JFS, Journaling File System, which was ported from AIX. So if you don't happen to have a $200,000 RS6000 box or whatever, you can still run the same file system.
I'd say *nix is doing fine.
Gamingmuseum.com: Give your 3D accelerator a rest.
you forgot:
and one for playing games
Read: http://www.microsoft.com/windows95/
Interested in learning Chinese or Japanese? check out Chinese/Japanese-English Dictiona
I find it interesting, if not amusing, that the Slashdot crowd always seem to overlook one powerful OS. I refer, of course, to IBM's MVS. I've been working (and fighting) this and its antecedents for almost the Unix lifetime. Don't forget that OS/360 dates back to 1966 or thereabouts.
True, it doesn't have the file system elegance of Unix and its Job Control Language is a horror designed by assembler programmers but nonetheless, as a secure, robust, transaction processing OS I suspect it can out-perform Unix.
I would be interested to see if Linux/390 (if that's what IBM will call it) can out-perform MVS but I doubt that that will happen any time soon.
I do! That way I can surf the web on my way out. We know that it would have a built in browser.
i thing that unix is only going to evolve as it has in the past. with the success of linux, gnome, kde and such, it might become more user-friendly. another reason that it will be around a while, is that mac os X will use a unix kernel (i think its based on bsd). windows is really based on the same technologies that dos was, which was for wimpy 4bit processors (originally). operating systems can change.
--- Hey, Jesus is coming! Everyone look busy
Agreed completely, EROS is way cool. Now, the orthogonal persistence could conceivably be added to a Unix kernel without changing anything that is user visible--thus it would still be basically Unix, except that you could do things like turn off the computer, turn it back on, and have your desktop up in front of you again in just a minute or two. This is an awesome feature, and it's one I hope more well-developed operating systems start making available.
The security differences, however, are deep and pervasive. Unix security works well for seperating users from each other, and for letting them safely use resources provided by an administrator. However, it makes it darned hard for users to collaborate with each other. Imagine you want to use A's compiler to compile B's program and save it in a file owned by C. A, B, and C have agreed to this. The compiler should only be able to access its own files, and the program, and the destination file.
This is a nightmare for Unix. I suppose you'd have to rewrite the compiler as a server and talk to it over sockets. But on EROS, if I understand right, a standard compiler would work just fine. Instead of giving an EROS compiler file names, you'd give it handles (capabilities) to the two files in question, and the compiler wouldn't be allowed to access any other files except for its own. Voila--users communicating, easily, without exposing themselves the way setuid or setgid would.
Communicating via the filesystem and via filenames is extremely ubiquitous in Unix. Once this goes to something better, IMHO it won't really look like Unix any more.
Conference of Australian Linux Users, June 1998
You'll have to say this quietly for now, but Linux IS the standard Unix
Hey, I'm still hardware limited. I spend and I spend on hard drive space, but still I can't have all my files on my hard drive at once.
Already we are seeing that the notion of a strictly one-dimensional hierarchical file system becoming archaic. Having a system of files is useless if you are so overrun with data that amongst the plethora of files nothing is meaningful.
I disagree. My personal files are gigabytes large, and include plenty of small files - interesting pieces of text, Dilbert cartoons, snippets of code - as well as big things like South Park videos. I pay attention to how I file it in my hierarchical file system and I can always find it again because its location just makes sense.
Hierarchical name spaces are incredibly scalable, if you bother to use them wisely. I think if it's difficult to find stuff, you just aren't using the file system's capabilities to their fullest. The limits of hierarchical file systems are a long way off.
This is why, I believe, UNIX's everything-is-a-file attitude is so successful. Files and directories just work.
everyone is a user /etc/inittab
there are user groups
everything is a file
users and groups own files
r means can-read, w means can-write, x means can-execute
all config files are plain text
start from
That is all you need to know to learn, use and admin Unix, if you're trully computer literate and have average intelligence. Er, sorry, it's not easy for other kinds of people.
UNIX, in its myriad forms, is ubiquitous for a reason. Not only has it evolved significantly since its creation, it has evolved intelligently, with many clueful people deciding its path. Realistically, there are few (if any) operating systems which match the ground-up approach to networking inherent in most every flavor of UNIX. This pretty much ensures its role as the backbone of the internet. It has a strong hold on servers for a reason, and it certainly would not have survived this long were it not architected well. The open-source derivatives which have captivated everyone lately are the most logical step -- they create a kind of 'forced' evolution. Weed out the good and the bad by passing every change through as many people as possible. This will ensure that UNIX in one form or another is around for many years, if not decades, to come.
--frood
Legacy: Adj. Any system that does not meet the minimum hardware requirements of Windows 2000/IIS5, or any hardware device that does not have Microsoft approved drivers.
.sig: Now legally binding!
Le Unix est mort! Vive le Unix!
well,
I read the subject line, and expected an Inferno disussion.
But, no mention so far.
Is there anything better suited than Inferno? I'm not a com. sci. person, more "a sophisticated user" as a friend described me, but at first glance it looks pretty yummy.
Ddraig
> I would be surprised if they weren't sneaking > > parts of Apache to their other OS's, which have > anemic TCP/IP suites to say the least.
What? They're using Apache's TCP stack in their other OSes? Apache doesn't have a TCP stack...
Oh, you use Photoshop, which means you're in Windows, so you can't :).
My Linux webserver is just that; a web server. It does nothing more. It can't play Quake, because it doesn't have X. Sure, it also has FTP, Telnet, and Samba, but that's just do I can update the web server. It cannot play sound, because I didn't put in sound support. I didn't put in a lot of stuff. If I don't want a router, I don't put in the routing code.
This is what Linux does right, and what M$ does wrong. WinNT and Win2K use the same kernel for their different levels (Workstation, Server, Pro, suXor haXor, etc). This is st00pid, because I am not playing Quake on the server. At least on Linux, I can take out that part of the OS out entirely.
Is this post not nifty? Sluggy Freelance. Worshi
I can only say that you are very lucky. Monday of last week, Tzu (our Linux internet proxy) had 11 unkillable Pythons running, 3 dead ls and a host of dead kill commands. None responded to SIGTERM, SIGKILL, or anything else I could think of. I suspect a hardware fault, though, so shouldn't blame the software... At home, I see zombies at least once a fortnight. Admittedly, I have an exceptionaly shoddy motherboard, but zombies arn't exactly rare. One of my textbooks has a 20 line program that creates a zombie process for you. In summary: Zombies and unkillable processes are common on shoddy hardware.
This sounds like the Condor system at the University of Wisconsin, Madison. I have only read a little about it but as I understand, it works as you describe. Check out their homepage at http://www.cs.wisc.edu/condor.
Thanks for the link. Just a quick question does this solely rely on fortran? I am not much of a fortran programmer (read can't even do a Hello, World thing).
Slashdot social engineering at it's finest
Because the topology of the network and the working model is changing from stand alone systems to clusters of systems to distributed systems, does not mean it is the death of existing OS's. OS's, such as UNIX, which are well suited for heavy loads and accesses, will thrive as servers in this new world.
Currently where I work, we are looking to install IBM Network Computers. We're using RS/6000's running AIX and NT boxes as servers, and the kernel for the NC's is NetBSD. And, the really cool thing is that IBM actually has a page on their web site for installing a Linux kernel instead!
UNIX is not dying with the change. It is adapting to fit viable roles on both sides, the client and server, as any good OS should.
Barbie of Borg - She doesn't just Assimilate, She Accessorizes too!
Unix is going to die eventually! After all, it's only a name given for a type of OS. The style of Kernel unix uses or the good ideas people can steal from the unix kernel is going to live forever. For instance, The BeOS is an Opperating system built from the ground up. The Developers were all home-grow on Unix. They made an all new Kernel based on ideas that are derived from Unix. However, the Kernel is so diferent that I wouldn't dare call it a Unix kernel, but a kernel it is non-the-less. Another example is Linux. now Linux is not unix, as we all know, but also uses a Kernel. So in a strange way Unix will live on through its kernel design. What won't live, thank god, is the 20+ year old unix architecture. What I mean by this is that all the baggage, or rather legacy, of Unix will have to change. It will have to change or die. This doesn't mean that Unix doesn't have a few good years left. in fact it won't be going away for like another 10 years in my best my estamite. The fact is that the market is probably going toward small Internet appliances. I wonder if this is what the author meant by distributed computing. These things will need to have an Embeded OS. A kernel is perfect for this. Unix is not! Enough said!
It isn't a lie if you belive it.
Much as I hate to say it, I've seen plenty of unkillable processes under Linux - most oft hem netscape. Not that NT doesn't suck.
"that's not encryption - it's a new perl script that I'm working on..." - from some Matrix parody
I recently did a research project on Ecuador, and used soc.culture.ecuador (if I remember correcly). It contained much clueless noise, but I posted my list of questions and received six comprehensive responses. I agree USENET contains too much garbage, but it's still useful and certainly not dead.
Talk is cheap because supply exceeds demand!
No Open Source OS will ever die. As far as I can tell, that's the biggest strength of Open Source. It changes all the time, better OSes are built from older ones, some distros die off, some outlive their usefulness; but they never die. Sure, maybe Unix won't stay around forever as the Unix we know, but that's the point, isn't it? It's always changing anyways, so if you're going to predict that Unix will die, the only way to be right about that is to say the Unix is already dead. We may not even call it Unix, eventually, but it'll still be Unix at its core.
The only way Unix could ever be said to be dead is if someone creates an OS that isn't based on Unix, but performs the functions of Unix better than Unix itself. I don't see this happening now (though if I'm wrong, I'd love to know about it), and it's not too likely to happen in the future. Unless we completely change the basic structure & function of computing (which would cause a complete rethinking of computers in general), Unix will always be around.
Now, Windows, on the other hand....
Eruantalon
Eruantalon
The Annals of Middle-earth
"Windows is the standard operation system"
I think the reason UN*X will be around for many years to come is that UN*X isn't wedded to any "paradigm." In other words: everything is extremely modular. This allows for evolution in the system to meet unforseen demands. Because of this, Un*x is used in everything from RT embedded systems to consumer desktop systems. So, while I think Un*x in 2050 will bear little resemblance to Un*x in 2000, I will be willing to bet that a unbroken chain of evolution will bind them.
And humans have kept finding better ways to HARNESS Fire! We shall do the same with Unix.
Mike Roberto
- roberto@soul.apk.net
-- AOL IM: MicroBerto
Berto
At this point of time within the Unix community, it[Unix and Unix-type systems] can be compared to a wild animal in the jungle. It is thriving, yet being naturally regulated by nature, and the open source movement. With linux and GPL, it is taking a true stride toward natural equilibrium.
:)
And as animals like this must survive in the jungle by evolving, so will the Unix system. Especially with such a broad background and the amount of powerful work that can be done with it. It will change to meet the future's demand, just as anything else that is free will do. This system isn't going anywhere except the possible changing-room once in a while.
and upon my preview, it seems that i must fix my sig
Mike Roberto
- roberto@soul.apk.net
-- AOL IM: MicroBerto
Berto
What no one seems to realize is that UN*X is a prime example of evoloution brought to an OS. Everytime a new need arises, someone (or a group of someones) ports that technology, thought, or otherwise function to UN*X. It is truly an adaptive OS that IMHO has never been rivaled.
Let's not forget NTv.4 was supposed to be the UN*X killer. Even with Win2k now released, UN*X is still alive and kicking ascii.
-- mikeDOTd
You really think so? I'd think that of all the big commercial nix'es that Solaris would be one of the survivors. IRIX is nice (I use 6.5.5 full time at work on an old Octane) but I think it will go sooner. SGI as a company is hurting and they have made some noises about a move to linux. Sun, on the other hand still cant build systems as fast as they sell (plus Solaris is avalable on x86 as well as Suns native SPARC architecture).
As for Digital UNIX, who knows? The Alpha is a phenominal processor wich I hope dowsn't go away, but DEC Unix tends to be flaky in my experience. Also there is no way to know what compaq's plans for the platform are long term...
-JeremyH
Unix is the cockroach of the computing world. Its too proliferated to squash it, and when humans eventually destroy themselves, aliens will be able to check their email when they get here on it.
Point is, the industry is so big that it is wayy to hard to re-invent the wheel. Its taken too long to watch Unix reach maturity, that even with the experienced gained.. it would take too long to produce a real competitor..Think of how long OS development takes.. some of the ideas for windows 2000 were on the boards before 95 left the gates. Unix has taken 20 years to come this far. Thats an awfully large codebase.
If there is a competitor, we'll see it coming, and so will our grandchildren.
Of course the star trek episodes will get a little more predictable:
[jlpicard to jlaforge: warp 9 laforge...]
jlaforge@enterprise ~ 22:22:11> s jlpicard "ill get right on it captain...it should be right here"
jlaforge@enterprise ~ 22:23:48> man warpdrive
--jay
Will it be replaced? two words: Plan 9, baby, Plan 9! :)
Here's what people are saying.
Where is Unix going? There seem to be three religious sects:
Aside from proclamations of adherence to these sects, some interesting comments include:
Mr. Slippery on the other hand, asserts Unix is basically static and optimized, with a few new gizmos now and then.
eries mentions UNIX is a way of organizing operating systems.
dsplat and jetson123 talk of UNIX's evolving features while maintaining backward compatibility.
jabber points out that Unix and other OS's evolve mostly from using each other's features.
Several people point out the ambiguity of "Unix". A philosophy, command line tools, GUIs, appications and even the organization of file system are candidates.
Others take the opportunity to cry to the heavens for mana:
My Religion
Now, for some words of my religion. Back in school, I remember reading of a truism from the 50's and 60's: Every time you make working with the computer twice as fast, people become four times more productive. Innovation is risky, and has high costs and profits. I believe in innovation, and am proud to have helped at CenterLine, JavaSoft, and others.
Before making new theology, let us consider our dogma.
Unix Dogma
Progress in computer science moves slowly. My computer science mentor taught me that: a new thing must be ten times better to get others to use it. Being thirty percent better doesn't even raise eyebrows.
So what can we do that is truly amazing?
Charles Merriam
Independent Business Consultant
Sunnyvale, California
Profit motivates invention.
Unix will die. As computers spread, more and more people begin to use them who assume windows is an integral part of every computer.
Nowadays you can do everything on windows that you can do on Unix. Whether you can do it as well, as efficiently or as reliably is not the point for most people. Point and click and MCSE are all that matters.
In business the people who hold the purse strings can see no further than Windows. All the new desktop users can see no further than Windows. All the kids learning to use computers are using Windows.
I can't tell you how much I hope I'm wrong...
Unix won't die - it'll be absorbed into other OS's. Look at what we have today - Win2K with almost-POSIX compatibility, BeOS, MacOS X, Linux, etc. None of these are unix (MacOS X isn't - it's based on FreeBSD which no longer has any AT&T code in it) but they all integrate UNIX concepts. Does that not speak for the superiority of the system? Hmm. Ponder that for today.
Oh - and another reason that it won't die - I plan on using UNIX-like OS's for the rest of my life, so there will be at least one person using unix!
"The romance of Silicon Valley was about money - excuse me, about changing the world, one million dollars at a time."
Visit
UNIX was built before networks really existed. It was built for connecting many terminals to one mainframe and working at the same time. And NOT distributing anyting out to other mainframes.
Yes, it can be distributed to a certain degree(a lot more than some other OSes) but not as much as OSes that where built for distributed computing, such as plan 9.
There's a list of other distributed OSes at this page.
Erik Dalén
sure, rsh/rexec works fine as long as you don't try to write to anything in
Another important feature is that you can mount two directories into one! So you could for example mount a fileservers /usr on top of you're own and access both at the same time :)
Erik Dalén
I have seen some implementations of having multiple users in Windows (95 series), where the environment is actually 'protected'. There is one at my high school. It's called FoolProof, for good reason: it takes a fool not to know how to get past it. Why, you ask? Well (1) You can't "Open" anything by using the context menu, but you can "Explore" all you want, (2) It locks you from using the Find dialog, but what about pressing Win+F?, (3) You can boot into 'safe mode command prompt only' DOS mode and edit config.sys and take out the -- get this -- DOS-level protection 'driver', and (4) (here's the kicker) boot into 'Safe Mode with Network Support' (no config files modified by this) and surf the web.
WinNT is another matter. But there are still lots of vulnerabilities.
That's why you need support for multiple users at the kernel level. Unix is staying. I'll bet even Microsoft will someday base its operating on Unix. On the other hand, they have to remain proprietary and closed source, so never mind.
Be, on the other hand, has the support at the kernel level. It should not be that hard to get multiple users really working (I guess you could recompile chown on Be, and su, login, etc. if you wanted to.)
Did you really mean to mention a Unix-based OS in your comment (hint: "Powers That Be")?
IMHO, I cannot see UNIX completely dying out, but I can see the UNIX field possibly condensing into distributions/flavors that are supported by corporations. I don't see Sun going anywhere, at least.
I don't think UNIX would disappear unless a company marketing a Linux distribution grew to the same level as Sun or IBM in terms of support services available. Large corporations using critical systems would not consider using a Linux distribution unless they knew they could have 24/7 support falling all over themselves to kiss their asses.
In addition, I don't see a great deal of "embedded" systems (phone switches, from PBX up to Central Offices) switching from UNIX to Linux. At the least, I imagine Lucent will keep on with Unix - they were the guys who came up with it in the first place. While there are NT-based switches available in the PBX and ACD market, the number of telco managers using them greatly declines each week due to them committing suicide when their switch crashes. :-)
bun-fhuinneog agam!
Just when we get things working they have to go and change it all and f*** it up. nah unix is here to stay.
I can see how this might be useful in some environments, but the fact is, I personally, and many of the people who I work with, don't really need or want that particular capability for most of our computing needs. Nobody else uses them except for me. Also, having a Root account in an OS almost ALWAYS leads to security holes.
I'm planning on trying BeOS when 5.0 ships and is available for free download. I want to try it out before I commit any $$ to it. It sounds fantastic, and so what if it doesn't have multi-user capabilities.
Free music from Jack Merlot.
I know there's a contest to guess the ship date...sigh...
Free music from Jack Merlot.
In a nutshell, I don't think Unix-like systems are disappearing anytime soon.
The open source Unixes (Linux, *BSD, etc.) are the least likely to disappear. Because so many hands are involved in creating these systems, the inner knowledge of how they work is disseminated very quickly. And since the target audience (power users and programmers) are the same people *writing* those systems right now, there's sort of a feedback loop building. The more I learn about Linux, the more I can do with it, and the less I want to use something else. Accordingly, the more I know about Linux, the easier I can hack it into whatever I want it to be. So the skills base in Unix is strong enough to ensure that it will stick around in one form or another. It will certainly evolve, just as Linux is nothing near the first AT&T release of Unix, but the basic concepts are still there.
Speaking of basic concepts -- there's something about good ideas, especially in computers. They're never forgotten, but they're almost never appreciated when they appear. Xerox had no clue how revolutionary the laser printer, the GUI, and Ethernet were. Atari, when it had a chance to buy Apple in 1976, didn't realize that personal computers would become what they are today. Yet all these things came about anyway. Without getting into too much detail (since I'm not an OS guru) I will say that there are some good ideas in Unix and that it is a well-thought-out system. This is why it is still around after 30 years. Contrast Windows 3.1 or DOS 2.0. No one other than hobbyists is a 3.1 "power user" anymore, since all Windows machines are 9* or NT. In five years, NT skills will probably be forgotten since W2K will be the standard Windows then.
Unix will certainly change, but it ain't going nowhere.
"Honey, it's not working out; I think we should make our relationship open-source."
I'm amazed, I figured extolling the death of unix was dead :)
Of course Unix will survive. It will remain a central part of computing for a significant amount of time. What role will it play is another question.
There is no arguing that the Internet was built around Unix. Although Microsoft would like you to believe otherwise, it will remain the foundation for the Internet, possibly forever. It will evolve to incorporate new technology and new methods, but like a true classic car, Unix will retain its soul through it's evolution. From a network and communications standpoint, it isn't going to die.
As a desktop operating system, we'd be silly to jump into this discussion here. Every media outlet with even a slight technical following has beaten this subject to death. Everybody with $35 for a domain registration is trying to "bring Linux to the desktop". How many buzzwords/companies can you think of that have their hat in the ring? You can't watch the news without hearing one of these names, "Red Hat, Apple, OS X, Corel, Linux..." Will it hit the desktop? Of course it will. Will it stay there? Maybe not. As long as the interface and applications evolve as fast as the network support, it has a pretty long life ahead of it.
Why is it that somebody have to come up every three months to foresee the end of unix? Shouldn't prophets see the future of stuff they know abouth? Just to avoir karma noise... :)
Yes, there systems that run Unix that don't any of the user interface that most people think of as Unix, (bourne shell, popular X Window managers, etc.) many having no user interface at all.
But few Unix systems can handle the requirements for "real time"
It is number 4,135,240. Notice since that they described the process in terms of the logic gates of the memory chips involved, since at the time the USPTO prohibited patents on software.
Unix die off? Not likely. Consider: There are over 10-15 million Linux machines, perhaps half to 75% of which are used for desktops. Don't forget the 3 BSD's - there are perhaps 5-10 million of those boxes. Then there's the hundreds of thousands of UNIX systems (Solaris, HP/UX, Digital UNIX, take your pick) used as servers on the Internet. If Unix ever dies off, I sure hope the successor is worthy of its new position.
GCS/E/S @d- s+:+>+:- a--->+++ C++ UB++++ P+ L- E--- W+++ !N !o ?K w--- ?O M+ V-- PS+ PE++ Y+ PGP+>+++ t++ 5 X+
> but now, many people believe that soon we will
> see the transition from that era into the age
> of the distributed/network system
I believe the opposite. Right now, companies are paying 2500* when they build their network, and that 700MHz processor on Jane from Accounting's desk is being about 5% used.
I think diskless, dumb, networked workstations are the future, and what better to use are the clients and servers for all these, then Linux, with GNOME/KDE 3.0. The workstation costs come down to under $1000, maybe even under $500, and management is completly centralized.
For the paranoid, there is always the "distributed computing" where your PC becomes a dumb terminal for "their" (government / Microsoft / MPAA / Satan -- your choice) network. You connect, run the applications from their servers, store your data on their servers, and sell them your soul.
I'm going to live forever or die trying.
Because it is mutable, and hasn't promised binary compatibility, UNIX has survived all the architectural epochs that have killed its competitors.
It's also worth questioning the premise that the distributed world in any way changes the demand for UNIX systems. There will always be a need for big computers with high locality. For the forseeable future, except those running MVS/Sysplex clusters, -all- those big machines will be running UNIX. Someday, many of them might be running Linux instead of the current flavor, but it will still appear to be UNIX.
-dB
"It if was easy to do, we'd find someone cheaper than you to do it."
If the market can't kill NetWare, what makes you think Unix will ever go away? Collectively, the switching costs are too big. Unix will be strong for many years to come. .......... kris
"I thought I could organize freedom. How Scandinavian of me."
Linux is just as bloated as windows
:) :)*
:)
Umm... no
To clarify, I can compile and place the 2.2.14 kernel on a floppy, still have room for a few basic tools, then boot from that floppy, and do just about everything I need to from the floppy (yes I can, I just did it over the weekend to troubleshoot what was going wrong with my new hardware, including the HDD controllers which, for a few bootups, were disconnected altogether. Turned out to be bad RAM, but I digress)
I couldn't have done that with NT. DOS, sure noproblem, except it had a bugger of a time finding the network hardware (which, incidentally, was one of the items compiled in mentioned kernel
believe it or not X and KDE and GNOME and their attendant libs are part of the OS to most people.
And to most people, lemmings are cute critters that will spontaneously commit suicide en masse. Your point?
*-Actually, if you think about it, the fact that *NIXes are inherently networking and distributable, it sort of negates the entire premise of the original argument
--
Xiphos
The groundwork was going to go into 5, but ended up getting ripped out. What multi-user stuff is most helpful for, I think, is shared permissions on files and printers and such like over a network. Until the BeOS gets more support for things like that, why bother? I'm much more excited about the new BeOS Networking Environment, BONE.
What he MEANS to say, is that is easier to crank out reliable code in java.
Plus, depending on your coding style in java, instead of the whole thing blowing up with a SEGFAULT!, the current thing you just tried to do will fail.. but the rest of it will keep running. So yes, in that situation, it IS more reliable
I agree. Unix has evolved to become nearly everything we need it to be and placed in nearly every corner it can possibly go; not every flavour maybe, but as a whole. To say that it will die when it is expanding in so many areas (like the mac) is ridiculous.
kill_9_1
*Shudder*
That adds a whole new level to the term "Blue Screen of Death
--
Intelligence is definitely a recessive trait.
heheh...that's a funny cartoon. appropriate for the topic, too.
but i must say, my first project was writing a serial port driver for a cp/m machine. back then, we used to laugh about how crappy dos was, and how the junky pc's would take 20 years to be as powerful and flexible as a s-100 and cp/m.
well, that was 1982 -- 18 years ago. about a year ago, a working cp/m machine was found in a secretary's office. it finally refused to boot, and someone from the help desk came and took a look at it. she had a notebook of sorts going back through several secretaries and about 15 years. the notes explained how to use the software and the machine. the help desk people were stunned. of course, cpm died many years earlier than anyone thought.
we used to also say unix was a dinosaur and would die within years. we thought unix would never outlast that modern digital research operating system. who needed multiuser/tasking when you had cheap $10K machines to throw at anything that needed one? one s-100 per hundred employees was approaching overkill. anything more was a useless waste of technology.
unix is pretty entrenched. the steep learning curve (like mshaft doesn't have one) leaves the phb's in terror. they want mshaft to sooth them with "look at all the mcse's, look at them, easy replacements for your overpaid employees". some employers believe the rhetoric, some still think it's the people that make a company work -- and a good, loyal employee is worth a lot.
times change. i've been seeing more and more of the "microsoft strut" going on lately. it's the kind of thing people do when they read something like "internet week", and see all those microsoft ads and glowing reviews of W2K, and all the testosterone-laced comments from stevie ballmer about freedom, innovation and vigor, and the various nimrod bylines. they kind of "get high" on it, and walk around repeating the things they read like some kind of holy mantra.
of course, i have a dark side of my personality, so as i listen to the speech, i can almost see the shirt turning brown, and see the swastika, and i think of that "bill gates as hitler" jpeg that everyone loves -- if i really relax, i can almost hear a german accent describing how much money microsoft will save us. it's kinda fun.
of course, the last company i worked for went microsoft. they had the pleasure of transitioning from a system with a bunch of cheap pcs and two unix servers, to a bunch of NT servers and a slew of high end pcs. their "microsoft bill" (software licensing) was $150,000.00 -- $100,000 more that for the proprietary unix machines.
overall, i wish i had done something other than be a programmer. i was pretty good at biology; i think i would be happier there at this point. i really dislike seeing my industry on the front page of all the magazines, and having my profession described by the various microsoft pundits as being overpaid, protecting my job by utilizing a mystic/cryptic technology, etc.
who knows what the future will hold. when i first started programming, they used to say any idiot could do it, and programmers would never be paid more than $10 USD an hour. they also said women were far better at it, and that nearly all future programmers would be women. women worked harder, were more stable, and easier to get a along with than men.
it can be rather difficult to predict the future.
Treatment, not tyranny. End the drug war and free our American POWs.
See my user info for links.
Hi, As a general comment on all the comments already posted: Todays UNIX systems work hard for POSIX compliance. It's the new hype. It seems like no UNIX today is really interested in earning the name "UNIX" (trademarked by X/Open for standard-compliant Unices as I recall), but saying you are POSIX compliant is really neat - so it seems. What's the result? In my opion, the result is that nowadays every UNIX ships ed and other things that really belong to the past. (e.g. who's interested, today, in the baud rate of a terminal. Maybe some, but you know what I mean). This is not only valid for POSIX; UNIX has anyway a habit of keeping op with old traditions. While I must agree that a lot of the basic principles behind UNIX will survive, I do fear that some of the traditional habits could cause it to get "old", if people will not alter them quick. What do you folks think? Whill this happen, or will we stick to the standards? Thanks,
It's... It's...
"We can confirm that Debian does *not* ship the version with the trojan horse. Our version predates it." [CA-2002-28]
I don't think Unix will die. Operating systems don't die, some merely are forgotten about, rare or even extinct. What's important with the free software revolution is that we have more than just Windows available to us, and as long as this is the case, and as long as we have free software, we have the freedom of choosing our OS, whether it be Windows, Mac, Be or Linux. Each has their strengths and weaknesses. I have a friend who has been standing by macs forever, whereas my dad refuses to use linux because the printer isn't 100% compatible yet. It's a matter of freedom to choose, and with that, nothing, not even Unix, will die.
icqqm [ICQ:11952102]
People have been saying that unix is dead for years, which shows a lack of knowledge of the fact that unix is still the dominant operating system in the world today (or flavors of it). As for network environments, the original distributed OS's were based on that, including sunOS. As of yet, unix seems to be one of a very few OS's that are keen to distributed/network environments. In fact, it seems to be the best suited all around.
Unix is the power of the future. Simple, elegant, adaptable. A command line can be made to do millions of things that it would take millions of separate windows programs to do, from smaller, beautiful component programs. Any drive/file/or piece of hardware is instantly transferred to a network or a new use via a pipe. Older software works with newer hardware, and this will stay the same. Choices of shells, mulituser, multitasking, multienvironment. This is perfect for systems that control our homes, that speak to us, that learn, and that work with each other, distributing tasks, or acting at the core of a household in which all appliances have computer interfaces or embedded systems.
Unix is here to stay, because unix is whatever you want it to be. Where windows says, adapt to me, unix says, I will adapt to you.
Eh...
If you're talking about unix running on servers, ignore this, it probably will change to distributed computing. But if you're talking about unix and linux running on all computers, you're wrong people (more specifically, me) will always want a personal computer running a stable OS that does what they need and can be customized to work better for them, even when their work computer is a bunch of peripherals connected to the LAN.
They that quote Benjamin Franklin on liberty and safety deserve neither.
If you think Plan 9 is "odd", check out Amoeba, from the father of MINIX. Amoeba is a "true" distributed operating system, much more so than Plan 9. (Also, the source to Amoeba is freely available, whereas you have to pay 250 bucks to get the source to Plan 9.)
Unix hasn't completed it's development cycle yet. I doubt it will until long after it dies -- if it dies before humankind.
I do believe it was 1971 when the first Unix system went online -- back when 'online' meant 'it works' (like the little green light on your printer).
Unix itself, as in the general phenomenon (if you will), as opposed to 'AT&T Unix', the distribution, therefore has been in development for 30 years. Any form of it that stagnates, dies. This became inevitable when unicies with freely distributable source started appearing, people (especially unix users) tend to try to make things better, unless they're paid to do something different.
Kevin O'Brien once said this in response to someone going on about how unix was ancient, deprecated, and dead. Kevin also said "It's not smut, it's data," when the question of webspace contents came up =).
I wouldn't worry about Unix dying off for now. I won't live to see it, I'm sure of that much. There's obviously nothing around now that can kill it. I do believe that it would take an entirely different paradigm to kill Unix -- so micros~1 obviously won't manage to do it, since all they seem to be able to do is take two existing paradigms and make them both worse.
In regards to specific tasks, if a certain kind of task is popular, then (as usual) Unix will adapt to accomodate. Yes, I know there are things that unix cannot do, simply for lack of the appropriate software. But this isn't a technical fault, it's a social problem, as all the examples I can think of involve user-friendliness, idiot compliancy, and streaming media. Linux is getting closer to a desktop OS all the time. Personally I dislike this (I don't want to use what the end-users are using), but it's the truth.
---
script-fu: hash bang slash bin bash
[ approaching AI ]
Meaning a file system (the BFS; you know, the 64-bit, multithreaded, fully-journaled, database-like one) with fully-working owner/group/all and read/wrtie/execture permissions is already in place (it just happens that root (or "baron" as they call it) owns everything for now). Also, home directories (for now there's only one) with both global and user-specific locations for settings, drivers, file system add-ons, and all that jazz are already implemented elegantly, beautifully, and functionally.
In addition, the BeOS's new networking stack, code-named BONE and due out pretty soon (not long after R5) will be "around twenty times (2000%) the speed of the current net_server" and "in the same league as Linux and FreeBSD" (BONE designer Howard Berkley from The Be newsletter, further explained in a BeOS Central Interview and a followup )
And let's don't forget the usual arguments of prevasive multithreading, multiprocessor optimization, high-efficiency modular I/O, the clarity and simplicity of a 100% object-oriented design, and those wonderful multithreaded direct access graphics capabilities.
Given enough market demand (not likely), Be could spend the time to finish the multi-user functionality, OR after what I see as the eventual (but not any time soon) open-sourcing of the cureently free, but closed-source desktop version of BeOS (keeping the appliance version (BeIA/"Stinger") closed-source) under something like the Mozilla license, we could just do it ourselves. Hopefully, if Be succeeds financially in the Internet Appliance market, they will be able convince the stockholders that open-source has enough advantages, and then we'll get to the really fun stuff
Come on, people, at least give the little guy a chance! I sincerely believe that Be could become a much larger player in the OS wars if people would be more open-minded to something that isn't open-souce (yet).
Long live the underdog,
~tide
"Linux is only free if your time has no value."
There's no reason to "convert" most of our existing Internet/networking infrastructure to anything else []
IMHO the libc network interface should be rewritten. All the sockaddr_whatever stuff is just insane.
And why did those guys give unreadable name to most functions? It seems that the more the function is important, the more its name is cryptic.
UNIX has several advantages over just about anything else for distributed networking. It was built for networking and I don't see any reason why, with only minimal improvements, it can't be used in distributed networking.
I'm already working on a program for Linux (it should be portable to other UNIXes) to make it work for this. Its in the very early stages, though. See http://hammer.prohosting.com/~linuxnet.
I apologize if this seems like a recruiting post for my project.
----------
Not a typewriter
I might veture an opinion on this if I knew what you were talking about. Mind posting some links?
----------
Not a typewriter
But UNIX is not one of them. I have been hearing that UNIX is dying since 1989 when I worked at Everex. They had an OEM version of ATT SystemV, Xenix was popular, and Windows was at verson 2.03 or so. Novell was pushing the distributed model, and held the top position in server software for years. But what happened? First Novell bought UNIX from ATT, later the internet brought UNIX back to the front line. The popularity and sensibility of unix is directly related to the growth of the internet IMO. UNIX is very unlikely to die anytime soon. It will more likely continuye to evolve at an ever increasing pace.
I think the thought behind unix is so simpel and powerfull it will adapt to any system that might be used in the future.....
Grtz, Jeroen
Secure messaging: http://quickmsg.vreeken.net/
Unix will evolve. Linux is a flavor of Unix. Many people use QNX on the Iopener internet device. Many people run OS/2 on their ATMs without realizing it. Many companies use cash registers based on DOS or OS/2.
You may not see lots of articles or advertising on Unix, DOS, or OS/2. That's not because it's not being used. It's that it is 'stable' and you don't have many new things comming out for it.
Fight Spammers!
whoa, sweet. can the homosexual men be homosexual spanish teachers who pretent to be mexican?
-cesar (i love you alysha)
It's my position that the same strengths that made UNIX a great OS in the 1970's will make it a great family of OSes well into the future. What are those strengths, you ask? Well, read on and find out.
1. UNIX kills at networking
Always has, always will. As the story mentioned, the 'net was built on UNIX, and continues to run on it. The Internet, the TCP/IP protocol... UNIX has the home feild advantage here, and I don't see any signs of an upset in the future.
2. UNIX has superior reliability
UNIX was originally created by a telco company, and was made what it is today with some very significant help from universities. When a phone grid goes down, or valuable academic information is lost, it can be devastating. Therefore, UNIX was coded for rock-solid stability. This, along with reason #1, is why UNIX runs most servers. As we move into the new era of embedded devices, reliability once again takes center stage. There is a preexisting standard of reliability for home appliances, and people aren't going to care if they can browse the web from their toaster if there's a driver conflict with the heating filaments. I happen to know that UNIX exceeds these standards, since I trust my Linux/BSD boxes much more than I trust my frequently misbehaving kitchen.
3. UNIX is modular
While most Unices are based on monolithic kernels, UNIX is still one of the most modular OSes out there. The UNIX design philosophy has always been one of small, fast, interchangeable programs over large, clumsy ones. Therefore, UNIX can be trimmed down to exactly what you need it for. A microcontroller for say, a pacemaker, may only need a kernel and a single user-space daemon. A PDA needs a suite of productity applications. A desktop needs more. A server needs the server program itself, but not most of the overhead found on desktops or even PDAs. UNIX can be adjusted to the needs of all but the most rudimentary computing devices.
4. UNIX is scalable
This ties in with the last question. If UNIX can run on all the platforms mentioned, that also means that developers can write software that works across all of them. It also means that users can enjoy a common interface on all computers.
5. UNIX is open
While not all Unices are open-source, many are. All the conventional open-source arguments apply here (all bugs are shallow, development is fast, reliability is increased, and hey, freedom's cool!). If UNIX is replaced in the future, it will almost certainly be by another open-source OS, otherwise it will be virtually impossible to compete on a technological or political basis.
6. UNIX reinvents itself
UNIX isn't a killer OS on microcomputers because it was a great OS in the era of minicomputers. It's an awesome OS today because it reinvented itself for the times. Since UNIX is in most cases an open platform, new features can be added whenever a hacker has a really neat idea. The BSD guys did it with TCP/IP. The X consortium did it with X windows. Tim Berners-Lee did it with the web. Better yet, someone can start from scratch and create a new UNIX clone with all the features of it's predecessor and none of it's overhead. Linus Torvalds and friends did this with Linux. Using absolutely no kernel code from other Unices, the flagship version of UNIX was built outside the "conventional" UNIX family. If Microkernels are as important as some people say they are, it will be done again with HURD. It is this continuous cycle of evolution, punctuated by revolution, that will drive UNIX into ever increasing dominance.
Anonymous Luddite: "What do you think of the dehumanizing effects of the Internet?"
Andy Grove: "Not Much."
Actually, I don't think GNU has ever pretended to be anything but Unix.
"The GNU Project was launched in 1984 to develop a complete Unix-like operating system..." --www.gnu.org
The 'Not Unix' merely refers to GNU being a Unix implementation that is not owned by AT&T, Novell, SCO or whoever owns the Unix brandname, copyright, or whatever, at a given time. A generic, free version (with, perhaps, some improvements) of a commercial product.
Well, some more knowledgeable people than me can comment but I thought that the Internet started on other OS's than Unix, mostly PDP/10 with TOPS-10, but when the hardware became obsolete it moved on Unix because TOPS wasn't cross-platform so the death of the PDP/10 was the death of TOPS-10.
So I don't think that the Internet was created on Unix, unless you mean the modern Internet, excluding what existed with Arpanet.
"The obvious mathematical breakthrough would be development of an easy way to factor large prime numbers." Bill Gates,
Of course it will adapt. It always has before, largely because its design lends itself to ad-hoc adaptation better than any other OS. To get an idea of how well it's adapted, try finding a UNIX V7 or earlier to play with. You'll hardly recognise it. TCP/IP? Hah! Shadow passwords? You make me laugh. GUI? Gooey is what happens when you leave your Mars Bar in the sun too long. Edit those password files by hand with vi. Shadow passwords? Nope. Wait a few years and download the source code for that one from comp.sources.unix.
UNIX adapts well because it's so open and easily modified. Linux is the extreme case of this. Because people can adapt it themselves, they do.
Or, worded another way: It will adapt. Resistance is Futile. You will be assimilated.UNIX servers will become a minortiy in twelve and a half years when Adam Mark's PassArch Kernal 1.3 Server SLC OS becomes the mainstay for everything from Strato-cruisers to watches... including global network servers.
This is an educated guess.
-Effendi
-Effendi
I was just yankin your chain (anthough I think it sounded realistic). Expended version:
Adam Mark's Passive Architecture Kernal v 1.3 Static Liquid Crystal Operating System
I made up the name Adam Mark too.
The point was: Why debate the issue? Venturing to say whether or when it will die or not is the very discussion that will determine whether or when it will die.
How about discussing something useful like what features need to be added or improved in UNIX to extend it's already increadible lifespan. The history of UNIX will never be enough to hold it in the air, only the usefullness of it today in comparison to other products. If there are currently no products available to fully replace UNIX, then we have just nullified the reason and effect of the entire conversation.
-Effendi
-Effendi
If you want to do something new, isn't it easiest to build it on Unix? Maybe you
have to wait for hardware improvements to make it practical
(like X-windows did).
Unix will be around longer than Moore's law.
If, say, security requirements made all the current popular
OS's unsatisfactory, the replacement would probably have POSIX compatiblity layer (as was already said) and might even be built on
some kind of *nix, especially if we can include GNU/Hurd.
Even after Unix dies, there will be a *sh prompt with all those familiar utilities.
Unix / Linux is made for a distributed networked enviroment. It won't die, it's evolving!
After reading the question two thougts came to my mind.
.sig I see floating around, " Those who do not understand it are doomed reinvent it ... badly ", that certinly is true but ask yourself this, does that mean understanding Unix as a bunch of stuff in /etc or as Unix the philosophy.
The first was that Unix will not die, to see that just look at *BSD and the growth it is experincing. Of course linux (linux is after all designed to be Unix like) can be used as an example but then that opens up the whole GNU is not Unix thing. So Unix will not die, but rather closed, propietary Unix systems will die, again FreeBSD is a prime example of how an open Unix system can succeed. To see Unix systems fading just look a SGIs irix, and how they're practically throwing it out the window for linux.
The second thought I had was in order for Unix to die would be for the Unix philosophy to die. The ideas that small is beautiful, programs should do just one thing, and programs should be filters is very important to Unix, and perhaps transends it. Sure, kernels may change, shells may disapear, and X might fade away but as long as there are people that addhere to the Unix philosphy, and there surely will be, then Unix never will die. That thought reminds me of a
15 years ago:
the 80386 processor is intoduced, MSDOS is the most popular PC operating system, the technically savy use bulletin boards and call into information services such as Compuserve at 4800 baud, and if you are lucky you can use this thing called the internet at a corporation or university to send email and read UseNetNews.
Linus will not start to work on Linux for 6 years
With the increase in networked systems, broadband access, storage densities, CPU power and abstracted layers,it may be impossible to guess what computing will be like in 10 years.
While I don't see UN*X as whole dying out, it would not surprise me to see flavors vanish.
More important may be what functionality will survive. With the advent of the TransMeta codemorphing techniques, the survival of Big Iron and VM, and certain aspects of X, UN*X and Windows UI, I can imagine many things:
An UN*X like OS that can act like VM, creating virtual machines that can run other OS's and purely virtual networks.
A UN*X system that has a smart configuration utility, allowing the average user to install it.
A truely portable and totally configurable UI. It could reside on a smartcard, can be used at any compliant unit, and allows secure access to your home and work systems through encryption, yet looks just like your desktop at home.
What ever the futur brings, we will not be guessing it more than about 4 years down the line, if we are lucky.
Of course, this is just my opinion, I could be wrong.
check out http://www.pdos.lcs.mit.edu/exo/ (sorry, but Slashdot isn't liking my html tags for some reason)
It describes an Exokernel based system, which is a break from tradition monolithic and micro kernels. The basic idea is that the real operating system handles only what is absolutely necessary for security, and leaves the rest, such as memory management, up to "libOSes". These libOSes, however, are user level processes, allowing users to customize anything that isn't absolutely required to be handled by the kernel. Unix, under this system, basically becomes just a library running on top of the real OS. It allows for increased performance for many apps because it gives more control to the programmer, should he/she choose to make use of it.
The page is a couple years old, so I'm not sure what's come of this. Still, it makes for an interesting read.
Plasma ?
Parents Against Kuro5hin
I disagree that the world is moving toward single centralized computer resources. I believe that, if anything, the world is moving toward distributed solutions.
Why? The cost of hardware power continues to drop according to Moore's law... and so do peoples' expectations of the power of their machines. What was one year your top-end server will next year be considered a reasonably good desktop unit... and the third year be left in the scrap heap.
I anticipate that any operating system which allows people to grow the power of their system by plugging in more CPUs, more drives, and the like, will gain a true market advantage. Such systems could be incredibly reliable (if one CPU goes down, the system as a whole does not), as fast as your budget allows, and extremely scalable.
In other words... I have seen the future, and it is like Beowulf.
I believe that companies will opt for many loosely-knit groups of tightly-knit Beowulf-like clusters. Each cluster would share data with the others (not necessarily a full synchronization), but will be largely autonomous of the others. This would provide for security: if one cluster were hurt (hacked into, had a denial of service attack, or whatever), the others could continue unscathed.
To bring this discussion back to its topic... I do believe that Unix is an operating system very close to that vision of the future.
Never play leapfrog with a unicorn. Or a juggernaut.
And, with free {speech | beer} software, its mindshare, not marketshare, that counts and Microcrap is tanking big-time. Sure, they aren't going to go away, but their monopoly days are OVER!
So ignore this troll.
Microsoft's VP of Customer Service is Helen Waite. If you are having problems with their products go to Helen Waite.
I remember reading an article about the symbols for nuclear waste dumps - I think it was in "The Futurist" in the late 70's or early 80's (I remember it from school.) Unfortunately, I don't see any old article index on The Futurist web site ( http://www.wfs.org/futurist.htm ) Perhaps this is worth a trip to an actual physical library...
Some versions of UNIX may die but when you say UNIX is gonna die well, there is a lot of different versions of UNIX. I think LINUX might take out some UNIX versions, but stuff like IRIX i don't see dieing anytime soon. LINUX may kill versions like SOLARIS and DIGITAL.
Well, NT is "based" on VMS (VMS -> WNT). Of course, the "Open" in OpenVMS is for POSIX compliant. POSIX has become an important standard to the point that OSes which never claimed to be "Unix-like" had to support it, at least to a degree.
What you are calling "multi-user" is what used to be called a Time Sharing System.
In fact, the earliest papers on Unix refer to it as a Time Sharing System.
Let's examine where the hardware has come:
I have over ten computers at home on my network, running a varety of OSes, including NetBSD, FreeBSD, Linux, various Windows.
I am almost always the sole user of all the machines. Do I need each machine to support multiple users?
No.
The 'many users/one machine' concept is obsolete.
It's one of the good things about BeOS that it gets a lot of the Unix goodness but shucks off the multi-user/group kludgework.
I like the way Neal Stephenson said it: "Unix, by contrast, is not so much a product as it is a painstakingly compiled oral history of the hacker subculture. It is our Gilgamesh epic." (from "In the Beginning was the Command Line", available on http://www.cryptonomicon.com/)
You have a better chance of killing Malaria than killing Unix.
The whole idea is that the native operating system is going to be the JVM. You don't put the JVM on the native OS. It can BE the native OS.
JVMs have:
- memory management
- thread handling (inc. semaphores)
- file handling
- graphics interfaces
- printing
- message handling
i.e. its an OS.
As to your argument that problems can occur in C/C++ code. I concur. That's why its important to write as much of the system as possible in Java.
Ultimately more code, quicker is always going to win. C++ is terribly UNproductive. C is better but less powerful. Java is more powerful than C and more productive.
-WolfWithoutAClause
"Gravity is only a theory, not a fact!"Unix is stronger than ever now. You could have had doubts about Unix's future in, say, the late eighties/early nineties. You could have pondered Unix's demise when NT was released (before people found out it sucked). But doubting Unix's power now is just silly. SysV Solaris (post SunOS) has only gotten stronger in the last ten years. Linux and the *BSDs are causing Unix to be more widely used.
No one is losing any marketshare. Unix-like OSes are gaining marketshare. There is a generation of IT managers who bet on NT in the early/mid nineties, and many of them are now switching to Linux, because it works better than NT, it's cheap as hell, it's open source, and they get to think they're part of the "new wave" of technologists.
Digital Unix (Tru64), HP-UX, and AIX haven't enjoyed the same publicity boom, but they all have a niche market, and AFAIK, Compaq, HP, and IBM aren't losing any money over them.
Unix isn't dying. It's entering a new Renaissance.
I suppose I'm a "conservative" hacker because I don't buy into the Redmond attitude that we need OS paradigm shifts every five years. Unix has been around for thirty years because it works. Saying that Unix is dying is like saying that cars with four wheels are just a fad.
I'm fully expecting that when IT managers are looking to do their next round of upgrades, Linux will be picked more often than W2K. IT people are tired of Redmond. Tired of the security problems of a bloated, closed-source OS, tired of service packs, tired of having to run around rebooting servers that forget to malloc their gig of RAM. W2K will be just another round of expensive, kludgely fixes and poor performance.
If Unix was dying, would Apple have chosen to integrate BSD elements into OSX? (Okay, maybe that was a bad example. Apple has never been stellar with their business plan. ;-)
This is going to be quite a year for Linux. 1999 was the year that Linux got ready to take on the world, and 2000 is the year that it will.
Just my two yen.
Too hot for CPAN!! Get PerlOS now from
I think we should better say that UNIX has never lived as a desktop OS. Anyway, to me the next OS for desktops won't be UNIX/KDE neither Windows 2000. I would bet better for Palm OS or for whatever operating system runs the Playstation 2.
Is Latin dead? In the sense that few are speaking it, yes. But if you spend any time in the Catholic church, read any medical journals or legal briefs, it would seem that it is very much alive. Can I drive a 1937 Chevrolet on the highway to any destination I choose? Obviously yes, but some would argue that the design of the 1937 Chev died a long time ago. Unix is a tool, like Latin, or a car. As long as there is an application for it's use, people still using it, and a context inwhich it is still useable, then I would argue it is not dead. But then there is a lot of technology that isn't really dead, isn't there?
Unix still has some steam left. Nothing will be able to touch it for at least five years (forever, in Internet time!).
That said, Unix has some major holes. It is definately not the be-all-end-all of OSs. What will replace it? Definately nothing from Microsoft. A replacement will have to be written from the ground up.
Off the top of my head:
1. Security. The access-list based security model is fatally flawed. In order to go to a capability- based security model, the kernel will have to be completely rewritten. Many apps will, too.
2. Multiprocessor handling. The macrokernel model used for Unix sorta- kinda works with multuple processors. A new design would be designed from the ground up for both shared- memory and mesage- passing multiprocessor machines.
3. Configuration management. The current crop of "package managers" is merely making the best of a near- impossible situation. Ideally, you'd just copy a "package" file to your disk and tell the system where to find it and who you want to be able to run it. None of this spewing files all over the system.
4. Dynamic reconfiguration. We need a system that will detect its hardware and run the appropriate code. (Plug & Play done right, in other words.) Current Unix/Linux systems have trouble with floppy disks and CDROMS, let alone esoterica like sound cards.
5. Real-time. I get the impression that the Powers the Be don't know what "real time" means and don't care. We need this for multimedia and general communications processing. Current doctrine is to spec hardware that is *real fast* and then hope for the best.
So who's going to write it? I dunno. Maybe it will be open source. Maybe it'll be something inside a sealed box from Sony. Maybe it'll be a closed-source project from NSA, required for anything touching the US Government.
It'll be interedting to watch.
Welcome to the Turing Tarpit, where everything is possible but nothing interesting is easy.
No to put too fine a point on it, I think I won't care much if somebody gets my "key" after they decapitate me. Sheesh.
--
Do daemons dream of electric sleep()?
I don't know what inspired the posting of this whole subject, but I think that UNIX will last for quite a while. I am a "Windoze Convert" - I was strictly a Windows user, until I found out about Linux. I am convinced that Linux will not die out anytime soon. The question we should be asking is, "When will Windoze be replaced by UNIX?" Argue as you might, but Linux is growing rapidly, and those users don't just appear. To sum it all up, I think that UNIX will be around for a lot longer than this whole Micro$oft Monopoly.
Unix is an operating system. Its purpose is to operate the hardware inside your computer, and to provide a programmatic and generalized interface to that hardwares' capabilities. As long as Unix continues to operate popular hardware and provide an interface that programmers like, Unix cannot die. I always wondered what 'die' is supposed to mean for software.. on old enough PC machines, there is still MS-DOS that runs just fine, doing what the user wants it to do, and it still works... get a C compiler, you can make it do even more things if you care to.. so how can any software that has a use of any sort 'die'? I wonder if software can get so old that no hardware that runs it exists and absolutely no one runs it, and never will... I guess unix, or msdos or whatever can somehow possibly have no future.. ms is killing msdos functionality in the next version of windows, but that doesn't even kill that old system, so why even doubt that something dynamic and important to the future of the past of all computing like unix is gonna just go away and not be improved or even used as is? The idea of software dying is silly. Support for a certain type of software dying is more likely, even inevitable, but when does an entire system 'die?'
@>-,-`-- Kandy Neko
If we aren't carefull, Unix will outlive humanity! It is just a matter of human stupidity! CU in Unix Heaven Shadowrom
More reliable? I don't know what logic you're using for that, but umm... no. just no.
Lets' see, the JVM gets compiled by a C/C++ compiler most likely, then runs on a native operating system. So you've effectively added a layer of complexity to the system where problems can occur compared to C/C++ on said native OS.
As for Java's usefulness, it's not a bad language, but let's be honest with our evangelism.
----------------------------
It is ALSO a truism that you don't get something for nothing. There's always a trade-off. With networked computers, your trade-off is in delays: delays within the network, transmitting the information, but also delays within the computer, which now has a real-time device to constantly monitor for potential traffic.
It follows that setting up a network of streamlined machines, each dedicated to a narrow range of tasks, will out-perform a single, monolithic system, IF AND ONLY IF the savings from the parallelising is greater than the penalty imposed by the network.
The future of Operating Systems, as I see it, will consist of intelligent distribution of tasks over a wide-area network, and where any program is parallelized at run-time by the primary OS. However, I also predict that 99.9% of all software will be kept local and run serially.
It would seem to follow that Unix will be extended to support run-time parallelization and fairly sophisticated task/net load-balancing.
It would also seem to follow that there will be a blurring between =PROGRAMS= and the network, NOT the machines and the network, OR the OS and the network. Why should there be? If it's not visible to the user, it won't last.
It's a small world and it smells funny; I'd buy another if it wasn't for the money; Take back what I paid (SoM)
I have often heard 'Unix is a Legacy system, old technology from the 60's and 70's.'
This is patently false. What Unix _is_, is a system of tools which have been refined for 30+ years. AFAIK, 'Legacy' should be used to describe something which is no longer developed or supported. At least this is the general use of the word.
New doesn't mean better, and in many cases, it can mean untested or unproven. Unix is proven technology. But above all it's a proven philosophy.
Call Unix whatever you like. Text-book Sys V, BSD, or clones like Linux. The underlying philosophy is largely the same. THIS is why Unix has been around for decades (besides techinal merrits).
In the philosophical sense, I doubt Unix is going to die, ever.
I've been using Linux now for almost 3 years, learning first that I knew nothing about Unix, and digging through lots and lots of documentation - books, LDP, man pages, et al. I consumed the Linux penguin/fox one bloody command at a time. I'm a mathematician who can program.
Now, between Awk and C, I can do virtually anything. I am absolutely impressed with Unix, Linux, the GNU tools, et al. Unix is the ultimate super calculator. It will do whatever I want it to do. Once I ran into a memory limitation with C, a limitation I was able to fairly quickly circumvent.
Unix and the GNU tools are the computer equivalent to being mathematically elegant.
Will Unix go away? Who knows? Who cares? I hope it stays, but if something better comes along, I'll likely follow.
At present I think Linux has revitalized Unix, and I believe a new generation of Unix Geeks will come out of this current revolution. There is and will always be a need for a superior computing toolset. Today that toolset is Unix. Tomorrow it may be something else. Whatever that something else is, however, won't be Microsoft Windows!
Graham
Graham
Linux - Fast Pane Relief
There is no such thing as "Iron-clad" validation, but validation tied to something physical (retinal scans?) will come pretty close to this in less than 10 years, IMHO.
Everyone always seems to think that retinal scans are a decent method of validation.
However, all they will do is provide a key. Furthermore, it's much more secure to remember your key than to write it down somewhere (think: decapitation). Until we have mind-reading technology, I'll stick with passwords, thanks.
Hamish
"Wise men talk because they have something to say; fools, because they have to say something" - Plato
Unlike many of the alternatives to UNIX mentioned by other posters, Vapour is an OS design that totally breaks with the UNIX concept - it has no kernel, no processes, no filesystem. Memory protection is implemented at the language level rather than the OS level. Persistent virtual memory (similar in some respects to EROS' design) replaces the traditional filesystem. All parts of the system may be modified at runtime. It's still very early in development, but the design is fairly comprehensive at this stage.
While I happen to think that all the Unices are probably superior platforms for this type of application, don't forget about some of IBM's offerings: AS/400, OS/390, and so forth.
Big Blue is active in the development of Apache, and I would be surprised if they weren't sneaking parts of Apache to their other OS's, which have anemic TCP/IP suites to say the least.
Just a thought.
Well "we" may have fought the powers that be, but on the whole "we" (whoever that is) don't matter much. For large coporations, and ISPs and for those trying to provide services of one kind or another (ASPs, if you like the jargon), centralisation is the only way to go.
The more stuff you have on the user's machine the more your support costs, as it breaks, needs to be replaced, etc, etc. When you look inside large companies you see a greater and greater tendency towards web-based and other centralised systems.
Simon
I don't think UNIX will die anytime soon!
The reason is simple: both the commercial UNIX variants and the "freeware" FreeBSD and Linux variants have extended the life of UNIX far beyond it was possible in the past.
The UNIX of 2000 can do multiprocessing, multithreading, SMP support, powerful networking, and even massively parallel processing via Beowulf clusters. And engineers are already pushing UNIX so it supports 64-bit processors like the Compaq (neé Digital) Alpha and the upcoming Intel Itanium CPU's.
In short, I just don't see any replacements for UNIX in the near future. The technology in Windows 2000 may be a major competitor, but it won't be a UNIX replacement.
Raymond in Mountain View, CA
Given:
/usr/bin and ls lists files. People make things similar to what they like. Witness Linux. Linus liked the UNIX he used, so he cloned it. Even BeOS (AFAIK) has a UNIX-ish filesystem hierarchy.
1 - Engineers / Computer People need a command line interface
2 - Regardless of the virtual reality smell-o-vision interface, there will be a command line at the kernel of a real (READ: non-consumer) OS.
3 - The idea of a filesystem hierarchy isn't going anywhere.
We should be able to plop down at any box and know that binaries are in
Um... Novell, SCO, Digital... ?
Never mind the scores of 'unix-like' embedded RTOSes?
And I know at least one vendor (my former employer, VenturCom (at vci.com)) had an embeddable version of Novell's UnixWare.
Unix and Unix-influenced OSes have always been out there in every market. The Unix-dieback was a dieback of the workstation and a little bit in the lightweight server area.
--Parity
--Parity
'Card carrying' member of the EFF.
I've looked through the comments, and seen a couple of what I regard as misconceptions:
...
1) that UNIX will be killed of by Windows or BeOS or
I think that if something is to replace UNIX, it will have to be something _significantly_ better than UNIX. I don't know of anything else currently out there that can claim this (although I will admit to not having looked at Plan9 and Hurd).
2) that UNIX is immortal because it is somehow 'above' operating system paridyms (sp)?
This is clearly wrong, is the whole UNIX model is based on a hierachy of files, where a file is a stream of bytes/characters. As a consequence of this, a most of the tools that make UNIX so powerful are text stream based.
I see room for a new sort of operating system that works with a much higer level of abstraction throughout, including type management and garbage collection as part of the OS (relegating streams of text to where you really just want to store text information), high level messages between components allowing easy distribution, different views of the file system (or maybe not even so strongly file-system based), and maybe taking some component ideas from Apple's failed OpenDoc to end monolithic applications.
However, I don't think that there is anything out there that comes even close to this, and there won't be for a while, and even if there is, it will take years to catch on.
UNIX will only be overtaken when the problems of the high level of complexity required by what people want to do need a higher level of abstraction than the basic model of UNIX can provide, and even then, UNIX will take a very long time to actually die - just go the same way as DOS that is still widely used but generally regarded as inferior by most people.
This sounds like the Condor system at the University of Wisconsin, Madison. I have only read a little about it but as I understand, it works as you describe. Check out their homepage at http://www.cs.wisc.edu/condor.
-- Remember: Wherever you go, there you are!
True. Nut unfortunately, nobody takes reports of the Amiga's rebirth seriously anymore either.
Two more weeks! BIG!
---
As copyright owner of this comment, I authorize everyone to defeat any technological measure which limits access to it.
I can guess based on what I've seen on the net about Apple's OS X, but it appears to be an evolution of UNIX (BSD 4, right?) :)
(I will state at this point that I'm an HTML and graphics guy; I am not a programmer in any sense of the term, and have desire to be one in the sense of writing big applications and pouring over code for bugs. I will leave that to the experts!
However, I see Apple's OS X as a *possible* evolution of what UNIX can be: it's got the friendly GUI on top and the creamy fudge of BSD on the bottom for those who want to delve into it. From what I hear, you can code into the BSD and it works well. I could be wrong.
In the same vein, with Linux, UNIX ain't going anywhere! More and more people everyday are installing and using and learning it, and there will be more in the future. The open source concept will live on, and I don't think UNIX is going anywhere: it does what it does to well to be replaces with anything else!
Pope
It doesn't mean much now, it's built for the future.
The death of unix is a false prediction that's been around forever. With the birth of linux & *BSD I would say that it's experiencing the largest surge in popularity for a good 10 years.
I see unix adapting in a few needed ways (and I think it will be lead by open source software):
1. Adaptation of the security model; file system ACL's are something that all IT personel appreciate. In todays enterprise it has become almost necessary for file server environments and is venturing way beyond that. As I tell people @ work NT has a pretty good security MODEL, but it's not the best implementation. I'm not saying stick NT's model into UNIX, but do like unix and OS software have always done....improve upon current implementations and develop new ones. This covers not just ACL's but many other things.
2. Fragmentation....Always a hot topic! Linux is starting to see fragmentation whether we like it or not and this is where I see companies like RedHat as good things. Take KDE and GNOME for example they are both wonderful desktops and the developers are working towards common standards for drag and drop, etc. This is the kind of thing I like to see, but unfortunately the effort hasn't produced obtainable results...yet. This is where a standards body would help us all out considerably because in the end it means more apps and functionality for everyone in the game.
This is just a short version of my $0.02....
-Aaron Dokey
Gainful Employee of Technology
----------------- Who is Jesus?
There are many things that could be done to improve the speed, scalability, security and reliability of UNIX systems (including Linux and *BSD). Some of these might require substantial kernel rewrites but they don't require a brand new operating system. Sun pulled this off when they switched from SunOS to Solaris.
I don't think Unix is an operating system. It is a family of operating systems.
;)
.sig, "He who don't get UNIX will make one that looks like it, only worse." I know I am way off but there may be some truth to the notion that there is something very pure and natural about UNIX, and elegant OS.
Look at GNU/Linux. It isn't really Unix, but is really close. Isn't it some kind legal problems with naming? And besides, GNU and Linux were actually two separate operating systems in development that eventually merged, neither really Unix but now we are all happy.
Now Apple is putting Unix inside of Mac OS X, which I would say the opposite: Apple is putting MacOS inside of Unix. Apple is known as an innovator in the software industry, who have seen far. But they have been standing upon the shoulder of a giant named UNIX.
And if you stare at DOS hard enough, it starts looking like UNIX (anyone know "TYPE FILE.TXT | MORE" ?) but I won't go there
My point is that you won't be able to look at a new OS and say, hey! that's not Unix (unless you are explicit like GNU). You will just kind of see it and think, hmmm, it's different but it's cool and UNIX.
Who was the guy with the quote from Henry Spencer as his
I don't want to start a flamewar, but what exactly is supposed to replace huge reliable Unix systems to run corporate databases, web servers etc.?!
While I'm unsure about Unix' (Linux') future on the desktop, I'm very confident about the professional part of the computer world.
Unix will not 'die'; neither will DOS, or MacOS, or Windows --any technology that it's innovative and gains enough mindshare lives on in other technologies. As long as a philosophy, concept of feature is liked enough, people will re-implement it under different systems... Do you forget that Linux was a Unix re-write?
Sadly, it seems that things aren't evolving as fast as they could. Linux *is* a Unix rewrite, and it's still not that much better than its foundation. NeXTStep re-implemented Unix a decade ago and it did a much better job --so much so, that its successor Mac OS X has still a big technological edge...
I wonder if the Linux community can look ahead far enough to stop worrying about backwards compatibility with older Unices and start innovating in the fundamentals of the OS. Security, administration, configuration, maintenance, documentation and quality of service are hopelessly krufty and kludgy in most Unices and Linux too.
There was a point that Linux needed to emulate its siblings, to remain relevant and useful. But in the next few years it's quite possible that Linux will become the dominant Unix-clone. Compatibility and tail-light chasing be damned --we need to innovate.
engineers never lie; we just approximate the truth.
The BSD family are firmly embedded in many vendors' networking infrastructure - both packet and TDM (telephony). Once it's there it will be there for some time. The penetration is increasing, as new entrants in these markets use FreeBSD, OpenBSD, and the like as convenient, stable, open platforms for their networking products. (It's particularly suited for packet routing, since BSDs are where the software was developed in the first place, and the BSD interfaces now serve as the default interchange language for exchanging software sources.)
Mainframes running UTS (mainframe-compatable clones of SVR2 and SVR4) now handle mission-critical functions for many large companies: All the Baby Bells, for instance, do their long distance billing data capture on it, and run their where-are-all-the-wires databases on it. (If it ever went down all the long distance calls would be free until it was back, which is why uptimes in years are mandatory.) Brokerages support their trading with it (even more $/second if it ever went away). Web sites run on it. (Apache has been there for a while.) And so on. And of course they fixed the Unix clock-rollover bug long ago, so they shouldn't have as many hiccups a few decades down the road when it finally rolls.
Semiconductor design is done with tools that run on Unixes. Some have been ported to Windows to try to take advantage of the cheaper crunch - but not many, and there's little demand for them, since they can't be easily combined into a design flow with scripts. Some of them are now being ported to Linux to achieve the same cost savings. This is easy. (For many, it's just copying the source tree and running "make", for some it's a little tweaking.) And on Linux you DO have the scripting tools, plug-and-play with Unix networks, and a familiar environment. So this IS being accepted - nay, demanded - by major ASIC design ooperations.
Billion-dollar companies in trillion-dollar industries are depending on hundreds of large applications written to run on unix. If they ever DO port them to something else, any bets on whether it will be something new, or another flavor of Unix?
(And right now Linux qualifies as a flavor of Unix for this discussion. Windows, NT, and OS2, of course, do not. What a pity.)
Bantam Dominique roosters crow a four-note song. Once you've heard it as "Happy BIRTHday" you can't NOT hear it that way
Supposeably that's in 5, which will be coming out before the month's end.
Saw it written and I saw it say, pink moon is on its way. None of you will stand so tall, pink moon is gonna get ye al
Since it's so much more versatile than an all-in-one GUI approach (like win9x but not really like the original NT architecture) it's unlikely to die out as a concept even if every single currently existing UNIX maker goes out of business.
Eric
Want to work at Transmeta? Hedgefund.net? Priceline?
Can your IM do this?
I use Linux these days. It is hardly the same Unix that I learned on in the mid-80's. And that wasn't the same Unix that was developed in the 70's. And it doesn't matter. I will switch OS's again. I don't know when. But as long as there are enough people who want what the Unix view of the world provides, irrespective of the name of the particular implementation, it will continue to exist.
I can run the same scripts under Linux, AIX or FreeBSD for the most part with very little portability problems. For me, Unix is a set of tools and a lot of leverage. It is the idea that I should be able to carry my tools and data with me until they are no longer necessary rather than making the programs that process them artificially obsolete.
The net will not be what we demand, but what we make it. Build it well.
Sorry.. I only got the lightweight corporate throwaways back in those days (89-91). Didn't see too much Novell or Digital, but SCO should have been up there.. Perhaps BSDi.. Naw, not BSDi.
Embedded UnixWare?!? What time frame?? I'd didn't hear of such a beastie until 94-95! Granted, that was when I was trying the embedd microcontroller thing, so I probably was just out of touch.
.sig: Now legally binding!
Microsoft invested enormous amounts of capital in self-configuration for Windows. As a result, despite complaints to the contrary, Windows _is_ less costly/risky to install/reconfigure with novel hardware. Aside from the encyclopedic knowledge base required, the logic behind this sort of autoconfiguration is fairly impressive. It approaches some of the best induction algorithms ever fielded. But it still is a pain to install new hardware/software components.
What this means is that the OS of the future is going to have to focus a lot more on dynamic autoreconfiguration with relatively sophisticated "truth maintanence" induction algorithms that draw on, and are sensitive to changes in an enormous rule base supplied via the net by the vendors of hardware and software components. Further, OS vendors are probably going to have some sort of minimal "boot up" network configuration, similar to the dumb video modes used to get your better video drivers installed.
The next stage beyond that is a very high level interface specification language that can describe the hardware and software to the OS which will dynamically re/generate the driver/libraries for its particular configuration (with appropriate roll-back safeguards). Such systems will eventually even have Hotspot-like dynamic optimizations built in so they can generate encached code on the fly to the spec of the high level rules based on the patterns of usage and other dynamic information. Much of the nasty code that goes into autoconfiguration and driver installation will be annealed via dynamic compilation and inference and eventually hardened and optimized -- always sensitive to changes in the high level rules and descriptions.
Seastead this.
There have been various attempts at distributed operating systems in the past; some, such as amoeba and plan 9 are actually usable, to a certain point. Unix is here to stay--it's not going to disappear until someone maxes the next "big step" in computing, and maybe not even then. There are a number of projects out there that are quite interesting. On the one hand, you have OSes like QNX which were designed to be entirely distributable (for lack of a better word) from the ground up. One of K&R (I can't remember which one) once said that one of the places where unix failed to take the "everything is a file" concept to its logical conclusion was networking. (In plan9, pretty much everything, including networked stuff is a file). I don't think that distributed OSes will kill unix, but that unix will eventually become a distributed OS. For example, GNU/HURD (which is getting along very nicely BTW), while not an attempt at a distributed OS, is designed in such a way that it will be easy to transform into a distributed system.
People now believe that birds are the descendants of dinosaurs. Stranger things could happen
;^)
I remember when Unix was a terminal-based OS. X-Windows and the Athena project seemed like a totally new world and a new way of doing things. Sure, it was still Unix, but it wasn't the same Unix.
Although I am sure that someone did something with that new fangled thing called curses. With curses and the like you can create a windowing system on a unix environment quite easily. This was a natural outgrowth of programming.
GNU has likewise changed what was Unix, and, despite it's acronymic denial, has become Unix. But not the Unix from before.
Other than costing a lot and having some little quirks that are anonying to people who are using linux now how is there a terrible difference?
The next Unix will not be today's Unix. But it will be Unix!
The unix that I use at home for the most basic things probably has not changed terribly from what a person in earlier times thought of. Although I do rely on various graphical input methods I could say take the base install for debian and have it pass as unix. True there are some extra bells and whistles but essentially they remain the same.
Slashdot social engineering at it's finest
If you mean distributed computing or spreading a task among multiple computers that are linked together, you probably mean Beowulf.
No I mean having say an application that could request data from a server and do some little thing. Now take that little chore and replicate thousands of clients that could transparently work on all of the machines on a network. Everyone from the secretary to the CEO of the company could have one of these little things on their desktop. Now say I want to figure out something really, very complex. How about how many times people have complained about product XYZ and how that correlated to the stock prices over the last 50 years. Now that data could be done on some mainframe with a high rate of failure or requiring special attention. However if you distribute the task to say 30,000 clients to work on in their spare time I would dare say that an answer could be easily found within the hour. All the main server or set of upper level area servers would have to do is just run the client solve their portion of the problem and then return the result back to the server they are responsible for and just correlate the final information.
Slashdot social engineering at it's finest
I find this idea laughable, we made this transiton long ago.. look at how many people use their PC solely to surf and as a really heavy typewriter.
What you will see in the future is more things like Napster, Quake 3, etc. Things which take advantage of the network but do so by using the power of the PC and not relying on something like a web server so much. I can definitely see the web shriking in the near future. Imagine how fast, efficient, manageable, customizable, etc, etc, eBay could become if it was a simple client application with the logo graphics cached, connecting to a distributed server farm, keeping track and tracking your auctions pulling data straight from the eBay databases, etc? Think of how efficient Slashdot could be as a distributed client application, relying on the PC to do a lot of the computation like sorting, getting the slashbox data, etc, etc...
Esperandi
And hopefully it will be fueled by adware. Programers get paid, users get to use for free, and the babies who don't realize that advertisement is subsidizing a life they wouldn't be willing to pay for get a fe more sharp kicks to the crotch.
(The scene - two cutting edge technologists talking about their favoured operating systems circa 2000 years BCE)
Gilgamesh (Bronze caster from the city state of Ur) : I tell you Nergal, this technology you're using, it's going to die out and nobody will use it any more ...
Nergal (Basket maker from the city state of Akkad): Ahh, away with you, everybody uses this technique of making baskets. Besides all the manuals are written in Akkadian, people will still be using Akkadian 5000 years from now!
Hahaha only joking, but hey....
Actually I heard the US military were looking at devising language / symbols to put on their high level nuclear waste bunkers, so when humans or whatever stumble across high level dumps 10,000 years from now they'll know what's inside is dangerous and should be treated with caution. When you consider that the history of human writing only goes back about 5000 years this is some task. Can anybody help me track down some of the literature/ urls/ research about this project? Many thanks.
This touches on a point that is relevant to most people here: that's why GNU is a unix-like system. At first RMS, what with his love for all things lisp, had thought about making the free OS he was planning a really big lisp environment. But he realized that in order to be a general-purpose system, it would still need to be built on top of a general-purpose OS; so he chose unix. And that's why we have our wonderful unix-like system with Emacs (= a really big lisp environment) running on top of it.
The best way, imho, to make your wonderful ubercool environment it to build it on top of (a subset of) unix. That way you can let unix take care of the mundane things (device drivers and whatnot) that you don't want to, and be quite portable.
If the move to network/distributed computers happens (which I really can't see at the moment, being in a contry where you still pay for net connections by the minute), it's an advantage, not a threat to Unix systems.
Microsoft still controls a lot of desktop machines, but the networking code in Windows 98 is so broken that people might consider upgrading to Linux or *BSD if they were doing more networking.
Embedded devices, another part of this move, are another big chance for Unix-like systems (primarily Linux and PicoBSD) - I think Linux is in use in more embedded devices than Windows CE already.
This message is provided under the terms outlined at http://www.bero.org/terms.html
Erik Dalén
Unix is not just an operating system. If it was just a kernel then maybe it could die. Or if it was just made by one company then it would die.
/home dir. I want a graphical user interface. I don't want to type anything anymore" But still inside the wrappings it will be Unix.
Really anyone can make a Unix with in a year.
Unix is a set of command line applications.
Is everyone going to suddenly realise that grep and wc were kind of silly and should be gotten rid of?
Unix is organised filesystems.
People are going to decide that placing libraries at random is better?
Unix is standards.
More than just the posix standard there are tons of standards. Even if they change they'll still be Unix a heart.
Unix is a philosophy.
A pretty good one. Pipes are good. Shell scripts are good. And small programs are less buggy.
Maybe someday people will say, "I only want to deal with the files in my
The beauty of Unix and its deriviatives(ie Linux) is their high level of customizability. If you have a device dedicated to a specific application, and you want a dedicated OS, you could write your own, or you could start with unix/linux kernel source and customize from there. That is in fact what many embedded device makers do. After all, who would want to run life support equipement on windows :)
It's been done: Plan 9
Unix is any OS that I can bring to a screeching halt by typing kill -9 1 at a root prompt.
Is NetBSD Unix? Is Linux? Solaris? AIX? Minix? GNU?
Sure, an OS can be "certified" Unix or certified "POSIX compliant", but that isn't an end in itself. Unix (however you define it) has eveolved through the years, as everyone has already pointed out, but is also modular (I can replace proprietary ls with GNU ls), and portable.
Where will Unix be in 10 years? I don't know. But I know where it won't be: lost & forgotten.
Well I mean think about it.. how many of us have more than one PC in our home? How many of us would rather have a single PC with a handful of "front-end" appliances scattered in convenient places?
Centralization of data is what we're moving to. This is why things like Hotmail are so successful -- easy access to e-mail regardless of where you're connected. I just bought a couple of i-openers (see the Slashdot article a few days ago) that I'm converting to cheap X terminals.
Some attempts to do this on a large scale (WinTerms and "Thin" clients) haven't really caught on, mainly because of the still high cost of the clients and the fact that it's not cost effective to switch. But who's to say that won't change?
Which leaves us with the question: what type of platform do we want supporting the network devices we connect? I don't think I've seen NT or any "recent" OS compete favorably on these grounds, which leaves us with OSes like Unix, still being developed as fast as the technology does.
Of course, that's not to say that other operating systems won't possibly step up and smack Unix out of the arena entirely, but we can't bet on that..
--
Do daemons dream of electric sleep()?
to ask "will unix come to an end" is meaningless. let me explain. if we look at unix in the context of normal operating system space and time, we notice singularities at the "beginning" and at the "end." i have been working on this problem for many years now and am now ready (coincidentally) to offer you my solution to this age-old problem of computer science!
to better describe my concept of unix, i have developed a new mathematics which defines unix in terms of "super-time." our beloved unix is embedded in this supertime, which contains no singularities! thus, unix itself is without bound in time and space... it simply always is.
thank you.
What do you mean by Unix? I can't really think of an all encompassing definiton. I'm not sure about everything here, but in my limited knowledge of Unix heritage and composition, it occured that a lot of things qualify in one way or another as Unix. Posix compliant? Nope - NT is Posix. Based on original Bell Labs code? Not even Linux qualifies. Containing Unix code? Fine, but don't forget to throw in MacOS X and BeOS as well. Command line based? No way - DOS et al aren't Unix. Unix grew from being an operating system at Bell Labs as in "Unix, the descendant of Multics" to standing for a whole slew of operating systems. Much like our DNA is different from primates by only about 3%, there are only a handful of things that separate a Windows machine from a Unix box. For the most part, they're a lot alike. Both have the same fundamental architecture - kernel, processes, etc. To me Unix means a really stable, secure multiuser OS that is remotely manageable and based on Open Standards. But I'm just pulling that out of my ass. It means different things to different people. Novell meets that description too. NT, in some ways, does too. So I guess my answer would be that everything qualifies, in one form or another, as being part Unix. Can anyone more clearly define what, exactly, Unix is?
--
I think there is a world market for maybe five personal web logs.
>NT evolves more
> toward Unix every day.
In the world I live in, it's years between NT Releases; months between service packs.
Not that I wouldn't love to see this daily evolution of course...
-fb Everything not expressly forbidden is now mandatory.
It's like when Apple was having financial troubles: "Don't buy Apple, what if they go out of business?" Who cares if they have $2 billion in cash reserves? Media sensationalism.
And Amiga. Amiga has died so many times, nobody takes the report of Amiga's death seriously anymore. Perhaps nobody defined dying: new products are becoming available, a new AmigaOS came out 5 months ago, and so on. So what is "dead"?
Now slashdot: anyone who has spent any time in the industry knows that Unix is the most dominant force holding the whole computing world together, so why pretend to take a question like this seriously? Media sensationalism. That's all.
Until I see a headline like, "Unix is dead!", followed by "a young man named Unix was gunned down", I'll stick to reality.
Now: a more apt question is: Is Windows dying? I have some compelling ideas about THAT...
Sounds a lot like any university's internal LAN. Which probably runs Unix (Solaris at my university).
Unix was already designed to address many of the issues that come up with network computing. I only see a few things that need to be fine-tuned:
Portability of user accounts across the entire network, checking of permissions/licenses for running applications, etc. This is already pretty much here; you just have to know what you're doing to set it up. The challenge is to make sure that all applications on your system work fine for all users, everyone can do what they're supposed to and have access to what they need to, but to make sure that nobody can do anything they shouldn't be able to.
If you are a validated user at one university, you would ideally be able to log into another university with guest privileges, and have it recognize you as a specific user ("foo at bar university"). Similarly, I'd like a validated user on my personal LAN to be able to access someone else's service while keeping an individual identity. Or through another network access their "home" network's services with their full privileges. The idea being that identities and permissions carry over robustly and securely over heterogeneous and possibly untrusted networks.
This ties into the whole "the network is the computer" idea. If I could just use the collective computing power of all of my hypothetical LAN terminals to distribute tasks, I might not need a central server at all (assuming that my tasks have low communications overhead). Similarly, it would be nice to be able to farm off tasks to another "friendly" network. Protocols and support for this is in development, but would need to be standardized for true "network computing" to come into its own.
Again - by and large, these are capabilities that already exist, or exist at least in part. They continue to be developed - and chances are, that development is happening under Unix.
Programs I wrote for UNIX 10 years ago still work just fine and take full advantage of the larger memory and faster processors. That is not at all true of systems like Windows.
UNIX is also a particular approach to writing kernels, based around a monolithic, fairly simple privileged core. UNIX kernels are also pretty closely tied to the C language. That's very different from microkernels, Windows, or other systems. And UNIX is a set of conventions for where and how to represent system configuration information, command line programs, etc.
That kind of continuity can't last forever, and eventually people will start using systems that can't realistically be called "UNIX" anymore.
Sooner or later, kernels will have to be written differently and in languages other than C, the file system will change into some different kind of database, etc. People will also build new replacements for system configuration information, the init/login sequence, etc.
I suspect, however, that the transition from UNIX to its successors will be fairly gradual, and that any successors will continue to offer reasonable POSIX implementations for a long time to come.
Plan 9 and Mach are both examples of successors to UNIX that really have some good continuity with it, and that's probably what's in store both commercially and in the open source community.
Since a unix core install can be made very small,. suitable for embedded systems, I don't see a reason to throw away a perfectly good model with a well-understood API and many thousands of man-years of refinement.
Unix is being used successfully on systems from mainframes to wearables, from super graphics boxes to the tiniest pinhead sized embedded systems. What design criteria do you envision that would contraindicate the use of Unix?
Montgomery's short "An Introduction to Unix" points at the Unix system family tree.
That 1997 document does not mention Linux, which grew out of the POSIX definition, System V, NetBSD, and GNU tools (developed on many Unix flavors). The Unix History segment of the Unix FAQ does mention Linux briefly.
Well, I was a bit surprised by the 5, but it wasn't meant to be a troll either. Damn it, I want Windows to live up to the hype. I'm in a huge NT shop at a large college where desktop security is important. Most of the pieces to make my life easier are there, but I see no light at the end of the tunnel.
Take applications for Windows. Damn they suck. Not the apps, but the design. Microsoft can readily fix this if they got their heads out of their asses and realized that the world isn't about one person/one computer with full control.
How can they fix it? By taking their already existing label standards for apps and strenghthing them so new apps at least behave properly. Don't follow the rules, then your app is not "compatible with Windows 2000." But then that would break all of THEIR apps too...
Example:
All of my bitching about NT/2000 comes from actual experience trying to make it work as advertised. When Microsoft's own fucking applications don't follow good design standards, how do they expect others to do so.
Do you realize how long I took to get that piece of shit IEAK to work properly? First of all, the IEAK book spends about 250 pages talking about how wonderful it is and all you can do with it, and then when it gets to the point where it talks about how to implement policies, it spends exactly 1.5 pages on what all 200 settings do. During an unattended install, it throws shit loads of stuff into runonce keys but, heaven help you if, before the next reboot you or another program invokes rundll32, because rundll32 triggers all runonce keys to be processed immediately, even if they were installed during THIS boot instance. That one killed me for a long time. Then, during a user logon, you have to ensure loadwc.exe can run and *IT* uses rundll32 to do a lot of the customizations with the policy settings installed. But, get this, rundll32 won't run if it can't for some unknown reason have write access to the runonce key. But allowing users r/w access to that key violates a KB article saying what a huge security risk that key is for users to have r/w access to it. So, I have to give admin rights to users who I want to policy control their computer?! Oh yeah, that makes a lot of sense... :( And none of this is documented in the IEAK docs. No, just hundreds of pages of marketing fluff. I'm a sys admin. I buy an administration kit to get technical details, not marketing hype.
None of this actual implementation stuff ever gets mentioned in the press. All the grand claims of how Microsoft makes administration tasks easier are taken at their word and I wonder if anyone actually tries to use these features. When I hit problems. dejanews and altavista searches for similar things usually turn up nuttin on these issues.
OK, I'm ranting as usual. I really really hate platform bigots, UNIX and NT or whatever. I go forr the best tool for the job. I just really get tired of doing careful research, picking a Microsoft solution as the best tool, then finding out I was an idiot for actually trying to implement the solution and expecting it to actually work.
How many times do I have to be abused before I learn my lesson? :-(
Yes, Linux and UNIX programs have their problems too. I couldn't believe the hassles I had to go through to get Amanda to work with my DG/UX box. sendsize silently kept failing to calc disk sizes cause the fork/exec of runtar was screwed up. But you know, fixing it wasn't a big deal. I had the sources right there, went through the code, found the problem as it relates to my OS. I fixed it and will send the patches back to the authors.
The difference is, when something breaks in an open source OS, I can always fix it myself. When something breaks in a proprietary OS, I'm shit out of luck and can only hope that the next version fixes it and the upgrade costs are not too prohibitive for my installed base of thousands of desktops...
The previous era of computing was hardware-limited. That constraint limited the domain of data that could be worked with and upon. We are no longer hardware-limited. We are overflowing with data, of myriad types and dimensions. The next era in computing will be filtering data and adapting it to the person...not adapting the person and data to the limited hardware.
/associative/ resource database, where resources are categorized not by arbitary position on disk, but by their very attributes and data characteristics. We are reaching a new level of abstraction.
Already we are seeing that the notion of a strictly one-dimensional hierarchical file system becoming archaic. Having a system of files is useless if you are so overrun with data that amongst the plethora of files nothing is meaningful. With the network-as-computer approaching I believe we will shift to the systems of "resources". We are already seeing this with distributed computing. URLs locate abstract "resources", whether they are on a local file system, or over the net, whether they are static data, or an executable service or agent. We have to move to an
That being said, a lot of Unix is based on the good old file system metaphor. At the time, addressing everything as a file was as novel as addressing everything as a resource (think system components as CORBA objects - check out AllianceOS). Because of the above reasons I think we need to graduate to a more abstract, associative model. Also, very simple security structures like ACLs are showing their age...they do not scale up well. We are seeing new security models, like capability-based systems, where each entity in a system, be it user, or program, has a set of "capabilities" which it can use to interact with other pieces. A higher level of abstraction. I think for Unix to stay in places requiring these features (associative data storage/filtering - desktop, new security paradigms - large networks, network OSs), it will have to change.
Where things like this don't matter one bit (like the mainframe), Unix will continue to reign supreme.
It's 10 PM. Do you know if you're un-American?
This theory:
:-)
http://home.xnet.com/~raven/Sysadmin/UNIX.html
says it very well. "People are confusing dying with age," and that brief article has a good idea why Unix will still be around for a long time.
(It was written shortly after Lose95 was released.
You cannot apply a technological solution to a sociological problem. (Edwards' Law)
UNIX was poised to die out in the late 1980s. That OS was floundering on the professional desktop in workstations--back when that meant something other than a PC--from Sun and the now-defunct Apollo. But the general UNIXy environment and Xwindows just felt so clunky next to something lighter weight; it looked like some form of personal computer was a good alternative.
Unfortunately, as personal computers became more complex, they also became more unreliable. If Windows 3.0 crashed and you could just hit a button and be back to work in three seconds, then no one would have cared. But you had to sit through an unbearable two minute boot sequence. Networking was messy. Arcane INI files were just as bad as anything from the seventies. As reliability dropped, UNIX began looking more and more attractive. It was still butt ugly, but at least it worked.
By now, we should have something better. We've had an additional ten years to deal with the problem. We should have something very small and very stable and very easy to take care of. A computer should be able to reboot as fast as a calculator. We shouldn't have to deal with driver issues and such as much as we do. But it didn't happen. PCs got faster and more varied, but nothing improved in the reliability or simplicity department. And now, to our horror, UNIX is looking like the simpler alternative. No one would have believed it.
I don't think it'll 'die' exactly. It may eventually evolve into something that bears little resemblence to what it is today, have a different name, have absolutely no code in common with what we know today, but it'll be evolution.
Licenses that make the source available for reuse make this more likely than ever. The open source movement is giving Unix a lot of strength. I'm fairly confident that people will still be typing "ls" to see their files 30 years from now (assuming the keyboard isn't dead by then
numb
There's no reason to "convert" most of our existing Internet/networking infrastructure to anything else in the forseeable future. I agree with the prediction that things will end up moving more towards centralized computer resources,
and lesser-equipped but ubiquitous terminals to access those resources, but Unix will still be there in some fashion.
I don't see the likelyhood of this. All you really have to do is increase the ability for the client to work properly and increase it's capabalities. For somethings (say perhaps tactical nuclear weapons simulations) you may need mainframes however this is not the norm nor very supportable. Applications are mostly writen for PC type platforms considering how much Microsoft has spent convincing people of this.
Who's to say Unix won't be the OS that drives the appliances?
But appliances are just so.. well boring. What would be nice is to have a large mainframe that you could optionally use and use for massive backups of the target machine (say copy the entire image of the client in case of power failure and such) and allow the client to have responsibilities.
Personally I don't want to have some rather fiendish god controlling my computing resources at one particular point. If someone would write an application API that would work like distributed.net and allow for say a complex process to be broken down into many smaller parts and work on any number of client machines that could be increased and decreased at will would be much better. Add into this a possibility to have "relay points" where the data could be copied for a particular portion of the network in case some machine failed or sent corrupt data.
Slashdot social engineering at it's finest
The trick of UNIX: it has always been availiable and highly adaptable to different environments. While this was changing in the 1980s (the UNIX wars), RMS, Linus and all of those open source programmers have insured that UNIX in some variant will always be in use.
If you look at recent corporate inroads, such as IBM-Intel, Phillips TiVo, etc., the market for UNIX like solutions is actually growing!
In the Epilogue of their book, The UNIX Programming Environment, Brian Kernighan and Rob Pike were not as arrogant to think UNIX would live forever, but did have this to say:
"The principles on which UNIX is based--simplicity of structure, the lack of disproportionate means, building on existing programs rather than recreating, programmability of the command interpreter, a tree-structured file system, and so on--are therefore spreading and displacing the ideas in the monolithic systems that preceded it. The UNIX system can't last forever, but systems that hope to supersede it will have to incorporate many of its fundamental ideas."
So long as this statement holds true, I'll gladly work with any future system which provides this set of core ideas.
Plan 9 is UNIX-like, but it treats all system objects like files. This includes objects that exist across the network. Because of this, it is very easy to distribute the OS across several machines with it being completely transparent to the user. We set Plan 9 up like this in the OS lab at my college a couple years back, it's very odd.
It probably isn't an OS that will pick up by itself, but it's an example of a way in which an OS can be distributed with a reasonable degree of transparency.
Unix will adapt and grow- it always has.
The Unix that we use now has little in common with the Unix of Thompson and Ritchie. It has been in a state of continual evolution and will remain thus until ^we^ stop caring about it.
Unix has transitioned from PDP-machine language to portable C, moved from minicomputers to microcomputers (and even to mainframes and PDAs), acquired thousands of tools and roles never dreamed of by its creators. The Unix user today has a choice of command line shells, a choice of GUIs, a choice of vendors, even a choice of fundamental architecture as the file systems and such have evolved quite differently amongst the different Unices.
I think we'll see changes in the coming years, but no greater change than we've seen in the last 30. Unix will continue to evolve until the Unix of our children is as unrecognisable to us as the Unix of our fathers. New hardware and new markets only create now opportunities for Unix to grow; it does ^not^ ring its death-knell.
~wmaheriv
~wmaheriv
"Shema Yisroel- Adonai Elohenu, Adonai Echad!"
I think I ran into that at work once... and it pissed me off. Let me explain... although most of our HTTP and DB stuff is served from IBM Big Iron, we do have a few Sun E450s and a couple of Sun E1Ks. The Suns are loaded with Solaris 7, as is to be expected, and most of the operators (myself included) use CDE. So I'm helping someone write some tapes the other day, and the machine had CDE, and so dumb old me (thinking I was in Solaris) started pounding out Unix commands. After nothing worked, I looked at the title bar on the Xterm. It wasn't an Xterm, but a "DECTerm". I looked under the table, and sure enough, there was a DEC Alpha box. Ah.
The point of all that ranting is, yes, I agree, DEC still makes money.
I don't usually go over to that side of the server room, so I looked around a bit after finding the DEC machine, and what I found scared me. A bunch of 10-year old DEC daisy-wheel printers and reel-tape machines. I had no idea the company was using such old crap. What's even more scary is that the tape machines are even needed: some of our clients refuse to rewrite their media distribtution software to accept anything besides large tapes. The more "up-to-date" clients get the same information by FTP, but these old fogeys -- and some of these are household names -- are using completely rotted software.
My horror was complete when I discovered something I never thought anyone from my generation would ever see in use: an NEC Astra machine, from the mid-seventies! Complete with a proprietary (read: not PC-compatable in any fashion) OS, loaded from eight-inch floppies. The machine was used right up to when it died, on January 1, 2000. (70s hardware and softare isn't quite as Y2K compliant as today's ;-)
I ran and hid behind the IBM fridge racks and haven't been over to that corner since.
I guess I started ranting again. Let me try and make a point out of all that... uh, VMS still pisses me off, and DEC can go to hell. :-D
Too hot for CPAN!! Get PerlOS now from
The Unix of today is very different from the Unix first produced in the 1970s. All things either evolve or die. Unix will evolve in some way or another. Just like mainframes, Unix popularity might wan (I think it already did) but it will come back (I think it already is, as Linux and FreeBSD).
Look at it this way Be *had* to put a level of Unix compatibility in BeOS. MacOS is now a variant of Unix and NT evolves more toward Unix every day.
On the other hand Unix/Linux must be lost to the user in the sense that Unix/Linux at the command line or Xlib level just isn't for the user.
EROS is hard to describe. It's capability-based and has orthogonal persistence -- and if that doesn't mean anything to you, I'm not going to be able to explain it much better. Check out the EROS project site and read the documentation. One thing this means that I can explain, though, is this: "snapshots" are taken of the current state of the system every five minutes. If the power goes out, the system is later restored to the last good snapshot. So you could have a text editor window open, never save your file, PULL THE PLUG on your computer and then plug it back it. Within 30 seconds (or however long your BIOS POST takes to complete), your text editor window is back on the screen, and you've lost no more than five minutes of work.
EROS is cool. I think it has potential to be the Next Big Thing. Check it out, download it (it's GPL'ed), play with it. Have fun.
-----
The real meaning of the GNU GPL:
The real meaning of the GNU GPL:
"The Source will be with you... Always."
Fire has lived well beyond the era in which it was born (the era of stone) and has survived and thrived in the era of bronze, but now, many people believe that soon we will see the transition from that era into the age of iron. In that case, has the Slashdot community at large ever considered what the future is? Will Fire finally die off, will it adapt as it did before, or will Fire find a way to remain the same trustworthy system it always has been? And if Fire will come to an end, what does the Slashdot community feel will be its succesor?
--
Infuriate left and right
I don't know what will become of Unix, and whoever says they do is not only a fool but a liar as well.
It was only a few years ago that I was mourning the apparent *nix recession; The only game in town was Xenix/AT&T (and a wee bit of Sun, but not in my neck of the woods), and their products were both languishing and confined to minis.. Linux and the *BSDs were infant, not worth a mention outside of academia. Now it has come full circle. People are using *nix [gasp] ON THEIR DESKTOP! I can run *nix on everything from my multi-million dollar IBM to my $100 garage sale throwaway. And it is adapting again. Embedded Unix? I would have laughed my ass off if someone had suggested running Unix on a microcontroller only a few years ago..
I kind of suspect that *nix is just too adaptable to die, but to say whether or not it will be beaten back onto the mainframes by PalmOS run-PDA's in a decade is impossible.
.sig: Now legally binding!
The advent of distributed, collaborative, pure-hype^H^H^H^Hjava, Active System Blaster 2000 will not make Unix obsolete. Even a revolutionary, fully distributed and autonomous network object system still needs to send bytes over the wire, still needs to access system memory, and still needs to accept input and create output. These are the things that Unix provides. This is why Unix will always be around.
I suppose that a newer operating system could supplant Unix. However, I don't seen any in the near future. Be has a bright future, because it has networking and other nice capabilities. But Unix has the trump card over BeOS: the idea of users. In a distributed network environment, the user concept becomes much more important. Information, files, interface configuration, and many other things are associated with a person. Those things must be secured from other people, and the other people must be secured from them. Therefore any OS that wishes to supplant Unix will need to supply the same kind of protection for users' information.
Cheers,
jwb
That's funny, I thought that Unix was based on a monolithic kernel... silly semantics
I would love to see some of the best coders and operating systems people put together a new OS from scratch using the latest techniques. Ideally this would create an ultra stable and very modular system. I would happily give up some extra CPU cycles for increased modularity and the ability to easily swap in and out OS components so that I could customize my OS to the task at hand. I find it rather ridiculous that I run the same OS when I am playing games, running a web server or working with Photoshop (etc). Rather than having a generically-good OS I would prefer a highly optimized OS for the task(s) at hand.
How often do I run run a game, Photoshop, compiler, and web server concurrently on my home box? Give me adaptibility and modularity or give me death!
Complexity Happens
Unix is not a product, it is a set of evolving ideas. As such, it is not going anywhere but up.
It's just as easy to ask: Is this the end of silverware, or the end of fire, or the end of any old thing that's proven to work... Is genetic engineering the end of agriculture? Is organ transplantation the end of death? Is The Bomb the end of War?
Yeah, there's micro-kernel based OSes out there like Qnx and NeXT, and Hurd... But they're still Unix. The NEW OS X from Apple is more Unix than it's predecessor. NT is more Unix than Win95.. There's new approaches like BeOS.
If one defines Unix in a very constrained way then Unix has been dead for a long time. When AT&T first released System V, and allowed it to mutate, Unix died and was reborn in a variety of ways. Umm, that was what? 1976?
If one defines Unix broadly, as a set of concepts, a layered architecture, levels of abstraction, sets of small uni-purpose tools working together, APIs and things 'grep-like' then guess what? Unix will live forever.
It's a pointless question, not because it's bad, but because it's completely subjective.
-- What you do today will cost you a day of your life.
Unix is too broad a family of systems to "die". It's not like AmigaDOS, or VMS, where there's just one Unix and it can "fall behind". Unix will be replaced, but it'll be replaced by more Unix.
My blog: http://www.seebs.net/log/ --- My iPhone/iPad app: http://www.seebs.net/seebsfrac/
I have seen the future. The future is filled with operating systems that demand that their system binary directories be writable to all, else they fail.
I have seen the future. It has an operating system whose applications, even those written by the OS authors, can ignore the TEMP environment variable and scribble temporary files where-ever they want and fail to operate if they can not do as they wish.
I have seen the future. The future is filled with continued support for legacy drive letters and 8.3 file names with rename.ini kludges during installs.
I have seen the future. The future is an operating system where you have to shell out serious dollars to buy third-party utilities to get around security deficiences in the design of the OS. After all, why fix that pesky virus problem when so many anti-virus companies would go under without that revenue stream coming in.
I have seen the future. It has operating systems whose file systems don't support the concept of being able to delete a file yet have it not actually get deleted until the last remaining process that has it open dies first. For doing so would prevent the need to put dynamic link libraries into a temporary space and have them "installed" during a reboot. Reboots are good. They clear up sloppy OS design problems.
I have seen the future. The future is filled with grand marketing schemes like "Administration Kits" that promise all kinds of abilities to deploy corporate policy restrictions to users yet neglect to mention that these policies are applied by a program that has to write to an area of the OS that was previously recommended be R/O due to the security problems it causes if it is R/W, hence making the ability to make the scheme work as advertised impossible for all users who do not have full permissions to their workstation.
The future, my friends, is about image and not function. UNIX is ugly. It's doomed.
Or in other words, resistance is futile. At least that is what they want us to believe... :)
There's no reason to "convert" most of our existing Internet/networking infrastructure to anything else in the forseeable future. I agree with the prediction that things will end up moving more towards centralized computer resources, and lesser-equipped but ubiquitous terminals to access those resources, but Unix will still be there in some fashion.
Who's to say Unix won't be the OS that drives the appliances?