Leo Laporte On UNIX As the Future

TractorJector writes "In a well-written interview with Mad Penguin, techmeister Leo Laporte (formerly of G4/TechTV fame) discusses his vision of the future of proprietary and open platforms: 'I think there's a lot of hope for Linux, although I don't think that Linux is the answer. I think that UNIX is the answer, in some form or fashion. It might be BSD, it might be Linux, it might be some third thing. But UNIX is such a well understood and smart to handle the issues that an operating system has to handle that it ultimately will prevail.'"

Comments

Unix is not the Future

[AKAImBatman]

Unix is very flexible, and it certainly outlive Windows. However, its development will only take it through the near future. In the long term, the very idea of unmanaged code will disappear. As will the traditional concept of the Desktop.

My predictions are:

1. Desktops will be replaced with Browser simulations of a Desktop that can work anytime, anywhere.

2. The traditional PC will then be replaced by a home server through which all activity will happen.

3. Components for Music, Television, Desktop, and Video Game consoles will (in many cases wirelessly) interact with this server.

4. The server itself will run an OS based on a managed code environment, making remote attacks difficult if not impossible. (Many Unix concepts would probably be reused in this system, but it won't *be* Unix.)

That's my thoughts anyway. Sometime in the near future, I'll get them blogged down in detail. :-)

Re:Unix is not the Future

[Anonymous Coward]

1. Desktops will be replaced with Browser simulations of a Desktop that can work anytime, anywhere.

2. The traditional PC will then be replaced by a home server through which all activity will happen. ....

Thin clients? Haven't we visited, revisited, gone back and suggested this again while implementing it? Citrix or X anyone?

Re:Unix is not the Future

[ravenspear]

The server itself will run an OS based on a managed code environment, making remote attacks difficult if not impossible.

Unless of course the server runs an MS OS, in which case it will actually open you up to attacks from all of the different types of devices interacting. ;)

But seriously, I'm not sure this managed code model is the answer. At the very least it needs to be designed very well. I could see it being very restrictive for a lot of legitimate uses. And I think having a model where the OS has to approve code before it runs opens the door to monopoly leveraging, unfair treatment, unauthorized runtime limitations, and a whole host of other undesired behavior.

Re:Unix is not the Future

[AKAImBatman]

Citrix and X are far too low level. You need to think higher level. Think NeWS. NeWS would have creamed X-Windows in the market if Sun hadn't tried to keep it proprietary. Today, it's all happening again with Web Browsers. The difference is that X and HTML are complementary.

Re:Unix is not the Future

[CrayzyJ]

Interesting commentary. One thing though:
> Components for Music, Television, Desktop, and Video Game consoles will (in many cases wirelessly) interact with this server.

This sounds a lot like Network Stations that were tried around eight years ago. They were touted as the next big thing. The idea, though sound, just didn't take off as some thought it would. Perhaps the Network Stations were ahead of their time (i.e. like OS/2)?

Re:Unix is not the Future

[AKAImBatman]

And I think having a model where the OS has to approve code before it runs opens the door to monopoly leveraging, unfair treatment, unauthorized runtime limitations, and a whole host of other undesired behavior.

That's not what managed code is. Managed code is systems like Java, .Net, and LISP that eliminate direct hardware access, thus preventing system bugs like buffer overflows. Java is a particularly good example, because it has a very flexible built-in Security system that could be leveraged to ensure that a given program ONLY has access to the resources it was given at install time. :-)

Re:Unix is not the Future

[Lost Found]

Yeah, I remember hearing years and years ago that in the next few years, all PCs will be nothing more than Java runtime environments, and you'll rent your applications over the Internet from providers.

Guess what? It didn't happen.

What you describe in your post would take a substantial amount of work from many companies (not to mention a very slow migration process of the end users to completely shift paradigms). Companies doing this will likely do it incrementally if they do it at all (because software comapnies can't afford to go in a cave for 10 years without making money). In any case, companies doing it incrementally are likely to leverage the excellent open source UNIX technology before them to make it happen, rather than reinvent computing from the ground up.

Re:Unix is not the Future

[dsginter]

2. The traditional PC will then be replaced by a home server through which all activity will happen.

This is what I've also predicted. Here are my thoughts:

A typical family might have two or three computers and a PVR or two. If the hard drives on all of these devices were aggregated into a single, logical server, then there would be benefits in terms of utilization, redundancy and speed - panacea. If we tie everything together with GigE, then we can PXW/network boot the PCs and PVRs with any operating system of our choice. Laptops would be a bit more complicated but I would envision a large, solid state cache on board. The hard drives in laptops are always failing anyway.

Re:Unix is not the Future

[ravenspear]

Ah, my bad. It seems I was confusing managed code with Palladium, TCPA, etc.

Re:Unix is not the Future

[AKAImBatman]

Perhaps the Network Stations were ahead of their time (i.e. like OS/2)?

That was certainly part of the problem. But as an admin who ran Citrix, I can tell you that the other half of the problem was Microsoft. After Citrix gained some initial momentum from their NT 3.51 product, Microsoft took notice and refused to license 4.0. Instead, Microsoft worked out a technology transfer deal where they would produce NT Terminal Server. Citrix was "allowed" to install their superior ICA protocol on top.

The result was that you had the initial price of Terminal Server, plus the price of each "Seat" (which was in number of users, not concurrent connections like Citrix), then the price of a full copy of Windows NT for each thin client that would access the system. If you wanted Citrix ICA, you then had to pay Citrix even more.

The result was that Thin Clients ended up costing *more* than a set of PCs, effectively killing the market.

Fast Forward to today, and we find that Windows now has the RDP client integrated and that Sun has been having reasonable success with their SunRay product. People are starting to become conditioned to the idea of thin clients. Wait a few more years for the WebApp revolution to shift all power away from windows and the time will be perfect to wretch the market away. ;-)

Re:Unix is not the Future

[Anonymous Coward]

I'm still waiting on flying cars.

Re:Unix is not the Future

[sgant]

Very interesting, very insightful comments...unfortunately we've been hearing the exact same thing for the past 20 years. You're basically regurgitating what the pundits and "experts" have been saying forever. We've yet to see any of these things....

That's my thoughts anyway. Sometime in the near future, I'll get them blogged down in detail. :-)

No, these are not your thoughts. These are the ideas from about a 1000 different people over the years that have been saying the same things. It's bad when you have to take the work or ideas from others and switch them around to try to make them your own.

You offer nothing new here...move along.

Re:Unix is not the Future

[AKAImBatman]

You're basically regurgitating what the pundits and "experts" have been saying forever.

Yes and no. Anything I say as a tech professional will ALWAYS be standing on the shoulders of giants. There's simply no way around that. However, these "experts" you're referring to have always been insensitive to the timing, and have offered no solid solutions to solving problems. While I'm making an abstract prediction now, I fully plan to make a solid prediction in the near future. :-)

We've yet to see any of these things....

Not true. It is becoming quite popular to purchase a computer with a Video Capture Card, use a LCD TV as the monitor/television, hook your computer up to your Dolby 5.1 speakers/stereo, download music and videos from the 'net, and use applications via WebApps. I'd say it's staring us right in the face.

Re:Unix is not the Future

[DartonW]

The problem is, not everything can be done with managed code. There will always be a need to get down to "bare metal" coding to get some things done.

For example: Has anyone ever tried printing actual TEXT to a printer (not an image created from text input) on the .NET platform? I have, and lemme tell you, managed code won't do it unless you consider creating managed libraries utilizing unmanaged code to be still in the realm of "managed code".

Re:Unix is not the Future

[Anonymous Coward]

mod parent up +1 insightfull

Re:Unix is not the Future

[Master of Transhuman]

I agree.

The main problem with the thin client solution has always been that if the server goes down, everybody goes dead. And the server ALWAYS goes down. This is unacceptable to any right-thinking CIO - and even more so to the people who are actually doing the work.

Of course, if you have proper system design, with failover and redundancy, this is less of an issue.

According to recent trade media reports, thin clients are now on the upswing simply because of Windows - no need to patch ten thousand thin clients with every new security hole.

Naturally, Linux is going to win this one, too, since you won't need to patch the thin client server (as often, that is - Linux isn't perfect) either...:-)

Re:Unix is not the Future

[Q2Serpent]

Remember that buffer overflows are not the only security vulnerability. Currently they happen to be a large one, but there were many exploits that were possible simply because of bad logic.

I remember an IIS flaw that was exploited because the server decoded a URL, checked to see if it was valid (i.e. not pointing to some arbitrary thing outside the document root), then before opening the file, decoded it AGAIN! This second decode was done without a second check, meaning that a URL that decoded twice into something harmful passed through.

Managed code does fix some things, but it's not a cure-all.

Re:Unix is not the Future

[AKAImBatman]

That's not a problem with managed code, that's a problem with the APIs offered by the managed code. There's nothing stopping a managed code environment from allowing you low-level control.

For example, JNode is a complete OS written in Java. It's still Work In Progress, but I'd imagine that you would have no trouble writing a simple text driver for your printer.

Don't confuse what you're currently allowed to do with what is possible. :-)

Re:Unix is not the Future

[OreoCookie]

Unix is very flexible, and it certainly outlive Windows.

That remains to be seen. Right now Windows sales are very strong.

From IDC:
Worldwide Operating Systems and Subsystems 2004 Vendor Shares: Microsoft Extends Its Dominant Revenue Position
Jul 2005 Doc #33719

"Microsoft has once again extended its revenue position in the operating systems and subsystems market, further distancing itself from the pack," said Al Gillen, research director, System Software at IDC. "The only other platform segment where we see appreciable revenue growth is Linux, but the disparity is significant, with Linux capturing a mere 1% of the market revenue compared to Microsoft's 70% share."

According to Wall Street figures the Microsoft Server and Tools division 2005 sales are up 34% over 2004

Re:Unix is not the Future

[acvh]

Thin clients have been well received by our corporate IT due to the ease with which applications can be updated, users can be given remote access, and local support people can be terminated.

Users hate them because there are weird sync issues, files change or disappear at random intervals, they can't listen to music via their "pc".

There is no one "server" that can go down to screw everyone up. A farm of three or more machines is standard practice here. Thin clients are NOT cheaper than PCs, until you factor in the headcount that gets cut.

Re:Unix is not the Future

[AKAImBatman]

I remember an IIS flaw that was exploited because the server decoded a URL, checked to see if it was valid (i.e. not pointing to some arbitrary thing outside the document root), then before opening the file, decoded it AGAIN! This second decode was done without a second check, meaning that a URL that decoded twice into something harmful passed through.

This is where Java's security model would have gotten in the way. When the file open request was received, it would have said "You don't have access to these directories, go away!" and it would have failed to open/run the file. Of course, that takes a token security system to implement, but Java is particularly well suited to this, as are many other managed code environments. :-)

Re:Unix is not the Future

[DartonW]

Good point. The issue I have with some of the managed code platforms out there (especially the current direction of .NET) is the closed nature of them, which leads back to a sort of proprietary interface where we're coding more to the platform and within its restrictions rather than opening the system calls themselves up with a managed interface.

This was certainly part of the goal with MFC, ATL, STL, GTK and a whole host of other class libraries and API's, even though they were written in unmanaged runtimes. I think the goal should be not to protect us from ourselves (the MS way), but rather to shield us from differences in hardware, filesystems, etc.

Re:Unix is not the Future

[Anonymous Coward]

My own prediction is that we will eventually be interfacing with the computer in an immersive VR environment. Sure, we won't have MS Windows but they'll still be around. Just with a new paradigm... take the idea of a desk to the next level (in VR space) and you'll wind up with MS Cubicle. I can already think of the advantages:

* New improved searching mechanism allows you to search through your documents (papers) in any directory (file folder inside of that filing cabinet beside your virtual desktop) intuitively. Just pick up a stack and start thumbing through them. You can scan as much or as little as the document as you need to in order to find something.

* Familiar apps such as calculator and notepad will be located right on your virtual desktop for easy access.

* Non-intrusive communication with others. You can set your IM application to make any number of ringing noises when someone is trying to contact you at which point you can choose to answer or keep working on what you're doing. Alternatively, you can easily put your IM app in away mode and have it visually signal you if any new messages come through.

* Expanded themes allow you to customize your workspace. Do you prefer a natural wood feel or is gaudy modern plastic your cup of tea?

And the list goes on. "MS Cubicle makes your workers more productive by increasing the ease of use and lowering the amount of training required for a new employee. For small to large businesses, the new version of MS Office for MS Cubicle will also let your employees collaborate on projects like never before!"

Sure, you're laughing now but just wait. As soon as someone invests a reasonable immersive VR headset with the ability to allow full interaction in a 3d environment, someone will ask the question: "How do we make a paradigm to allow user's to interact with the computer in VR in an easy to understand and intuitive way?" This will most certainly be followed by at least one person saying, "Easy. Just take the 'desktop' GUI paradigm that everyone is already used to and expand it to fit this new virtual 3d space."

Re:Unix is not the Future

[hendridm]

Very insightful, if only I had mod points. I agree. The computer of the future will be sitting next to the furnace and water softener (hopefully on risers), and LCDs will connect to them for functionality.

How convenient it would be to connect my LCD to an ethernet port in the wall and have full access to the services of the main server in my basement (which would include virtualization capabilities, if I were in charge).

My house currently has 7 computers in different parts of the house used for different purposes. 4 Windows desktops, 1 Windows laptop, 1 Macintosh, and 1 Linux box for firewall/proxy. Granted, I'm no average user, so we can eliminate the firewall/proxy. That leaves us with 6 machines. I want one server with multiple monitors throughout the house, just as easy at it is to plugin an extra TV.

Oh, and before you mainframe geeks chime in, a green screen isn't acceptable. I want the capabilities of my desktop on a central computer combined with the simplicity of plugging a monitor into an Ethernet port.

Re:Unix is not the Future

[Anne Thwacks]

LISP that eliminate direct hardware access

In what way does LISP eliminate hardware access? Not in a "LISP machine", surely? In any case, its two fundamental instructions "CAR" and "CDR" stand for Content of Address Register, and Content of Data Register! These were certainly hardware registers sometime in the 1970's when LISP was invented!

My other CAR is a CDR is fine on your bumper, but don't try to execute it!

Re:Unix is not the Future

[photon317]

The problem is that these high level managed code environments tend to breed dumb programmers. I have found it easier to trash content and/or steal data from java servlet based apps than perl cgi's from the outside (I do this on the side for freinds, to test their code before they give it to some unwitting customer). Java programmers (at least the ones I know) tend to not think through all of the security ramifications of what they are doing - and the fact that they are so abstracted from what's really going on is partly to blame.

Re:Unix is not the Future

[MartinG]

Managed code is the answer for the majority of applications, but right now it has drawbacks for certain scenarios that make it a problem.

The non-deterministic behaviour of current garbage collectors mean is just can't handle some types real time high fidelity audio processing for example.

Also, the code manager (i.e. the vm) has to run somewhere so the whole OS cannot be managed right down to the hardware level. (unless you implement the VM in hardware, but that's another discussion)

Re:Unix is not the Future

[DartonW]

Real-time does seem to be an issue, but I guess we're getting off-topic here. Is the future *nix with a VM in which all apps run (like a cellphone or PDA)?

BTW, props on the email munging, Martin.

Re:Unix is not the Future

[Bob Uhl]

I think what he was probably thinking of was the fact that most modern languages prevent buffer overflows and the rest. Lisp actually partakes of both natures, though: by default the language is safe but not as fast as it could be; you can tweak it to make speed more important, and you can even tweak it to make safety less important. It's kinda cool, actually.

I urge anyone who's not read it to take a look at Practical Common Lisp, which is an excellent introduction to an excellent language.

Re:Unix is not the Future

[benjamin_pont]

PS3's descendants talking wirelessly to a chip in our brains is our future. ;)

Re:Unix is not the Future

[Anonymous Coward]

Note that in the process of mandating Terminal Server Microsoft killed NCD, who had a much nicer solution for Windows in the form of integrated X in the Windows desktop, and X terminals capable of talking to Windows servers.

The bright side is that this freed Keith Packard to get back to core X work. Not a consequence Microsoft would have intended, I think. :-)

Re:Unix is not the Future

[NickFortune]

Ah, but the question is: what does the managing?.

A lot of the security of having a virtual machine lies in the fact that is is in fact virtual, and that all an attacker can do is crash the proram he's attacking. But if your runtime becomes the OS in fact, then the machine stops being virtual.

Re:Unix is not the Future

[LWATCDR]

This is an interesting idea excpet for a few issues.
1. Home Server == single point of failure. Think about if you had one box that ran your phone, all your televisions, and your home automation. Now imagine that box going down.
2. Managed code? Good thought and it can be added to unix with out much of an issue. Mono and java are prime examples.

I do see a place for a home server but I see several servers. I can see one file server, a phone server, and a video server. I can also see a home network that ties all of of your little boxes into a home network.
A single centralized home server is just too risky.

Re:Unix is not the Future

[iabervon]

You don't get direct hardware access from userspace under a moderm OS regardless of the language or environment; the actual processor hides that from you without needing a virtual machine. What Java and Lisp hide from you is arbitrary access to memory in your address space, which means that the system can safely store data in your address space, you can have multiple programs that don't entirely trust each other in the same address space, and that buggy code can't screw up the program state into an incoherent state relative to the programming language model.

Re:Unix is not the Future

[TheRaven64]

Java and JOGL provide an interesting possibility for re-inventing NeWS. The main limitation of NeWS was that views had to be written in PostScript, while the rest of the application had to be written in something else.

If, however, every terminal had a Java VM and OpenGL, then the view could be sent to the display and the controller / model logic could remain in the server, and the whole thing could be written in Java (or some other language that targeted the Java VM).

Re:Unix is not the Future

[AKAImBatman]

A lot of the security of having a virtual machine lies in the fact that is is in fact virtual, and that all an attacker can do is crash the proram he's attacking.

That's not actually correct. If a managed environment is correctly implemented, it should be altogether impossible to crash the environment. The worst that an attacker could do is cause an Exception (which travels back up the stack until it's handled in some form or another) or a DDOS.

But if your runtime becomes the OS in fact, then the machine stops being virtual.

The Virtual Machine is an OS. The key is that today's VMs rely on the host OS for basic functionality. Add the necessary features to the VM, and you no longer have a need for a host OS.

Also, do not forget about the Java Security System. Any attempts at a breach of security can be easily walled off, trapped, and traced back to the source. This functionality isn't used much today (mostly due to the attempts to make Java programs act like Desktop Apps), but its potential is staggering.

Re:Unix is not the Future

[AKAImBatman]

I've actually done some work in this area, and I've come to one conclusion: It's far too heavyweight to work on a large scale. All you really want is a shell that can render the screen well. Java does not provide these facilities, but instead focuses on the details of the filesystem, network, threading, and other powerful concepts.

IMHO, the model that works best is:

HTML/CSS/JavaScript (Client) Java (Server)

The client is as dumb as it gets. It can only display information and take input. The server, OTOH, is capable of advanced processing and I/O functionality. So the application you see on your screen is just a webapp entension across the network, onto your screen.

Applets would still be used, but only for very rare heavyweight functions such as video streaming. :-)

Re:Unix is not the Future

[AKAImBatman]

Gah! Stupid SlashCode. The model should look like this:

HTML/CSS/JavaScript (Client) <-> Java (Server)

Re:Unix is not the Future

[NickFortune]

That's not actually correct. If a managed environment is correctly implemented, it should be altogether impossible to crash the environment. The worst that an attacker could do is cause an Exception (which travels back up the stack until it's handled in some form or another) or a DDOS.

Or, if it isn't handled at all, the application crashes. At least, that's been my experience of a lot of Java programs, anyway. Still, perhaps I should have said, the best an attacker can hope for is to crash the application.

The Virtual Machine is an OS. The key is that today's VMs rely on the host OS for basic functionality. Add the necessary features to the VM, and you no longer have a need for a host OS.

I understand that. Look, suppose a black hat finds an exploit that lets him tweak the parameters of the Java runtime. In the vast majority of cases all he can do is inconvenience himself. He can't get root access, unless the runtime was running as root, becausehe still has to crack the underlying OS. He can't mess with another user's activities, because that other user is running on a separate VM and changes he makes to his VM have no effect on hers.

Now if the entire OS is a java runtime, then these protections no longer apply. The runtime has access to superuser privileges, and changes made the runtime affect all users.

Also, do not forget about the Java Security System. Any attempts at a breach of security can be easily walled off, trapped, and traced back to the source. This functionality isn't used much today (mostly due to the attempts to make Java programs act like Desktop Apps), but its potential is staggering.

That's great, to the extent that it works as it's supposed to work. It's not, after all as if all those buffer overflows in C were coded in because the coder thought they were cool[1]. The secutity model was designed to be watertight.

In a complex system, and IMHO obviously, there are always going to be exploits.

[1] OK, maybe one or two.

Re:Unix is not the Future

[jav1231]

Okay, you're digging up 10 year old predictions here. You may very well be right but nowhere near in the near term or within say 10 years. As long as you have gamer/devel types hungry for powerful desktops, you'll have hackers (strict sense). As long as you have hackers banging code and working on their projects, you'll have a desktop. The Pandora's Box has been opened on hardware and getting it back in (i.e. thin clients) is going to be very tough. I think you can make a great case in corporate environments but for the average user, there's going to have to be a huge catalyst to make this happen.

Re:Unix is not the Future

[jonabbey]

I have never understood why Sun, upon designing the Graphics2d imaging model for Java 2, version 1.2, SE, didn't immediately propose and implement a standard X11 extension that would allow that imaging model to be carried out inside a standard X server.

We're finally moving in that direction now with the XRender extension and the like, but Sun could have given Java on *nix a nice boost 5 years ago if they had had the foresight to improve Java's performance on Solaris, Linux, and BSD in this way. They would also have seen the development effort that has gone into the underpinnings of Gnome and KDE focused more on the semantics of their APIs, making it easier for desktop developers to switch over to Java.

Java's performance over a remote X connection plummeted when 1.2 came out, and it didn't have to be that way at all.

Re:Unix is not the Future

[shmlco]

Wait a few more years for the WebApp revolution to shift all power away from windows...

Ain't gonna happen. Or rather, there are still large sets of problems that need robust applications running locally. I'm supposed to upload my 400MB Photoshop image to a remote server and work on it there?

The fact is that we had "dumb" terminals before, which gave away to smart terminals which gave away to PCs running applications and client/server applications.

Why the change? Because the user experience is several orders of magnitude better when things happen instantly, and because offloading a good portion of the work to the edge of the network means that everyone is not competing for the use of the same exact set of resources.

Now we have browsers that are increasing in complexity with AJAX, scripts, embedded components, xml, xforms, and other technologies. Why? Because we want better and richer and more productive user experiences.

Re:Unix is not the Future

[bofkentucky]

And how are you going to play Doom on that setup? Are you going to trust the "server" with your excel files, quicken logs, email, etc? I think not.

Re:Unix is not the Future

[AKAImBatman]

Nope. You're going to walk over to your game console gizmo and use that. The server will automatically pipe the video output to the LCD television you're using. The console will pull its software from the internet (No media!) via the server, and use the server as a gateway for internet gaming.

Use the right peripheral for the right job. :-)

Re:Unix is not the Future

[AKAImBatman]

Or, if it isn't handled at all, the application crashes.

No, actually. If it isn't handled at all, the current thread dies. If you're using a single threaded program, then your program is dead. However, networked programs are almost never single threaded. That's why C/C++ servers used to come down all the time on errors whereas modern J2EE server handle the error as best as possible and move on. The biggest problem with servers today is accidental memory leaks.

I understand that. Look, suppose a black hat finds an exploit that lets him tweak the parameters of the Java runtime.

Never should happen. That's what the Java Security System is for. Even without that in place, each program should be running in a semi-separated environment. I actually have such a system on my laptop which allows telnet logins and servers to all run under the same JVM with no impact between programs. The System class has been subtly modified to prevent programs from changing the system wide parameters. If your program writes to System.out for example, I guarantee you that the message will get to the right place.

Now if the entire OS is a java runtime, then these protections no longer apply. The runtime has access to superuser privileges, and changes made the runtime affect all users.

You're asssuming a single user environment when you shouldn't be. If you extend Java down to the OS level, you have to add standard multi-user protection. The key is that it's much easier to add this protection in Java than any other language. For example, starting a program in its own classloader would prevent it from interfacing with any other code other than system classes.

That's great, to the extent that it works as it's supposed to work. It's not, after all as if all those buffer overflows in C were coded in because the coder thought they were cool.

No, they weren't designed that way on purpose. However, C/C++ have no built-in security model. There's no way to check for that other than attempting to build your own security framework on top, and/or relying on the OS to do its job. Both options are far from perfect. Java has no such problems.

A Java security manager can check every file passing through the system. A Java security manager can check every network connection. A Java security manager cannot be uninstalled by an attacker unless you give him permission to do so. Java's Security model is inescapable as long as it's used and not ignored.

Re:Unix is not the Future

[Lost Found]

Well, and keep in mind that the move to service-oriented systems in the home user sector means that you'll be asking those users to pay monthly fees for something they used to get for free (hey, Windows and Works came on the PC, didn't it?)

The idea that all apps except the browser will move off the desktop and centralize flies like an anvil... it has nowhere to go but down.

Re:Unix is not the Future

[DA-MAN]

1. Desktops will be replaced with Browser simulations of a Desktop that can work anytime, anywhere.

Highly unlikely. With hardware getting cheaper it makes no sense to outsource your applications to a server.

Having your data available everywhere is not going to outweigh the inherent security risks associated with putting your data on the public internet. And encryption can only go so far, without you locking yourself out. No company is going to implement proper encryption due to the irreversable nature of a lost password.

Combined with the fact that this will probably cost money. Free e-mail is one thing, ads are part of the business. No one is going to want to be blasted with ads while trying to write a paper or a report. Why pay for a company to have your word processing docs and other stuff when disk is so cheap. Why pay monthly for a word processor when you can get OO.o for free or buy Office and be done with it? In addition I've had server unavailable messages from Gmail, Hotmail and Yahoo. I don't know any business or person who would take that risk with their word documents.

2. The traditional PC will then be replaced by a home server through which all activity will happen.

So desktops will become servers? Huh? I'm just not sure what you are talking about.

3. Components for Music, Television, Desktop, and Video Game consoles will (in many cases wirelessly) interact with this server.4. The server itself will run an OS based on a managed code environment, making remote attacks difficult if not impossible. (Many Unix concepts would probably be reused in this system, but it won't *be* Unix.)

Possible, although what the server will run is anyones guess since I'm still not clear on what you mean by the server.

Re:Unix is not the Future

[AKAImBatman]

Think of it this way:

I purchase a new home computer package from Best Buy and bring it home. When I open the box, I find a box that's similar to an integrated stereo/receiver, CD Player, and DVD Player, but with an integrated network switch and Wireless hub support. I plug my LCD TV into this box, all my speakers, and my internet cable.

I now have a home entertainment center where I can watch television, listen to the radio, download MP3s, watch TV Shows and Movies off the 'net, and other cool stuff. I just have to pull the remote out of the box to get started.

But wait! What's this next to the remote? Why, it's what looks like a super-slim laptop! I open it up, and it automatically connects to my "home server" device using the wireless hub that I noted earlier. The desktop is projected to this laptop via a wireless Browser session. From this gizmo, I'm able to surf the net, edit text file and spreadsheets, and do pretty much most of the Desktop Office work I could do with a full computer. The only difference is that all the files are actually stored on the "home server" device.

So then I go out and I purchase a game console. There seems to be no place to put a Disc, but there is a network port. So I hook it up to my "home server" and turn it on. There I'm presented with a shopping list of games I can purchase! Super-cool!

So I select a game and wait for the server to load the first level. When it does, I start blasting away monsters. I get tired of single player, so I kick it into multi-player play. It immediately starts playing games over the internet. No console configuration required!

So then I get a TV in another room. I buy a little "Wireless TV" box to plug it into. This Wireless TV box connects back to the "home server" in my living room over wireless! Soon I'm watching TV, playing music, and other entertainment stuff in another room!

Once a secure, managed code OS is used on this "home server", there's no reason why you can't automatically share your Desktop on the internet, so you can always access you stuff at neighbor's, coffee shops, hotels, and work! Your "home server" could even run its own mail server and website!

Does that clarify the idea? :-)

Re:Unix is not the Future

[NickFortune]

If it isn't handled at all, the current thread dies. If you're using a single threaded program, then your program is dead. However, networked programs are almost never single threaded.

Can't assume an attack's going to be over the network. Could just as easy be a trojan. They do tend to be single threaded. All of which is a bit beside the point.

Never should happen. That's what the Java Security System is for.

Should, certainly. The way I see it, the Java security system is like a theoretical cryptosystem. It works fine on paper, and a fair few gedankenexperiments have been performed. However, until someone tries to crack it in earnest and out in the wild then we can't be sure that nothing has been overlooked.

You're asssuming a single user environment when you shouldn't be. If you extend Java down to the OS level, you have to add standard multi-user protection.

Yes, I get that. It's just that in doing so you forgoe the protection previously afforded by the OS. It's not solving the problem, it's just moving responsibility for the problem.

The key is that it's much easier to add this protection in Java than any other language.

The question is whether the language is the proper place to add such protection. I'll come back to that.

However, C/C++ have no built-in security model. There's no way to check for that other than attempting to build your own security framework on top, and/or relying on the OS to do its job. Both options are far from perfect. Java has no such problems.

Actually, relying on the OS to do its job does seem to me to be the perfect solution. Granted, you need an OS that does its job properly, but then a buggy java runtime would have the same problems as a buggy OS.

I think there's a lot to recommend the idea of letting the OS handle security and leaving the language to describe the logic needed to solve a specific problem. One of the BSDs, OpenBSD I think, demonstrates how well this can work. Currently it has zero outstanding security advisories and a policy of full disclosure.

You can't get much better than that. (And I say this as a linux guy with no axe to grind).

Java's Security model is inescapable as long as it's used and not ignored.

And C's parameter passing mechanism is secure so long as the programmer always checks buffer lengths. Everything is secure if it's done properly. Maybe I'm misunderstanding you here?

Please understand: I think it entirely possible that Java could turn out to be the Best Way To Do Most Things(TM). On the other hand, as must be apparent by now, I also have strong reservations about the wisdom of intergrating OS and language, and I think it's sometimes too easy to get carried away by Sun's hype and Gosling's enthusiasm.

So for me, what it comes down to is:

1. Java could well turn out to be the answer to our security prayers
2. On the other hand, it's been shown that conventional techniques can work near perfectly.
3. I'm going to wait and see how the architecture does in the wild before I get too excited.

Re:Unix is not the Future

[VE3MTM]

Bull.

A few months back, I wrote a web service wrapper around a Zebra 2844 thermal printer, which sends commands (in ELP2, a printer command language that Zebra printers use) to the printer through the parallel port.

Yes, it involves using the Win32 API to open a stream to the LPT1: device, which is less than ideal, but it does not involve any unmanaged code.

Wrap that in a library, and it's as good as managed.

Re:Unix is not the Future

[bubbaD]

Many people totally misunderstand LISP and its features. Its sad, but in any case the OP's case about the future of unix and managed code is wrong, anyhow. PS Did you know that lemmings don't mass suicide http://www.snopes.com/disney/films/lemmings.htm

Re:Unix is not the Future

[spectecjr]

This is where Java's security model would have gotten in the way. When the file open request was received, it would have said "You don't have access to these directories, go away!" and it would have failed to open/run the file. Of course, that takes a token security system to implement, but Java is particularly well suited to this, as are many other managed code environments. :-)

If you're running IIS as the internet account instead of as local system, then you can get the same effect just by correctly administrating your system, and locking the internet account out of the parts of your filesystem it's not supposed to have access to.

Java's security model doesn't help you at all here.

Re:Unix is not the Future

[AKAImBatman]

Can't assume an attack's going to be over the network. Could just as easy be a trojan. They do tend to be single threaded. All of which is a bit beside the point.

But what would the trojan do? Would it simply run a program just to crash it? That seems kind of pointless. The point isn't that threads don't die. It's that it's impossible for an attacker to use this in any meaningful way.

However, until someone tries to crack it in earnest and out in the wild then we can't be sure that nothing has been overlooked.

It HAS been tested in earnest. Applets are an example of an area where the Java security model is in effect. There was exactly ONE semi-successful virus (see: Strange Brew), and it was only able to spread on systems where the Security Manager was not in effect. i.e. Your standard desktop applications. There is one other issue that I'm aware of, but it was a flaw in the JVM->Native mapping (specifically the JavaScript support). On a fully code-managed system, this is impossible.

Trust me, crackers would LOVE to use Java for malware. Unfortunately (for them), no one has yet managed to break the Security Manager.

The question is whether the language is the proper place to add such protection.

Whoa! Hold up there! The protection is not in the language. It's in the platform. The Java Language is independent from its platform, and provides very little in the way of security features. However, the Platform is as secure as it gets, no matter *what* language you use in it. Python, Ruby, BeanShell, JavaScript and many other languages have been made to work on the Java Platform.

Remember, your OS/CPU combination are one type of platform. Java is higher level platform that solves many of the issues with previous platforms.

Granted, you need an OS that does its job properly, but then a buggy java runtime would have the same problems as a buggy OS.

It's far easier to prove the correct execution of Java Bytecode than it is to prove the security of today's OSes. In fact, even the most secure OSes (e.g. OpenBSD) have been shown to have root exploits. You can't do that in Java. You just can't. There's no ledge on which you can grab a purchase. The best you can hope for is something like the TENEX flaw which allowed programs to hook into the paging notifier to check if something was paged from disk. By aligning the password characters with the end of the page and swapping the next page out to disk, an attacker was able to know that a password character was correct based on if a page fault happened.

Of course, security has moved on quite a bit from there, so I seriously doubt such flaws would be present.

I think there's a lot to recommend the idea of letting the OS handle security

If Java is the OS, it WOULD be handling security.

One of the BSDs, OpenBSD I think, demonstrates how well this can work. Currently it has zero outstanding security advisories and a policy of full disclosure.

You can't get much better than that.

Yes, yes you can. You can have an OS *never* have a root exploit, or even a critical exploit. Java can do that. Think, with all the J2EE servers running out there, and all the webbrowsers with Java installed, how many have experienced major flaws in the Java architecture or VM? The answer is a resounding ONE. (The one I described above.) Even programming flaws in J2EE systems fail to lead to system security issues like gaining root access. Usually, it's a matter of allowing web clients access to data they shouldn't have. (That's a whole other problem unrelated to system design.)

And C's parameter passing mechanism is secure so long as the programmer always checks buffer lengths.

And Java's is secure even if the programmer DOESN'T check buffer lengths. "ArrayOut

Re:Unix is not the Future

[AKAImBatman]

If you're running IIS as the internet account instead of as local system, then you can get the same effect just by correctly administrating your system

No, you wouldn't. The grandparent stated that there was a flaw in IIS where the URL was decoded a second time and used without security checking. That broke IIS's security. Java's Security Manager sits outside the program's security and checks on EVERY file access in the system. In an OS setting, the server could come pre-packaged with Security Manager rules saying that files outside directory WEBAPPS cannot be accessed. If such a bug exists in this mythical application, IT WOULD BE IMPOSSIBLE TO EXPLOIT!

Java's security model doesn't help you at all here.

Yes, it does. It helps a LOT. :-)

Re:Unix is not the Future

[DA-MAN]

Does that clarify the idea? :-)

Yes. So you want to get rid of all the redundancies between different components. However when your "server" has a bug, it's not just your computer. It's your TV, your Stereo, your PS4, and your computer.

In addition this sounds like a jack of all trades and a master of none, trying to get all that functionality into one device will make it convoluted and hard to use.

Oh oh, and you want to clog up the little free wireless spectrum we have left. All around this thing you describe has many fundamental flaws.

Maybe one day those flaws will be addressed, but you had stated "near future" in your post. To me that means within 10 years MAX, and I don't see that kind of convergance happening within that time frame.

Then again, I never thought I'd own a Mac either so I have been wrong before ;)

Re:Unix is not the Future

[spectecjr]

No, you wouldn't. The grandparent stated that there was a flaw in IIS where the URL was decoded a second time and used without security checking. That broke IIS's security. Java's Security Manager sits outside the program's security and checks on EVERY file access in the system. In an OS setting, the server could come pre-packaged with Security Manager rules saying that files outside directory WEBAPPS cannot be accessed. If such a bug exists in this mythical application, IT WOULD BE IMPOSSIBLE TO EXPLOIT! ... and setting security on the file system so that INETSRV_ANONYMOUS (or whatever the IIS account is) only had access to places it's meant to fixes the problem, without the need for a security manager. If you were talking about anything other than filesystem protection, you might possibly have a case - but in this case you don't.

Re:Unix is not the Future

[AKAImBatman]

However when your "server" has a bug, it's not just your computer. It's your TV, your Stereo, your PS4, and your computer.

That's a bug? You obviously haven't seen some of the setups these days. I just built a computer that is my CD Player, DVD Player, Television, Radio, and Computer all in one! And I've been considering purchasing a large LCD TV to replace my 17" LCD screen that I've been using. I hardly think that's a bug. :-)

BTW, the PS4 is still separate (for its graphics hardware and such), it just feeds off the server for networking and video streams.

In addition this sounds like a jack of all trades and a master of none, trying to get all that functionality into one device will make it convoluted and hard to use.

Have you SEEN the home entertainment systems that people have setup? You can't tell me that this would be more complicated. If anything, it should simplify things quite a bit.

Oh oh, and you want to clog up the little free wireless spectrum we have left.

How is that different than the little wireless spectrum we're already clogging? Many users already have wireless in their homes. This would simply replace it with a new hub. There's nothing stopping someone from turning off the Wireless and running cable instead, but that is such a PITA.

Maybe one day those flaws will be addressed, but you had stated "near future" in your post.

Read it again. I said that development of Unix would take it through the near future. I said the long-term future belongs to something else. :-)

To me that means within 10 years MAX

Pfff. We'll have the first offering of this sort of home entertainment in 5 years.

Re:Unix is not the Future

[AKAImBatman]

... and setting security on the file system so that INETSRV_ANONYMOUS (or whatever the IIS account is) only had access to places it's meant to fixes the problem, without the need for a security manager. If you were talking about anything other than filesystem protection, you might possibly have a case - but in this case you don't.

Except for one minor difference that you failed to note. In the Java Server I'm suggesting, the security settings are built into the shipped copy. There's NO CONFIGURATION to do! With IIS, you have to explicitly set the system permissions. That's always a good idea, but how many admins fail to take this basic precaution?

Re:Unix is not the Future

[Anonymous Coward]

To quote John Titor (interesting topic in & of itself, that name, look it up if you have not read up on it... some of it, SHOULD spook you some, because much of it CAN be debunked, but some? Cannot...)

ANYHOW:

"UNIX HAS A PROBLEM IN 2038" - John Titor

* :)

APK

Re:Unix is not the Future

[Luminary Crush]

Yeah, I predicted this too - about 5 years ago. I even tried to get enough interest to start a company to produce such a 'home server' product, which would interface with entertainment, security, networking and home management functions. No one was quite interested yet. That day will come, sooner or later.
Your home server will be like your refridgerator - something you buy, drop in, and turn on. And, rarely, call the Maytag repairman to come fix it when it has a problem.

Re:Unix is not the Future

[Master of Transhuman]

That's one of the reasons Linux will win on this: Microsoft and everybody else will be charging for Web services - and Linux won't(since everything comes WITH Linux!) Of course, services will be available on Linux as well and for a fee, but anyone not wanting to use them can opt for free stuff loaded right on their system.

Linux and open source always stands for choice, and I believe this will continue no matter which way the industry as a whole swings.

Eventually there will be balance between the Sun mantra - "The Network IS the computer" - (which I've never agreed with on simple grammatical grounds if nothing else) and the rich client (or multiple rich clients as PDAs get more powerful or more varied personal digital devices appear.)

Also, the alternative to bigger and bigger servers serving up services will be smaller and smaller and yet more powerful, easy to carry and embed, devices that may connect to those servers but can also do useful work independently.

I'm not worried about it going to far in any direction.

Re:Unix is not the Future

[NickFortune]

But what would the trojan do? Would it simply run a program just to crash it? That seems kind of pointless. The point isn't that threads don't die. It's that it's impossible for an attacker to use this in any meaningful way.

I was contrasting the current, typical use of Java as an application programming language against its hypothetical use in a Java VM as OS context. In a typical contempary scenario, finding a java 'sploit avails the cracker naught, since he then still has to crack the OS. Contrariwise, on a system where the OS is unified with the Java VM, there is no limit to the damage an exploit can do.

To summarise: one of the main reasons Java is so secure is that it has the system OS as a second line of defence. Unify OS and VM and that advantage is lost.

It HAS been tested in earnest. Applets are an example of an area where the Java security model is in effect.

On the one hand, applets are a very specific and specialised case. The malware artist not only ha t get past java, he also needs to defeat any browser security measures and there is still the OS to contend with.

On the other, current JVMs rely, as you yourself said, upon the OS for a lot of facilities. When those features are migrated into the VM, the complexity increases, and there's no guarantee that security is maintained.

Whoa! Hold up there! The protection is not in the language. It's in the platform. The Java Language is independent from its platform, and provides very little in the way of security features.

And yet I seem to recall you saying that "it is much easier to add this protection in Java than in any other langauge". Perhaps you meant to say that "the Java RTE has features that enable programs which compile into Java bytecode to better implement such protection".

We can use the term "Java" loosely (as we have been) or we can be precise. I don't mind which as long as we both follow the same rules.

If Java is the OS, it WOULD be handling security.

You mean "if the Java Run Time Engine is the OS". We're being precise, remember?

You can have an OS *never* have a root exploit, or even a critical exploit. Java can do that.

Never happen in the real world. That's my prediction.

Think, with all the J2EE servers running out there, and all the webbrowsers with Java installed, how many have experienced major flaws in the Java architecture or VM? The answer is a resounding ONE.

Yes. For a JVM insulated from the bare metal by at least two levels of abstraction, that's not bad going. It's still unproven as a full on, standalone operating system.

And Java's is secure even if the programmer DOESN'T check buffer lengths. "ArrayOutOfBoundsException: Element 100002 does not exist."

You mean the Java Runtime Environment, surely?

As I said above, we're not talking about the Java Language. We're talking about the Java Platform. You have to make a distinction between the two or you'll fall into the same trap you just did. :-)

Let he who is without sin cast the first stone, that's what I say :P

But I'll tell ya what: Let's conduct the remainder of this in terms of Perl 6 and Parrot. I'm a Perl fan, sh that'll factor out my admitted bias against Java, and since all you were originally arguing for was managed code, Parrot should do as well as the JVM. Also, I can write "Perl" to mean "Perl" and "Parrot" to mean "Parrot". Does that sound fair?

Re:Unix is not the Future

[DA-MAN]

Read it again. I said that development of Unix would take it through the near future. I said the long-term future belongs to something else. :-)

I misread this line:

That's my thoughts anyway. Sometime in the near future, I'll get them blogged down in detail. :-)

I read it as:

That's my thoughts anyway, sometime in the near future. I'll get them blogged down in detail. :-)

Punctuation makes a big difference. Not sure what was going through my mind actually.

Pfff. We'll have the first offering of this sort of home entertainment in 5 years.

Good luck getting all of those vendors to agree on standards. Sony will always push their own crazy shit.

Re:Unix is not the Future

[AKAImBatman]

If you don't mind my asking, how much experience do you have with Java Platform/Language? These concerns are silly. You can't crash a J2EE server with a remote exploit, and you can't crash a Java OS with a remote exploit. The language was designed from the ground up to prevent exactly those sorts of things. Once you write a program loader in Java using a classloader, it becomes apparent how impossible it is for a program to break out of its sandbox. :-)

Re:Unix is not the Future

[spectecjr]

Except for one minor difference that you failed to note. In the Java Server I'm suggesting, the security settings are built into the shipped copy. There's NO CONFIGURATION to do! With IIS, you have to explicitly set the system permissions. That's always a good idea, but how many admins fail to take this basic precaution? ... unless you're running Windows Server 2003, of course, in which case it's the default.

Re:Unix is not the Future

[16K Ram Pack]

I remember being in a company that switched from green screen to PC, and the most seductive part wasn't local working. It was drop downs. Users loved that you didn't have to know a product code, you could just look it up.

The question isn't about you doing 400MB photoshop work on a remote server (although one day, this may happen), it's more about someone running CRM or fault management on a web based project management tool instead of a desktop app.

Re:Unix is not the Future

[NickFortune]

Not a whole lot. A couple of uni projects. I've got a half written JEdit plugin lying aroound somewhere. I've not done any J2EE at all.

All the same, it is in the nature of security considerations that they are designed for today's challenges, but must inevitably meet those of tomorrow. There is no silver bullet.

I presume you keep framing this in terms of J2EE because that's part of your vision of the future? Everyone has a machine with a browser and all their files are on a server somewhere else and all their apps are applets?

Re:Unix is not the Future

[Anonymous Coward]

My house currently has 7 computers in different parts of the house used for different purposes.

Let me guess: one of them is a dedicated porn server.

Re:Unix is not the Future

[AKAImBatman]

Not a whole lot.

That's what I was afraid of. If you want to learn more about how secure Java is, I highly recommend trying to write a shell for launching Java programs. You'd need to become intimately familiar with classloaders and the rest of the java.lang package. Then you could write as many programs to attempt to crash the VM as you like. Good luck on that.

All the same, it is in the nature of security considerations that they are designed for today's challenges, but must inevitably meet those of tomorrow. There is no silver bullet.

There will always be security problems. For example, there is no solution to social engineering. However, we *can* solve the problems that cause nearly all of the remote exploits today. Java was built from the ground up with near-perfect security.

Most new programmers to Java assume that there must be a way out of the JVM, because there are always ways to break systems. Any programmer worth his salt soon learns that this just isn't the case. The Java Platform is as inpenetrable as modern technology can make it.

I presume you keep framing this in terms of J2EE because that's part of your vision of the future?

No, I frame in J2EE because it's the perfect example of where things *could* be broken. J2EE is a very large system with more components than many OSes. There are thousands of these servers deployed in the wild, yet remote exploits just don't exist against them. Any exploits have to target the specific holes in the company's codebase. (e.g. Some people are dumb enough to write servlets that allow access to arbitrary files.) Even these holes can be closed off if a Security Manager is used.

Java is used in Cell Phones, Web Browsers, High End Servers, and just about anything else that an attacker might want to penetrate. Yet day in and day out these systems withstand all attempts to be exploited by would be attackers. If that doesn't say something to you, then it should.

it might be some third thing...

[Anonymous Coward]

um os x?

Re:it might be some third thing...

[FLAGGR]

OSX is unix, buddy.

Re:it might be some third thing...

[Anonymous Coward]

Yeah right, ask the Open Group what they think about that. Is every mp3 player an iPod too?

Re:it might be some third thing...

[FLAGGR]

You reversed your anology, thats like saying is every unix OSX? The proper question is "is every iPod an mp3 player", to which the answer is "yes". It doesn't matter what the Open Group thinks, they don't get to decide what is unix and what isn't unix.

Re:it might be some third thing...

[Anonymous Coward]

Unix -> Trademark
iPod -> Trademark

OS X, unixlike but not a unix
iRiver, iPodlike but not an iPod

I didn't reverse shit.

Re:it might be some third thing...

[The Lynxpro]

"Yeah right, ask the Open Group what they think about that. Is every mp3 player an iPod too?"

Over 80% of them are here in the States. :)

Granted, I really don't like that argument if you substitute the word "MP3" with "personal computer" and "iPod" with "Microsoft Windows."

Re:it might be some third thing...

[Transcendent]

OSX stems from BSD (FreeBSD). BSD stems from UNIX. Therefore, OSX stems from UNIX.

When leo said "UNIX", it is just the environment and design of the operating system (I think that UNIX is the answer, in some form or fashion.). In that sense, OSX, BSD, Linux, Solaris... are all "unix".

And you did reverse shit: "Is every mp3 player an iPod too?"

Re:it might be some third thing...

[C0vardeAn0nim0]

according to ESR theres 2 kinds of Unix: the "genetic unixes" derived from the origninal AT&T code (wich include all BSD variants such as apples darwin) and the "trademark unixes", the ones not neccessarily derived from the AT&T or BSD code but that passed an opengroup certification.

macos x is a genetic unix, so the generical term "unix" applies. GNU/Linux is nor genetical nor trademark, so its usualy refered as a "unix like" OS.

see here: http://www.opensource.org/sco-vs-ibm.html#id308718 0

I agree

[MacFury]

MacOS X and operating systems that can marry the power of a good command line with the ease of an excellent GUI shall inherit the earth. I'm interested in how the new windows command line stacks up.

Re:I agree

[MrShaggy]

HMmmmmmmmmmm

AMIGA Anyone ??

>> MacOS X and operating systems that can marry the power of a good command line with the ease of an excellent GUI shall inherit the earth

Re:I agree

[mnemonic_]

I wonder why MS is working on a new command line at all. Do people buy Xserves so that they can use the OS X command line? Do people run linux because they love staring at those grey characters on a black screen? No one really likes the command line... plenty of people get by with it, but it's obviously the most primitive computer interface. So why is Microsoft developing it? Do they really believe that *NIX users like their OS because of the command line?

Re:I agree

[Anonymous Coward]

Every time you top post, God kills a geek.
Please, think of the geeks.

Re:I agree

[Virak]

Do people run linux because they love staring at those grey characters on a black screen? No one really likes the command line... plenty of people get by with it, but it's obviously the most primitive computer interface.

Yes, some people (myself included) actually do like the command line. And as it's one of the most primitive interfaces, it's much faster and more reliable than a GUI, uses less memory, and for many operations is many times faster than a GUI. Until we can control our computers by thought, the CLI is here to stay.

Re:I agree

[Anonymous Coward]

Or Syllable?

Re:I agree

[kesuki]

Do people buy Xserves so that they can use the OS X command line?

Yes, powerful Command lines are more than 'just' for end use, they open up the entire core functionality of the OS to non-interactive scripting. By having a powerful, flexible shell you can have powerful scripts that run fast, do everything you want, and can be quickly edited, they run as fast as compliled code, but since they're just a text file that gives comands to a precomplied binary you can modify them much more easily than a full fledged program.

System administrators need a powerful command line interface, and while standard 'unix' tools sometimes have areas that need improvment. for instance chroot on BSD require the setting of a shell variable to change shell, but linux chroot which accepts it on command line, but can't change the user or group(s) that you're chrooting them to. That means you can't create a chroot jail to disable (remote) root access on linux (that allows remote logins)... but you can on FreeBSD/MacOSX

Re:I agree

[Daniel Dvorkin]

I wonder why MS is working on a new command line at all. Do people buy Xserves so that they can use the OS X command line?

They buy Xserves so they have a choice -- use the nifty OS X Server GUI admin tools (which are really good, I have to say) if they fit the task, and use the command line if that fits the task. Choice is a Good Thing.

Do people run linux because they love staring at those grey characters on a black screen?

Very often, yes; (usually multicolored, these days) characters on a black (or whatever) screen may seem primitive to you, but to many people they represent an extraordinarily efficient way to get things done.

No one really likes the command line...

*falls over laughing*

plenty of people get by with it, but it's obviously the most primitive computer interface.

No, manually unplugging and plugging in vacuum tubes is the most primitive computer interface. It may not be obvious to you -- or to Neal Stephenson, for that matter -- but today's Unix shells represent an extraordinary level of abstraction from the underlying bare metal.

So why is Microsoft developing it? Do they really believe that *NIX users like their OS because of the command line?

In a word: yes.

Look, not everything is best done on the command line. GUI's are wonderful things, if they're done right. (Which pretty puts any flavor of Windows out of the running, but that's a whole 'nother argument.) But as I said above, they are not the right tool for every task. For power users, especially admins and developers, the command line is very often a better tool. And the best of both worlds, as in Apple's current OS, which Microsoft is again trying (and no doubt failing) to emulate, is being able to switch seamlessly between them as the task at hand demands.

Re:I agree

[irc.goatse.cx troll]

What most of us CLI users dislike is graphical input. A lot of us don't mind graphical displays, as a lot of time they are better, but theres nothing better to having to find which of the 1,920,000 pixels my cursor currently occupies so i can move it over to click on something. The best input is the right amount of key bindings, with a command mode like vim (eg, :make).

What I personally would have switched to had it ever been feasable is xmlterm. XMLTerm was a mozilla project to create an xterm clone that can drop into mozilla's renderer so that you could throw html/xul/images into it, so that for example your cli script could return a pretty table, or a full color coded logfile, or a dir index with thumbnails. This way you'd still have the ability to do a quick for i in *.jpg;do convert $i -scale 800x600 ${i}.resized.jpg;done or some mass regexps or whatever, but while being able to graphicly see whats going on.

xmlterm.com doesnt mention that its dead, but theres been no updates for over a year. R.I.P xmlterm

Re:I agree

[qcomp]

No one really likes the command line... plenty of people get by with it, but it's obviously the most primitive computer interface.

i disagree; i do like the command line - for many tasks it is the most advanced and suitable interface. Everytime I'm forced to use a Windows machine, I notice how much I miss a decent CLI and bash.

Re:I agree

[kesuki]

i meant a 'secure' chroot jail. sorry
you can make an 'insecure chroot jail' on linux that is vulnerable to buffer overflow bugs in the os what not... because the chroot jail still leaves you as root, even if you have no access to a shell or a directory tree.. you're connected to the machine via a connection protocol, that may have a remote vulnerability in it as root access. if chroot can switch you to user none, or guest or something else locked down even if they exploit the jail, they still end up as a user who doesn't have permission to read/write anything, so ot doesn't matter if the code they exploited you with is a small installer that would download a backdoor to your system, because it doesn't have permisssion to do any of that. It's called Security through Security.

Re:I agree

[sdpuppy]

Sure, some like command lines, others like iconic interfaces.

rm -rf /*

ooops....

Re:I agree

[richlv]

commitment to commandline on a workstation might be looked at as something weird - but for servers it's a lot better.
servers just have to sit there and do their work, so :

1. cli takes less resources;
2. cli is a lot faster over slow connections (like resolving a problem from a deep countrside over crappy phone line);
3. it allows more flexibility - like mass converting user mailboxes, converting a bunch of images and a lot of other jobs.

for some more information-intensive tasks there can be gui applications that are run on a workstation but connect to some process at server - but what's the reason to have a gui on a server if you anyway tackle it only twice a year ?

Re:I agree

[AddressException]

What's that asterisk doing there?
Shows what you know! ;)

Re:I agree

[cp5i6]

Barring the trolls I'm probably going to get from this, I 'd like to say that Choice is a Good Thing


Mac OSX is a very slow OS.. because of it's true monolithic approach it's no where near as fast as windows will be.

As for power users/admins/developers ... I'd really doubt you would use Apple's OS to control anyhting else but a mac .. because we all know that macs are really just sitting in every server room through out this world.

For developers ... as much as I like using textpad. tools like eclipse even netbeans are just so much more efficent (if even just for the library look ups) that there is no reason NOT to use a gui.

regardless there are benefits of the command line and yuor examples really suck and you need to learn how to use windows properly.

Re:I agree

[PWatson]

No, manually unplugging and plugging in vacuum tubes is the most primitive computer interface. It may not be obvious to you -- or to Neal Stephenson, for that matter -- but today's Unix shells represent an extraordinary level of abstraction from the underlying bare metal.

I'm not sure how Neal Stephenson got into this, I can't find a reference to him above. Anyway, Stephenson did write about the Metaverse in Snow Crash, but that doesn't mean he doesn't appreciate the command line. In fact, he wrote In the Beginning... Was the Command Line. It's been a while since I read it, but I remember it being quite pro-CLI.

Re:I agree

[SolarCanine]

It may not be obvious to you -- or to Neal Stephenson, for that matter -- but today's Unix shells represent an extraordinary level of abstraction from the underlying bare metal.

Actually, in Snowcrash, Stephenson specifically notes that when a hacker (in the classic sense) wants to get something done, he does it not in a virtual environment or a GUI, but in a command line.

Re:I agree

[AKAImBatman]

You know, I always wanted to try XMLTerm, but I could never get it working. When it was active, I didn't have a Linux machine available. (I did have Solaris and FreeBSD, but neither one was "Good Enough".) By the time I got to a Linux machine that would work, XMLTerm was too out of date to use on a modern Mozilla client.

Man was I peeved. :-/

Re:I agree

[sootman]

Assuming you're not trolling, I'll answer. Command lines are good for lots of things. Here's one set of reasons to like the CLI. Anything you can do at a command line...

• can be done remotely over even the slowest network link.
• can be put into a script...
• ...which can be scheduled with CRON
• produces textual output, which can be
  • instantly sent to a printer (hack-proof!--hard to delete logs when they're already printed*)
  • emailed
  • shown on a web page

Without a command line's texty goodness, how could I do something like this?

* without physical access, of course.

Re:I agree

[temojen]

su (nonprivileged user)
chroot (newroot) (server)

Each tool does one thing, and does it well.

Also, chroot jails as non-root still leave you vulnerable to a combination of privilege escalation and chroot breaking vulnerabilities, no matter what OS you're on.

Re:I agree

[Daniel Dvorkin]

My point was that In the Beginning ... Was the Command Line presents a semi-mystical view of the command line as The One True Way Of Interacting With The Computer, whereas in fact it's just another type of abstraction, and a pretty sophisticated one at that. IIRC, Stephenson acknowledges this at one point in the book, but then ignores it thereafter. The command line isn't the bare metal; it's not even close.

In the beginning ... was the vacuum tube. These days, of course, logic gates are much too small to move around by hand, but the point remains.

Re:I agree

[Anonymous Coward]

Vacuum Tubes??

In my day, we had relays, and we liked it. Except de-bugging with dental floss. Yuck.

Re:I agree

[TheRaven64]

Note, however, that these properties are not intrinsic to the command line. OS X through AppleScript and GNUstep through Steptalk expose the model objects in the model-controler-view pattern to external scripting, allowing them to be directly manipulated. Done well, this is far more flexible than a command line, since it is much easier for objects exposed to a dynamic language like Smalltalk to negotiate a connection with each other and interact. You could do the same thing as your example if system monitoring objects were exposed to your scripting language - easily, if your scripting language supported remote access. Actually, you could easily display graphs of the output (for example) if they were exposed to a proper scripting model, rather than a command line like the UNIX one.

Re:I agree

[kesuki]

see the problem is, su just needs you to type 'exit' to revert you to the shell and login you had before, unless you have a different version than me, hense the lack of security, when chroot does it it terminates the previous shell, leaving you no easy route to privaledge escaltion. of course buffer overflows in the kernel are always gonna nail you...
that's why HURD has the microkernel architecture, why linux uses loadble modules as much as possible..

if you can disable/remove the portion of the kernel that has the vulnerability without rebooting it's vastly superior to a system where you have to 'reboot the server' to apply a kernel security patch, if one even exists..

Re:I agree

[temojen]

notice where I said

chroot (newroot) (server)

There is no shell to type exit in.

Re:I agree

[kesuki]

but then you just need a buffer overflow that allows you execute code, which can kill the process of su...which is why i said linux is less secure.. if all you need to do is kill a single process to get back to root... thats a hell of a lot easier than permisions escelations etc etc..

Re:I agree

[temojen]

killing the shell created by su does not give it's child processes root priviliges.

Apple

[avalys]

I'm rooting for OS X, personally.

It's funny, because I absolutely hated Mac OS = 9.

Re:Apple

[Lost Found]

OS X winning would be a huge step back for humanity. Mach is terrible and performs like ass.

Re:Apple

[AddressException]

Always nice to read such well-reasoned arguments on Slashdot.

Re:Apple

[rubycodez]

which is a BSD. which is really a Unix, though the people who made Unix into Unix(tm) won't let us say that

Re:Apple

[Lost Found]

Oh yes, a microkernel that up until the very last release of the OS had the so called "funnel locking model" of having one big kernel lock, then one smaller one for a corner of the functionality is good.

Their increases in lock granularity in Tiger simply moved the locks to subsystems. "Oh, I've got IO now... it's MINEEE"

In fact, I remember Anandtech (I think it was Anand) benchmarking it against Linux and found Linux to be *substantially* faster at anything that invoked the kernel.

Re:Apple

[jpc]

OS X could be ported to another Unix. There are rumours that Apple code is being made more portable (well, things like iTunes need to run on Windows too). NetBSD has an almost finished Mach emulation layer; when OS X is mainly an Intel platform other Unixes will probably do this too.

Apple are also making the non-Mach parts entirely FreeBSD compatible, rather than the slightly partial compatibility there is now.

OS X on Solaris might be an option but probably too much work. Migrating to FreeBSD is more likely.

Most apps dont address the Mach layer directly anyway, only the core system libraries.

Re:Apple

[Anonymous Coward]

Damn, I accidently modded you down. This post is only to remove my mistake.

Re:Apple

[sickofthisshit]

The Anandtech benchmarks were useless. They benchmarked applications tuned for Linux on Linux vs. applications tuned for Linux on Mac OS X.

Ignoring the Apple-provided versions of the same applications tuned for OS X.

Re:Apple

[The Lynxpro]

"I'm rooting for OS X, personally. It's funny, because I absolutely hated Mac OS = 9."

I find it neither funny nor odd. Back in the 80s, I couldn't stand the Mac platform - and its inflated cost - in comparison to the ST (and the Amiga). When both of those platforms ceased, I gave in and switched to the VHS of the industry, Windows, because it looked like Apple was nearing its end (back in 95/96). However, since OS X 10.2, I've really been rooting for OS X and Mac hardware in general.

Who knows? If Longhorn - excuse me, Vista - fails and Apple sells millions of Intel-based Macs in 2006, perhaps Dell, HP, Gateway and Sony will be forced to devote a large amount of monies to getting Linux up (or down, depending upon your perspective) to "average computer user" abilities and then the industry will once again pick up and innovation will return.

Re:Apple

[Lost Found]

Fine, I don't remember the article well enough to confirm or deny, so point conceded. But the differences in lock granularity are still *substantial*. Linux locks data as a rule (granted it wasn't always this way), and OS X, even with the "great advances" in Tiger locks subsystems. Hence, lock contention is higher. Spinning pinwheel of death.

Well written?

[Anonymous Coward]

"In a well-written interview with Mad Penguin..."

"'But UNIX is such a well understood and smart to handle the issues that an operating system has to handle that it ultimately will prevail.'"

Yep, seems pretty well-written to me ;-)

Re:Well written?

[imroy]

"Whether you're writing an open source browser or you're righting a symphony, I don't think there's that much difference."

I've also seen a few other mistakes, and I'm only part-way through the first page. Well written? nope.

Re:Well written?

[Enrico Pulatzo]

Wow, that was close. I would've told you to duck, but then I realized it went right over your head anyway...

Re:Well written?

[Anonymous Coward]

Hey at least they don't have any Microsoft ads. Oh wait...

Re:Well written?

[MrHanky]

Yuo must be new here. On this site, "well written" means well intended. If you can guess what it means, and it means well, it's good enough for us.

Linux is still growing

[ylikone]

I don't see why Linux would not be THE operating system of the future for server as well as desktop. It is still growing and getting vastly better every year. As long there is an involved, interested and active open-source community, I think Linux will be around for a long time to come... unless something like HURD ever properly materializes.

Re:Linux is still growing

[Lost Found]

HURD still has pipe dream status. Yeah, you can download it and use it, but given the rate of development of Linux, plus the investment by all of the large firms, plus the rising deployment rate, no one has the momentum to keep up.

Linux gains new ground every day, and it seems particularly hard to lose it. I'd say that the only threat Linux faces that is "serious" is te fact that it is the only threat Microsoft faces that is "serious" (Microsoft has said as much themselves). Expect Microsoft's moves to restrain against Linux to become legal ones, from lawsuits to "campaign contributions".

Re:Linux is still growing

[ciroknight]

But until Linux's GUI developers get a contract with Video Card manufacturers to produce better device drivers, or until GUI developers get their act together and make the GUI's faster and more user friendly, I'm afraid to burst your bubble.

Mac OS X is in a position no operating system has been in for 10 years. In 1995 when Microsoft brought out Windows 95, the operating system shattered the market because it was faster, prettier, and just plain cool. Now, OS X is in the same position. And they're going to have to screw up just as bad as Microsoft did with Windows to lose this creative lead.

On top of this, OS X is based on Unix, meaning that it's going to stay secure for a long, long time. As time goes by, bugs will be found and squashed, as no software is perfect, but UNIX by design has less issues with bugs, and the bugs are harder to exploit.

Linux has promise, but it's being held back by developers that simply don't care, because they aren't paid to care. They're doing their own, individual thing, and not working towards what a User wants, they're working towards what a developer wants. And if they took a minute to have an objective look at things, they're not too different. Mac OS X mastered this with XCode. And the only good IDE I've found for Linux is KDevelop, which requires me to install another set of GUI libraries, just to use it.

It's really past time to have Linux desktop-ready. It's time to replace X with something that renders faster, or to simply get Cairo and the other eyecandy, GL-rendering, bad-ass GUI systems up and running. You guys are five years behind Mac OS X, and about a year or so behind Windows in this department.

Re:Linux is still growing

[ucahg]

Yes, but Linux effectively is UNIX, if not technically. It conforms to the POSIX standards for the most part, and it is these standards, and these UNIX philosophys (that Linux developers also adhere to) that will live on. Be that through the actual Linux kernel or not is really a side-issue.

At least that's how I interpreted the original article.

Re:Linux is still growing

[ylikone]

One extremely important point to consider that everyone seems to ignore in the Linux vs. Mac debates... Linux is open source, Mac is not and never will be.

Re:Linux is still growing

[ciroknight]

Why does this matter at all in the debate of who's better? While Linux has the ability to improve faster (more eyes on the code, yadda yadda), it has repeatedly failed to live up to the promise. Meanwhile, a few good coders who are getting paid to make their operating system better and to innovate at any costs are creating an operating system that's better and growing at a rate faster than Linux.

The fact that Linux is open source is it's Achilles heel and its most shining gem. Because it's open, not enough people are doing the nessicary work to get the operating system useable and competitive. But because it's open, the potential for anyone who wants to, is there.

In my eyes, OS X is open enough, having all of Darwin, most of Safari, and a bunch of other well documented APIs. Meanwhile it's competitors are all scrambling to try their best to beat it, but they simply can't match the momentum that's currently pushing Apple up the hill of beans Microsoft has accrewed over the past decade.

Re:Linux is still growing

[*SECADM]

Isn't the NT kernel (which win2k, winXP and now vista are based on) also POSIX compatible?

I guess it's a pretty safe prediction to make, when all the major OSes in the market (windoze/OSX/Linux/*BSD) are POSIX compatible.

I am still waiting for L4 to take off any day now.

Re:Linux is still growing

[b1t r0t]

Linux is open source, Mac is not and never will be.

The people who care about this are about 1% of the market, if that much. The fact is that OS X is open enough source. The foundations are open source. The graphics library (Quartz) and UI library (Aqua) are not. The apps are generally not. If being 100% open was so important, Linux would have taken over years ago.

If you care so much about having something like the OS X graphics and UI that is open source, then get involved with the OpenStep project.

Re:Linux is still growing

[krmt]

In my eyes, OS X is open enough, having all of Darwin, most of Safari, and a bunch of other well documented APIs. Meanwhile it's competitors are all scrambling to try their best to beat it, but they simply can't match the momentum that's currently pushing Apple up the hill of beans Microsoft has accrewed over the past decade.

One day Apple will change.

One day, Steve Jobs won't be around any more and someone else without even a touch of his abilities will come in to run the company.

One day, Apple may well revert to the old Apple of the mid and late 90's.

Then what will you do with your "open enough" operating system?

Re:Linux is still growing

[ciroknight]

That looks like blue smoke to me. Sure, one day Steve will retire, and I'm 100% sure he won't leave without taking care of business before he leaves, and putting someone in charge to do the business the way he wants things done. Look at Microsoft for an example of that one.

Good job on the dramatics though.. "I'll get you yet Captain Planet!!"

Re:Linux is still growing

[delire]

Kind of. While I personally don't find OSX(Tiger) to be the high-performing, uber-intuitive, victoriously user-friendly OS people keep telling me it is, it's impossible to ignore Apple have created an incredibly successful, reasonably brand-loyal consumer base.

And I agree, they are in an advantageous position, however one stricken by a crippling glass ceiling.

OSX simply isn't the kind of platform nomadic OS it needs to be to reach ubiquity. Linux, being non-proprietary, has developed a talent for adaptability innately - arguably a the bulk of Linux kernel development itself is invested in sustaining this critical attribute. While some complain about KDE and Gnome not having the bells-and-whistles of OSX, Linux (if ever one can sum it's parts) has quietly been frying more important fish, and very wisely too.

Nowadays, it's increasingly hard to move data around a network without it being touched by Linux. Now phones, kiosks, ATM's and game consoles.. The word 'Linux', not being publically obliged to carry a brand, increasingly represents the set of robust and future-proof tools a given service provider deploys to get the job done. Now we're seeing it extend to public utility, government and administrative desktops en masse.

'Linux' simply doesn't appear on the kinds of brand-conscious radars we're used to using when tracking technology and culture, yet quietly continues to be the fastest growing operating system in the world. Apple however, is a company built on brand innovation, initially as the 'popular minority'; a vain, techo-arian multinational that has boldy engineered a bridge between the computer (as appliance) and the public space of popular culture. Here, the Apple brand is both a boon and an anchor, something to protect (not just any x86 machine), yet something to wear out (iPod).

In the context of popular operating systems, OSX is thus a social, yet exotic animal. As such it is desired as pet, but quickly dies when transported from it's natural habitat. Comparing OSX to others, outside the context of it's crippling architectural dependence, produces false and unrealistic prospects.

Re:Linux is still growing

[SmittyTheBold]

one day Steve will retire, and I'm 100% sure he won't leave without ... putting someone in charge to do the business the way he wants things done.

Like he did the first time, with Sculley?

Still, I agree with you in that I think it's ridiculous to judge Apple on what a future CEO might do. If we're going to take that train of thought, let's just assume that Microsoft will become entirely non-evil sometime next March as well.

Re:Linux is still growing

[sloanster]

But until Linux's GUI developers get a contract with Video Card manufacturers to produce better device drivers, or until GUI developers get their act together and make the GUI's faster and more user friendly, I'm afraid to burst your bubble.

There may be some validity to this, however, it is very very easy to find video cards with great linux support and drivers - just buy nvidia. As for fast, user-friendly linux GUIs, where have you been, rip van winkle?

On top of this, OS X is based on Unix, meaning that it's going to stay secure for a long, long time. As time goes by, bugs will be found and squashed, as no software is perfect, but UNIX by design has less issues with bugs, and the bugs are harder to exploit.

Agreed, OSX foundation is solid, but so is Linux, for the same reasons.

Linux has promise, but it's being held back by developers that simply don't care, because they aren't paid to care. They're doing their own, individual thing, and not working towards what a User wants, they're working towards what a developer wants.

No idea where that bizzarre idea came from - the continuous incremental improvement in the linux GUI environment over the past 5 years, the user-friendliness, the eye candy, the easy customizability, all attest to a great deal of care.

And the only good IDE I've found for Linux is KDevelop, which requires me to install another set of GUI libraries, just to use it.

Hmm, kdevelop is certainly a good ide, but not the only good one, not by a long shot. I't's unclear what you mean by "requires me to install another set of GUI libraries, just to use it". OK, let me get this straight: You want to install a qt-based GUI, for creating qt-based apps, and you complain that you must have the qt libraries installed in order to use it? It's unclear exactly what your complaint is here. In any event, my installation of kdevelop consisted of checking the box next to "kdevelop", during the OS package selection phase of the linux install. That took what, a second or so? How could that be made easier?

It's really past time to have Linux desktop-ready. It's time to replace X with something that renders faster, or to simply get Cairo and the other eyecandy, GL-rendering, bad-ass GUI systems up and running. You guys are five years behind Mac OS X, and about a year or so behind Windows in this department.

FUD, I'm afraid. I use linux primarily (suse 9.3 ATM) but I also spend a lot of time on my wife's OSX box, and random expee and w2k boxes in the course of my daily work. I'm sorry, but I just can't see any reason for all the microsoft hype, I really can't. Where are the microsoft advantages you speak of?

OTOH, OSX is nice, but IMHO the interface preference is more a matter of taste. I like OSX better than any microsoft desktop I've seen, and the unix underpinnings are a huge win over the microsoft pc OS internals, but I actually prefer my suse linux desktop to OSX.

Ultimately, apple will capture some market share from microsoft, but the idea that OSX will kill linux is just silly.

Re:Linux is still growing

[TheRaven64]

Comparing OSX to others, outside the context of it's crippling architectural dependence, produces false and unrealistic prospects.

I don't know quite what this is supposed to mean. OS X is directly based on OpenStep, which ran on a variety of hardware architectures. It has been shown that OS X is portable between x86 and PowerPC, and it is almost certainly portable to other architectures without much of a problem.

If you mean it is limited to using the Mach+BSD+IOKit XNU kernel, then this is probably not true either. The NetBSD Mach/Darwin compatibility layer can run Darwin programs including XDarwin. XNU is an interesting kernel - it trades performance for a few extra layers of abstraction which make developing it easier. If it had been started from scratch now, it would probably more closely resemble L4 HURD, but at the time it sacrificed many of the benefits of a microkernel in order to run at a reasonable speed on a 25MHz 68040.

Re:Linux is still growing

[delire]

And I'm sure Aqua can run on Linux.

Of course OSX is portable to many target architectures, however it just isn't being ported. This is under duress of of Mr. Jobs, whose tactically retentive business agendas are exactly what I'm talking about above.

Five years to x86. Linux is only three times that in age.

This will always be the case for software one doesn't own (the right to modify and pass on), but rents, like OSX and Windows.

Re:Linux is still growing

[bored]

Ah, how the mac people forget rhapsody
http://toastytech.com/guis/rhap.html

Which I know runs on x86 because I have a copy.

Re:Linux is still growing

[alucinor]

The Linux desktop isn't going to take off in the first world. No chance. Now, developing countries ... I believe that's where Linux will make a dent. Then, after a generation of new coders in these countries grow up with it, Linux will storm on to the desktop scene. Until then, just watch it take over the server and smartphone markets instead.

Re:Linux is still growing

[The OPTiCIAN]

I don't support these eye-candy arguments. Aqua is one of the things I dislike most about my mac, and the finder is another strong contender. Integration is important and responsiveness, but eye-candy - well - I think it's counterproductive. I prefer gnome to any other desktop UI. If I could swap aqua for gnome tomorrow and still have the great java integration and application support that I have on the mac I would.

Ho ho ho

[gowen]

UNIX is such a well understood and smart to handle the issues that an operating system has to handle that it ultimately will prevail

That's right because, as we all know, the solution that is technically the best will always win out in the marketplace...

Re:Ho ho ho

[http]

the solution that is technically the best will always win out in the marketplace...

You must be referring to the use of proper grammar in /. articles.
Zonk, did you (or a family member) have surgery today or something?

Re:Ho ho ho

[g2devi]

> That's right because, as we all know, the solution
> that is technically the best will always win out
> in the marketplace...

Actually, that *is* the case most of the time. You just need to increase your time horizon.

Take for instance the microkernel versus macrokernel debate. Linux succeeded because it was a macrokernel but increasingly, it's becoming a microkernel. More and more stuff is being shoved into user space. Xen is only the latested push into achieving this.

The same thing can be said about Microsoft. Increasingly, they're borrowing the Unix ideas. They're calling things different and doing things differently, but the technology is drifting in the direction.

It's not

[Arthur B.]

Playstation, XBoxes, Mobile Phones, DVDplayers type of operating system are the future. The OS has been developped far ahead of most people abilities. The future is going towards less and less user control on this OS. Quite the opposite of UNIX.

Re:It's not

[FLAGGR]

OSX is unix. 'nuff said.

Re:It's not

[mrogers]

Which OS family has been ported to three out of four of those platforms?

Re:It's not

[Anonymous Coward]

wow...you are an idiot. nuff said.

Re:It's not

[poot_rootbeer]

Playstation, XBoxes, Mobile Phones, DVDplayers type of operating system are the future.

And all of these devices can and do run embedded versions of Linux (though against the manufacturers' wishes in the case of the game consoles).

One of the nicest things about UNIX-style operating systems is how well the design can scale up and down. If you want to run a UNIX in single-user mode on a 10-year-old IBM PC/AT, you can. If you want to run a UNIX on a cutting-edge minicomputer with thousands of users at thousands of terminals located throughout the world, you can.

Re:It's not

[mikefe]

Which OS family has been ported to three out of four of those platforms?

Are you trying to refer to NetBSD or Linux?

Control

[Anonymous Coward]

PC as a thin client browser?

I don't know about you, but that doesn't satisfy me and I think there will always be room for people who want a traditional desktop.

As a gamer and just fan of controlling the computer in front of me completely without all this abstractness, I don't think that everyone is going to bite on this kind of stuff.

I'm sure it has its place, but for everyone?

Re:Control

[guaigean]

But we here at /. are not average users. The masses, however ignorant or educated, are what determine the future of goods, and as we all know, the masses are easily decieved.

some third thing?

[intmainvoid]

It might be BSD, it might be Linux, it might be some third thing.

Talk about ignoring the elephant in the lounge room.

Re:some third thing?

[cloudmaster]

Umm, OS X is (Free)BSD with a few relatively minor changes. Those changes don't change that it's still BSD.

Re:some third thing?

[Otter]

Umm, OS X is (Free)BSD with a few relatively minor changes.

Yeah, some minor changes like the display system, the libraries and APIs, the utilities and pretty much anything else with which a user interacts... ls and cat are the same, though.

In any case, this is an interview with a Linux site. Laporte is just being polite.

Re:some third thing?

[Anonymous Coward]

Category: Before and After

Apple's New OS and Long-time Unix GUI and OS

Mac OS X Windows running on BSD

Re:some third thing?

[cloudmaster]

I guess you didn't read the differences, and you've never used OS X (or BSD).

Ok, so changing the display system changes the OS. Are you saying that my Linux system running X.org is different from one running Xfree? Or that mine running Windowmaker is not Linux, while one running twm is? No, those are all Linux, or all BSD, and the article wasn't about the user interface.

The utilities are all BSD, except for the *additional* utilities for OS X-specific stuff (like the things for netinfo, the disk imager, etc). The libraries are all the same, except for a few changes (can count them on one hand), except for the things they added for the GUI. The API is the same, except for the extra GUI stuff.

OS X is not a new OS. It's a new GUI on top of BSD, with a couple of minor changes from the specific BSD they chose as their OS. It may be a different user experience, but it's certainly not a new OS.

Re:some third thing?

[Anonymous Coward]

That's nice, but it's not really UNIX. It may run some BSD services, but it's really XNU/Mach. NOT UNIX.

Re:some third thing?

[jessicavampirehunter]

I'm certainly not a BSD girl, so take anything I have to say about them with a grain of salt, but it seems to me that while a lot of those changes are minor, some are definitely not. Like the security system - there are similarities because they're both Unix, but beyond that the differences are significant. Furthermore, the BSD's don't use Mach. Mach was based on their kernels, but they haven't been the same thing for 20 years. That alone makes it not a BSD - we call Red Hat, Gentoo, and Debian Linuxes because they share a kernel, and they differ because of platform-specific applications like Portage, YaST, and APT.

In short, OSX and BSDs are all Unix, but they're not the same thing.

Re:some third thing?

[MrHanky]

It uses a bunch of the same code, but the Darwin kernel is a different architecture (XNU/Mach), it uses Mach-O instead of ELF binaries, the init system is different, and the main system administration utilities are different (no niutil for every other task on FreeBSD). If you're used to FreeBSD, and boot pure Darwin with no GUI, you're not going to feel at home.

Re:some third thing?

[jonabbey]

Leo was speaking specifically to open source implementations of the Unix tradition. Mac OS X is significantly open-sourcey compared to Windows, but as long as it is limited to running on hardware with an Apple label, it will never truly be in a position to challenge Microsoft.

Linux and FreeBSD run on Dell, HP, Lenovo, Intel, AMD, Joe's White Boxes R Us.. that's what makes it threatening to Microsoft.

Volume, volume, volume. On the hardware side as well as on the software side.

Re:some third thing?

[larkost]

Hmmm.... lets see here...

user information stored in NetInfo

inetd not process 1 (and not present in many configurations)

no ports or packages system (other equivalents though)

display model (the primary one) not X-11 at all.

Mach kernel...

the framework system

90% of the API completely different (AppKit, CoreFoundation, Foundation, Core*, *Kit)

driver model completely different

It is obvious that you have never actually admined or developed on either FreeBSD or MacOS X.

Re:some third thing?

[Heretik]

Linux and FreeBSD run on Dell, HP, Lenovo, Intel, AMD, Joe's White Boxes R Us.. that's what makes it threatening to Microsoft.

Not to mention Macs, and Sun, and MIPS, and PDAs, and wristwatches, and....

Re:some third thing?

[cloudmaster]

Yeah, I should take OS X and NeXTStep off of my resume, because some poster on Slashdot - who thinks that a display system equals an operating system and thinks that the internet super server is the same thing as init - wants to argue about semantics. OS X is just another *nix with a pretty GUI on top - it's not the revolution fanatics want to believe.

BTW, NetInfo is going away, and inetd isn't process 1 on anything common. Normally, init is process ID 1, on BSD and OS X (and Linux). Here's the OS X machine sitting on the desk behind me:

cloudmaster@yyy ~ $ ssh zzz
cloudmaster@zzz's password:
Welcome to Darwin!
zzz:~ cloudmaster$ ps 1
  PID TT STAT TIME COMMAND
    1 ?? Ss 0:02.04 /sbin/init

It's the same on all of the 10.2, 10.3, and 10.4 machines, and on the FreeBSD box in my basement.

Maybe one day soon I'll tar up Linux with QTopia as the UI and use LDAP to store user information, and remove RPM from my distribution. Maybe I'll add a couple of printks to the Linux kernel, and a hook to activate them (I'll call it the "spice" kernel). Then I'll inlcude the libraries to support QTopia and LDAP, and call it cloudnix. Then I can just sit back and wait for some people on /. to get into a discussion about how cloudnix isn't Linux.

Re:some third thing?

[mikefe]

. . . it seems to me that while a lot of those changes are minor, some are definitely not. Like the security system - there are similarities because they're both Unix, but beyond that the differences are significant. Furthermore, the BSD's don't use Mach.

In the case of OSX, the Mach API is not exposed to userspace. The security system is standard BSD so I think you are adding a little too much congecture based on "it's based on Mach".

Arghh

[realmolo]

Unix is fine. It works.

But....it's 40 years old! Wouldn't we all like to see a completely MODERN operating system? I know I would. Keep all the good stuff from Unix, update it, and throw out the bad stuff.

Of course, in the end, we'll still be stuck with Windows and MacOS and Linux because they're the only 3 that have developer support.

Re:Arghh

[Lumpy]

you mean like BeOS or Plan9?

yeah those modern OS's sure did take off.

it's not about modern, it's about who can market the hell out of what they are selling, and how fast can you sucker the other people into buying what you sell.

Re:Arghh

[tomstdenis]

Naive.

ReiserFS and O(1) schedulers and IPv6 and ... were not in UNIX 40 years ago. Heck ReiserFS is a relatively new addition. I recall using ext2/ext3 and having to "fix up" the drive after every unclean shutdown.

I have yet to lose a single file to a ReiserFS on a medium that still operates. Even through several blackouts [before I got my UPS] and other shutdowns [emergency and otherwise].

I think you just need to reflect on what is actually in the Linux kernel to realize it is nothing like UNIX of 40 years ago.

Sure the overall design (e.g. monolithic with a touch of micro going on) is similar but that's about it. Sure the userland is the similar [with many UI enhancements!] but why not? Do you know a better way to remove files? What does "rm" lack? ... etc.

I think in the near future any OS improvements [e.g. new mm or FS or taskscheduler] will just find their way into Linux just like many new compiler breakthroughs find their way into GCC.

What Linux [and GCC in this case] give people is a well studied and tested framework to base new ideas on.

Tom

Tom

Re:Arghh

[Anonymous Coward]

stuck with Windows and MacOS and Linux because they're the only 3 that have developer support

And what about Solaris (including the open sourced version)? They may not be the booming company from the 90's but they are still developing the OS and tools/applications. I don't think you can leave them out of the picture.

Jim

Re:Arghh

[plopez]

yeah, and we are in the 21st century primarily using 19th century transportation technology. It doesn't make sense but things are so stable (in an economic and cultural sense) that a radical change would require a huge destabilizing event.

Re:Arghh

[Anita Coney]

Merely because something is old does NOT mean it should be replaced. We're still building houses out of wood after thousands of years. Our cars run on internal combustion engines. And after all these years we're still carbon based life forms.

You admit it's "fine," that it "works," and that there is "good stuff" in it. If all of that is true, then why replace it merely because it's old?! That kind of mentally makes no sense.

Re:Arghh

[rubycodez]

yeah, we've got nothing on those 19th century horseless carriages with their electronic fuel injection into multiple valves, high energy ignition, automatic transmissions and torque converters, radial tires, air conditioning, power seats and windows, safety belts and airbags, power brakes and steering. No sirrie bob.

Re:Arghh

[hungrygrue]

The bicycle is over 100 years old and is still the single most effecient form of transportation ever devised. Materials have changed, but the basic design has not. Innovation and improvement are wonderful a wonderful goal, but replacing a proven design based on nothing but an immaginary shelf life is silly.

Re:Arghh

[homebrewmike]

There is no such thing as old techonology. There are only appropriate and in-appropriate applications of techonology.

Writing is a hell of a lot older than Unix, but it's still pretty appropos.

Re:Arghh

[i7dude]

While it its old, it adresses the issues associated with current system design. One thing people are overlooking is that the basic architecture of pc's (non supercomputers) has remained relatively static for quite some time. Perhaps in order to see a radical change in operating system design we also need to rethink the hardware architectures that we have used for soo long.

dude.

Re:Arghh

[lxs]

BeOS was a marketing disaster, and Plan9 was never more than a proof of concept, that doesn't mean that their core ideas were bad, just that their time hasn't come yet.

Squeak is another 'proof of concept' system with a lot of promising ideas. (Finally a system where you can rotate your windows 37.5 degrees anticlockwise. Pointless but very cool)

It will probably take another 50 years or so, before a viable OS is built based on these ideas (Squeak for example needs a host OS and has no security at all, making it useless for real world applications), and until that time, we'll have to get by with Unix variants. Just remember that computing today is extremely primitive compared to both the visions and working prototypes of the 1960s and 70s. We have made enormous advances in hardware, but progress in software engineering seems to have stalled in the early '80s.

(Personally I blame Bill Gates and Bjarne Soustrup for that, but I have no hard evidence to back it up)

Re:Arghh

[lisaparratt]

To be fair, us Europeans have a word for American buildings:

Sheds

Re:Arghh

[ciroknight]

Merely because something is old does NOT mean it should be replaced. We're still building houses out of wood after thousands of years. Our cars run on internal combustion engines. And after all these years we're still carbon based life forms.

Oh boy did you bark up the wrong crowd with those words.

Houses would be better built with steel and concrete as they do less environmental damage, have a better resistance to natural disasters and depending on where you are from, it might save you some money on your energy bills.

Cars need to be running on electricity, because those carbon-burning, environmentally destroying devices we all have in our garages are not maintainable.

And I'm sure a hundred slashdotters would tell you that we'd rather be silicon-based lifeforms, and there's enough of us working on that possibility that it may become a possibility in the future.

Of course, I'm being very brief to get my point across, but still. Because something is old, it needs to be evaluated for replacement. If it's still the best thing for the job, use it. Else, replace it.

Re:Arghh

[varmittang]

But computers change much faster than bicycles do. And cmoputers rely on software and not just hardware. In a bicycle, the frame and wheels are like the hardware of the computer. While the gears and the chain is the software of a computer. You can change the gears, which some people in the Tour de France were not just changing gears but using a new oval sprocket that the peddles connect too (suppose to help the rider start the peddle downwards when the foot reaches the top of the stride), to change how the bike works. Change the computers code, change how it works.

Re:Arghh

[deander2]

You admit it's "fine," that it "works," and that there is "good stuff" in it. If all of that is true, then why replace it merely because it's old?! That kind of mentally makes no sense.

yes, but lord knows it sells widgets. which is what keeps the US economy running my friend.

what are you, a communist? :-P

Re:Arghh

[tootlemonde]

Because something is old, it needs to be evaluated for replacement.

On the surface, the criteria for replacing something old is the same as the criteria for replacing something new: is there a better way to do it.

In practice, things like amortizing existing investment, vested interests and training are decisive. These economic and psychological issues cannot simply be dismissed as a failure of imagination since innovations have work in the real world, whatever else their merits are.

Sometimes, like the end of analog television signals, someone just has to legislate change. It is a chicken-and-egg problem. Some things only make sense if every one adopts them at once.

To some extent the reason we still have wood houses and the internal combusion engine is because no government has demanded something better.

Re:Arghh

[GuyZero]

Merely because something is old does NOT mean it should be replaced. We're still building houses out of wood after thousands of years. Our cars run on internal combustion engines. And after all these years we're still carbon based life forms.

Oh boy did you bark up the wrong crowd with those words.

Uh-oh. Mixed metaphors are a bad start.

Houses would be better built with steel and concrete as they do less environmental damage, have a better resistance to natural disasters and depending on where you are from, it might save you some money on your energy bills.

Hm. Iron mine plus a steel plant versus a tree farm. Which does less environmental damage? And relatively few people make houses out of 100% wood - I can hardly imagine a situation where you need to absolutely avoid using wood at all costs.

Wood is light, strong, can be grown in a very environmentally sensitive manner (unless you build a house out of mahogany or something) and is easy to work with. It's really a very useful material that is unlikely to be supplanted anytime soon.

Cars need to be running on electricity, because those carbon-burning, environmentally destroying devices we all have in our garages are not maintainable.

Well, my car burns gasoline, not carbon. If you have a design for an engine that runs on carbon, please go away and enjoy your patents worth a bajillion dollars. And my car is pretty maintainable, thanks a lot. Changing the oil once in a while is about it.

I assume you mean that you don't like the idea of buring fossil fuels. Ask yourself, however, with electric cars: where will the electricity come from? Is it really more efficient to transmit power over lossy copper lines for hundreds of miles versus just converting the fule at the point where it's needed? And what about the environmental impact of creating and disposing of all those exotic chemicals in large electric car batteries? Your "solution" is hardly a clear improvement.

And I'm sure a hundred slashdotters would tell you that we'd rather be silicon-based lifeforms, and there's enough of us working on that possibility that it may become a possibility in the future.

Not my bag, but whatever, sure. Just don't shuffle your feet on the carpet in the winter. You might accidentally short-circuit yourself.

Of course, I'm being very brief to get my point across, but still. Because something is old, it needs to be evaluated for replacement. If it's still the best thing for the job, use it. Else, replace it.

Sure, fine. Just don't assume that anything newer is automatically better. Pretty much all of your examples demonstrate items in which the incumbent technology (wood, gasoline) is still the best solution for many situations.

Re:Arghh

[HyperChicken]

The bicycle is over 100 years old and is still the single most effecient form of transportation ever devised.

I would think sailing is more efficient, no?

Re:Arghh

[Anonymous Coward]

Bah, I usually hold a policy not to respond to people like you, because you're only out for the sheer arguement of it, as I didn't spend any time on my arguments at all. I simply wrote the first thing that came to my head. I don't care what little nuances in my language you peel out, or the options you hold. In fact, I'm not even going to finish reading your post because I don't care to deal with people like you.

My original point stands, no matter what you want to take from my argument and twist in your own way. In the end, we're on the Internet and arguing on the Internet is like winning the Special Olympics. Have a good life.

Re:Arghh

[Anonymous Coward]

Mod Parent up! I've listened to Alan Kay and Mark Guzdial and then looked at stuff like the video from the mother of all demos from the 1968 by Douglas Englebart and wonder how the software industry got so far behind. The stuff like Sketchpad, NLS, and the Dynabook concept which led to Smalltalk and today's GUIs. It took the industry until the mid 90s to get to where the researchers were 30 years or more earlier.

Re:Arghh

[sootman]

There are two kinds of fool.
One says, "This is old, and therefore good."
And one says, "This is new, and therefore better"
-- John Brunner, "The Shockwave Rider"

Re:Arghh

[hungrygrue]

Derailers and gears have been around nearly as long as the bicycle, and the concept of the eliptical chainring is also far from new. It is one of those ideas that comes and goes and comes again http://www.sheldonbrown.com/biopace.html The basic design and functioning of the bicycle is unlikely to change much. We do have rather new additions like suspension, but that is more of a specialization than a general improvement: for riding down a mountain trail, it is a good thing. For normal riding it adds weight and adds flex to the drivetrain which wastes energy. There have been other "innovations" such as automatic transmissions for bicycles that provide little benefit versus the added complexity and weight and have therefore never caught on. Bringing the analogy back to the original topic however: Sayining that UNIX should be replaced with something entirely new just because it is 40 years old is like saying that the bicycle should be replaced with an entirely new technology just because it is 100+ years old. What you wind up with with that kind of thinking is something like the segway - It is much slower, heavy, expensive, has to be charged up, has a very limited range, depends on batteries which will wear out over time, etc. etc. Sure, the segway wins on cool and neat points, but pitted against the bicycle in terms of just plain utility, it doesn't stand a chance. Immagine someone trying to ride around the world on a segway :-)

Re:Arghh

[ml0fl1n]

Like the old joke in my Cobol class in 1983, "No one knows what we'll program computers with in 20 years, but we'll call it Cobol."

While the name "Unix" is indeed 40 years old, Unix on newer computers is very, very different than it was 40 years ago!

Unix is an OPERATING SYSTEM; it is adapted to the hardware it is running on.

Re:Arghh

[The Lynxpro]

"yeah, and we are in the 21st century primarily using 19th century transportation technology. It doesn't make sense but things are so stable (in an economic and cultural sense) that a radical change would require a huge destabilizing event."

Careful about the "19th Century transportation technology" comment. Fuel cells were first developed in the late 18th Century so if they replace the combustion engine in cars, we'd technically be swapping out 19th Century tech for 18th Century tech.

Re:Arghh

[krell]

'Hm. Iron mine plus a steel plant versus a tree farm. Which does less environmental damage? '

The iron mine, for sure. It ends up taking a lot less room. The tree farm ends up being sterile "plantation", not friendly to wildlife, covering a very large area of land.

'Ask yourself, however, with electric cars: where will the electricity come from?'

It comes from very few points, which are stringently controlled for emissions: much more stringent than the thousands of tailpipes all over the place. It is a clear improvement.

Re:Arghh

[JabberWokky]

I agree with him. I work with wood (just built a Saxon Lyre), do medieval encampments (the authentic kind, not the turkey leg kind), and bind my own books. Not only am I using old technologies, I am using old manufacturing methods, ignoring assembly line practices and other efficient ways to do what I do. I even play music and sing instead of owning an iPod and like to take walks. I enjoy these things quite a bit, while I don't like, say, video games.

It is not even an issue of "older or newer". What is *best* is not always most desired or preferred - new things just give a wider range of choices. Things getting "better" is a phrase loaded with subjective criteria. I don't think that "old is better" either - I enjoy the internet, programming, international relief funds, polymer molding and modern medicine. I'd even like to colonize the universe with mankind. But there are plenty of concepts and practices that are not new or even "better" by your criteria that are quite vital and enjoyable.

That said, I'll take your final two comments and point out that I spent several years as a Special Olympics volunteer, and the joy and determination from those people is inspiring. And thanks - I am having a good life.

--
Evan

Re:Arghh

[TheRaven64]

I think you just need to reflect on what is actually in the Linux kernel to realize it is nothing like UNIX of 40 years ago.

Synchronous system calls, no standard ABI for cross-language function calls, no standard interface for accessing command-line parameters, no support for record files, no support for automatic versioning in the filesystem, directories stored in a hideously access-ineficient way. Looks remarkably similar to me. Do you know a better way to remove files? What does "rm" lack?

An undo feature? Seriously, MacOS had this in 1984, it's hardly rocket science.

I think in the near future any OS improvements [e.g. new mm or FS or taskscheduler] will just find their way into Linux just like many new compiler breakthroughs find their way into GCC.

Right, like how GCC's only just got SSA? Seriously, this is such an obvious feature I implemented it in the compiler I had to write as an undergrad because it seemed like the most logical way of handling things, without reading about it first. It allows simple graph-optimisation to be used to create optimised code and was first developed 20 years ago at IBM. Hardly a stunning piece of innovation, and yet touted as a great new feature in the latest GCC.

Re:Arghh

[zkn]

Its thinking like that, that made people buy windows me. (And some would argue XP)

Re:Arghh

[tomstdenis]

Cross platform ABI? .... application ***BINARY*** interface...

Yeah that'd sell well, just make sure you don't overwrite the link register on your x86!

Linux uses STANDARD syscalls to do all of the work from the application point of view. How the syscall function is implemented varies with the platform but the actual syscall number and it's arguments does not.

Versioning in the filesystem? Why would I want to waste my space on that? I CVS my source code, my /bin directory doesn't change often enough [or in any uncontrolled fashion] to require that.

As for an undo ... Gnome and KDE can do this... you can alias "rm" to your own command, e.g.

alias rm='safe_del'

Where safe_del moves the file to /tmp or something....

That's not a feature of Linux though so perhaps I shouldn't have brought up rm.

As for GCC...Well tell me how many other compilers out there target dozens of platforms, optimize fairly well [even better on x86/arm/mips] and still manage to grow over time.

SSA is not a totally new concept but combined with all the other stuff in GCC it is. If it wasn't you'd be pointing me to the URL of another cross-platform open source readily accessible compiler.

And at anyrate, suppose GCC is 20 years behind the times [which I think is not only untrue but a very unfair thing to say] ... it's still moving forwards. More so then say MSVC or Borland C.

Tom

Re:Arghh

[sickofthisshit]

Your excuses for UNIX behavior are simply apologies.

Versioning and safe deletion in the file system have real advantages. Version control is not a substitute for all the uses of versioning, and aliasing rm in a shell does not take care of all the shell scripts and make files that use rm directly.

Believe it or not, other operating systems have had behaviors that are useful that UNIX does not have, and really can't have without leaving the essential spirit of UNIX behind.

Simply denying these features are generally useful is silly.

Re:Arghh

[tomstdenis]

You have to balanced resource consumption vs. useful ness.

If I say do

rm -rf /usr/bin

That would be painful but I could have an operational box again in 8 hours [from scratch with Gentoo from bootstrap!!!]. It isn't the end of the world.

In fact faster than that since I backup my system drive every so often...

Wasting gigs of space so I can have 10 copies of ggv and 200 copies of some file in /etc doesn't help me any more than having my own archive.

And really, if you manage your OS properly [e.g. have non-root users] it's hard to accidentally kill the system. As my default "tom" user for instance I can't write to the bin directories or /etc or ... so I can't really screw up even if I tried.

And really having an archiving "rm" is ***NOT*** an operating system task. That's totally userspace.

So I think you need to take a class in "operating systems" at your local college and maybe pay attention?

Tom

Re:Arghh

[mikefe]

These features are available, but do not belong in the kernel.

Check out LibTrash that does just what you ask for, and sits between your apps and the kernel.

Re:Arghh

[mikefe]

That's funny, and I'm an American.

But from my point of view, I just have to laugh whenever I hear the word "flat" when someone referrs to where someone lives.

Re:Arghh

[mikefe]

I like that one. Thanks.

I'd say ...

[hawkeye_82]

the future is the HURD. Even in the future, the future will still be the HURD.

Re:I'd say ...

[shis-ka-bob]

And the HURD will be running on GaAs processors so we can finally up the clock rate into the GHz range and use optical fiber rather than copper.

Re:I'd say ...

[Anonymous Coward]

"The future is the HURD, and it always will be."

Re:I'd say ...

[ph43drus]

Ah yes, as my stat mech professor taught us, Gallium-Arsenide, material of the future. Always has been, always will be.

Do you think they'll port DNF to it?

Jeff

OpenSolaris

[Anonymous Coward]

I recently downloaded OpenSolaris and started using JDS. Right away I realized that it was the future. As far as Unix goes, this is the best I've used and the most polished desktop built on X hands down.

Let go

[Anonymous Coward]

"Not only is UNIX dead, it's starting to smell bad."

                                                --- Rob Pike

Re:Let go

[timbloom]

Unix is as dead as Apple.

The irony to MS blowing his argument away...

[gatkinso]

...is that back in the day when really only technically savvy types owned or operated computers, is when MS gained their stranglehold on the market.

We would like to think MS somehow bamboozeled the teeming masses, but that is BS. It was us they bamboozled with MS-DOS of all things.

We did this to ourselves.

Re:The irony to MS blowing his argument away...

[Brunellus]

Let's not forget why MSDOS won:

• It was cheap . Competing operating systems were not. The Apple and Amiga platforms were generally regarded as superior, but who really wants to spend money?
• It was fast . I loved macs. They were neat little machines, capable of doing a lot. They also seeemed downright sluggish in comparison with a DOS machine, which wasn't running so much overhead.
• It got the job done.
• Loads of applications--from Lotus 1-2-3 to Lesiure Suit Larry--were developed for MSDOS. Never mind that we all had to learn how to use the command line--the apps that you needed to get the job done (or have fun) were on DOS.

Within its limitations DOS was and is a good platform.

But what about Microsoft?

[durbnpoisn]

This all sounds fine and dandy on the surface.

But, doesn't it worry any of you that Microsoft is currently doing their absolute best to squash out every other OS on the planet?!?

At the rate that they are strong-arming hardware manufactures, and this DRM bullshit, there may not be any hardware available to run this wonderful new system on!

I don't know about you, but that's about scarey concept to me!!

Re:But what about Microsoft?

[HyperChicken]

But, doesn't it worry any of you that Microsoft is currently doing their absolute best to squash out every other OS on the planet?!?

Microsoft is doing their best to create the best product possible. If that means "squash out every other OS on the planet", than no, it doesn't worry me.

Re:But what about Microsoft?

[nsayer]

Microsoft is doing their best to create the best product possible.

No, they're doing their best to prevent anyone else from successfully selling a better one. That's not quite the same thing.

Let's not forget that they have already been convicted once of violating federal law in the furtherance of this goal.

Not likely

[ylikone]

Remember when they first announced opensolaris and everybody on slashdot was like "it's going to kill linux, bla bla...". Where is it today? It has a small handful of users and nobody really cares. Like it or not Linux has you pwned!

Re:Not likely

[oxygene2k2]

opensolaris was announced less than 50 days ago, and the process of open sourcing isn't even finished yet.

what do you expect?

as for solaris, 2 mio downloads in 6 months are pretty impressive in my opinion (and yes, you can resume broken downloads in sun's download system)

If UNIX is so well understood..

[concept10]

How long does or will it take to put build/develop a layer on top of Linux that will make the underlying technologies and processes transparent from the average user?

I agree with his assumption and that was an easy one to make with the sucess of Mac OS X and Linux.

I would rather see Linux succeed because of the freedom and the ability to customize it for any given application.

Re:If UNIX is so well understood..

[Anonymous Coward]

I would rather see Linux succeed because of the freedom and the ability to customize it for any given application.

This is in part why Linux hasn't succeeded. Simple and well defined systems appeal to the masses. Not that I've looked, but has anyone ever considered standardizing the user interface and allowing the underlying code to be whatever a user cares to select? Think about it in the sense of a car. All cars have the same basic user controls but the actual mechanics will vary in a way transparent to the typical driver. Maybe this has already been accomplished and I just haven't had to deal with enough open source systems to really be "in the know".

Jim

We know all there is to know

[SporkLand]

This sounds like one of those statements that future generations look back at and laugh.

The Quote?

[Saggi]

Is it just me?

I cant find the quote: 'I think there's a lot of hope for Linux, although I don't think that Linux is the answer. I think that UNIX is the answer, in some form or fashion. It might be BSD, it might be Linux, it might be some third thing. But UNIX is such a well understood and smart to handle the issues that an operating system has to handle that it ultimately will prevail.' ... in the article?!?

It has spawned a discussion, but the linked article is much more about Open Source, than UNIX. Try search for 'hope for Linux'. Am I the only one who try to read the article before posting?

Re:The Quote?

[debest]

Try "Page 2".

Re:The Quote?

[Saggi]

Thanks.
It was not easy to see the layout...

Re:The Quote?

[Anonymous Coward]

Am I the only one who try to read the article before posting?

You must be gnu here. Try, you fail, succeed you must.

** very idea of unmanaged code will disappear**

[oldwarrior]

those who bet their $$ on this will get waxed by those who squeeze the most raw power out of their CPU's/hardware well into the future. The goalposts will just keep moving. But some well behaved apps will live nicely in JVMs/CLR's but the edge will be in real code. They have made claims about byte interpreters since the Series/1, VM, UCSD P-System, and Java.

Linux is not UNIX ?

[randalware]

This is a distinction I do NOT understand.

        The underlying code open or not is just the implmentation.
        And some implentatoins have different switches on the commands.
        Like BSD,Irix,SYS V5, didn't

        Sticking to a design (what UNIX standard, POSIX?) is the bigger issue in my opinion.

        But the end result is the same.

        "Everything is a file" ( read,write,block or char)

        Or is there really a major difference ?

Re:Linux is not UNIX ?

[Anonymous Coward]

Linux is a clone of UNIX, unlike BSD who's source code can be directly traced back to UNIX .

What an ass

[DartonW]

Why should we take seriously anything that Leo Laporte has to say? He's not involved in the OS community; he's just a TV and radio personality. What, exactly, qualifies him as a pundit on such issues?

Re:What an ass

[genrader]

I've met the man, he kicks ass. Shut up. Been an avid fan of him for the past six years and the guy is pretty smart.

Re:What an ass

[MightyPez]

So fans proclaiming he "kicks ass" is a qualification? On that note, I'd like to proclaim my love for Tucker Carlson and Bill O'Reilly on their insightful political science knowledge.

Re:What an ass

[DartonW]

Yeah, whatever. Name-dropper.

Re:What an ass

[daperdan]

Leo has been hosting call for help Radio and Television shows for almost 10 years. He's written a number of computer books. I respect his opinion more than a slashdot troll.

I was never a big fan of his Call for Help show on TechTV but when I did watch he seemed to be very bright. He answered Linux, Windows, and Apple questions. Why the hate?

He seems to be a big proponent of Open Source. His site is proudly release under the Creative Commons License. He recomends a number of open source projects on his shows.

Re:What an ass

[DartonW]

That's precisely my issue with him. People think he knows what he's talking about simply because he's a celebrity or a "nice guy". But if you read his interviews or listen to his shows, he's clearly misinformed on some important issues regarding technology.

Time is cyclical after all!!!

[nigel_q]

This just proves that time IS, in fact, cyclical! Consider trends with fashion? The 70s came back, the 80s came back, the 90s are coming back... Remakes of movies and music, too... The same is true with computers! Remember how we used to have these big centralized machines that occupied cabinets or frames in rooms? And people used these things called 'Terminals' to interact with the 'MainFrame'... Now we call it a Server, and the Terminals "Thin Clients", but its the same thing! If this was such a good idea AND the future, why did we abandon it when the PC came out?

Re:Time is cyclical after all!!!

[Master of Transhuman]

Of course, we abandoned it for exactly the same reasons we'll abandon it AGAIN in the future:

Because the PTB that ran the mainframe were incompetent assholes who couldn't support our computing needs properly from a centralized position.

And Microsoft is EXACTLY the worst ITS department anybody could have. And Sun and the UNIX vendors aren't far behind.

So, yes, most people will use thin clients and complain about network response time - just like every terminal user used to do on an overloaded mainframe.

The smart ones will run their own rich client PCs on the side.

It's not going to happen soon. The buildout to allow the Internet to have enough bandwidth to do EVERYTHING over the Net will take another ten years - maybe twenty. These idiots can't even make getting your email or accessing a Web site reliable yet - how the hell are they going to make mission-critical Web services reliable?

Re:Time is cyclical after all!!!

[databyss]

It was only a good idea back in the day because it wasn't economical to be able to provide every desktop with the processing power that each user required.

It makes alot more sense in the corporate environment though when you get into multiple permission levels and access issues. A centralized server with terminals allows centralized control over access to various resources.

This will not be the way people arrange their home networks in the future. File servers will probably become more common as a central storage location for music, movies and potentially some programs, but nobody will be dependant upon the server for anything more than that.

People want their own machine do to their own things on.

It's probably even cheaper to buy multiple desktops than it would be for a central server and thin clients.

Re:Time is cyclical after all!!!

[dodobh]

PCs give a lot of freedom. However, in the hands of clueless administrators, PCs are a large cost.

Remove control from those people who are not willing to accept the responsibility that goes with it, or capable of it. They get the thin clients and dumb terminals. The rest can use PCs.

Just because you can drive a SUV does not mean you should.

Re:Time is cyclical after all!!!

[Rich0]

Because the PTB that ran the mainframe were incompetent assholes who couldn't support our computing needs properly from a centralized position.

I think the problem was an ivory tower one - the IT group's goals were not aligned with the business units.

I've seen this pattern re-emerging with the re-discovery of shared services in many companies. Here is how the cycle goes:

1. Start with lots of departments running their own mini-data-centers, help desks, etc.
2. Somebody comes to the realization that by centralizing these services the level of service can be increased (24x7 monitoring, professional staff, etc) while the cost is decreased (fewer servers, less headcount overall, staff freed for more strategic work).
3. Company centralizes services - cost of running servers goes from $10M to $500k. Somebody gets promoted.
4. New guy comes in - shaves budget from $500k to $400k. Gets a raise. Lays off 10% of staff, gets a raise. Aims for cost of $100k. Looks into moving servers to elbonia.
5. Business units find help desk cases going from 3 days under #1 to 1 day under #2 to 5/7/9 days under #3. Servers go down - IT gets to it when they get to it. Developers release new code, IT staff installs six months later.
6. New application is being rolled out. Business unit builds own "computer room" to host it to avoid dismal IT group.
7. Wait 5 years - go to step 1...

#2 is where you want to be - central services just make sense. The problem is that once you have one IT budget everybody starts hacking away at it until they actually raise costs by causing everybody to roll their own. Management needs to realize that #2 is optimum, and fund it well. Often budget increases in a central group LOWER overall costs, as it deters business units from reinventing the wheel.

Of course #3 is always acompanied by management edicts forbidding the creation of non-central IT groups. Business areas always find loopholes (10 workers who all do it in their "spare time", "small productivity tools" that aren't classified as IT projects, the Access database that grows to 10 million records, etc.).

In 15 years we'll be right back where we are now...

Re:Time is cyclical after all!!!

[Master of Transhuman]

I've always felt that IT staff should be seconded to operations departments for direct support - they get to know the real needs of the departments because they WORK there daily and can design and implement systems that are more likely to be accepted by the end users because "they" built it.

At the same time, they all should be required to adhere to corporate interoperability standards set by a central IT group that looks at IT operations of the entire corporation as a whole.

Of course, humans being primate hierarchical empire builders, this obviously rational scheme will never work.

I see it every day at City College of San Francisco where my boss is the only guy supporting the users of the Student end of the database and ITS is blithely flailing away trying to keep up with annual installs of the huge college COTS software at the recommendation of consultants who are feathering their own nest.

Centralization of SOME things in IT makes sense - servers and standards may be one of them. For others, like development and support, it doesn't.

Re:Time is cyclical after all!!!

[Anonymous Coward]

If this was such a good idea AND the future, why did we abandon it when the PC came out?

I think the answer is that the limitations of early PC technology created an environment where most PC users worked from a different paradigm.

When the PC came out, you still had a 'mainframe' and a 'terminal', it's just that both were sitting on the same desk. The thing about the PC was that it was (at first) a single-user batch-mode mainframe with absolutely no connectivity of any sort. And, it stayed that way for almost two decades (early 80s to late 90s). The AmigaOS was multitasking in the late 80s, but windows was still batch-mode well into the 90s. Thus even though 'serious' computers (running 'serious' operating systems like UNIX) had long been multi-user, multi-tasking and network-centric, the average PC user had no conception of this.

As a side-line, this is also part of the reason the windows operating system had so much trouble, first with stability and second with security. Both stability and security had to be solved early on with multi-user, time-sharing OSs on a network, but for a batchmode, standalone single-user DOS machine - hey, let's cut some corners! Decades later, those early design decisions (and the paradigms that went with them) came back to haunt ms and all its unfortunate users.

In a nutshell, we didn't abandon the idea when the PC came out, it simply wasn't possible when the PC first came out. Decades later, people are beginning to understand what it means to be connected to a pervasive network of services and servers. Thus the paradigm can move again.

Re:Time is cyclical after all!!!

[MidnightBrewer]

The corporate environment is not the home environment. The entire draw of the personal computer - emphasis on "personal" - is that you could have your very own. People want to possess, and they don't want other people messing with their stuff. This is why Apple has pushed laptops so much in recent years, recognizing that people are not only straying further towards having their own computer, they want to be able to carry it around with them, too.

That being said, I think there will be a natural evolution towards a more centralized system per household. Wifi networks are the beginning of this trend, in a way. However, until a server becomes as easy to install as any other appliance (turn it on and forget about it), then it's not going to become mainstream.

Re:Time is cyclical after all!!!

[16K Ram Pack]

I don't think it's cyclical.

This issue is much more about trade-offs. In this case, the cost of hardware+software+bandwidth vs cost of labour.

At one time, processing power was very expensive, servers were expensive, as a percentage the OS was cheap, network traffic was very expensive. PCs cut the cost of that. At the same time, they often required more costs in terms of labour.

Now, PC hardware is cheap, servers are cheap, network traffic is getting cheaper, but labour has gone up.

Poor Use of English Alert

[Anonymous Coward]

Ooops, forgot, this is Slashdot.

Uh....

[Dracolytch]

But UNIX is such a well understood and smart to handle the issues that an operating system has to handle that it ultimately will prevail.

-----

WTF??? I mean, really, come on now... WTF!

~D

Re:Uh....

[tkr2099]

Apparently English is not the future either.

Do One Thing Well

[wild_berry]

Laporte says:
"It's funny, because in the early days of UNIX, the philosophy of a program was, "do one thing well, and then pass the result along and interface with others." We've gotten to the complete opposite, which is do everything kind of okay, and interface with nobody. That was clearly a wrong turn. It's a response to market forces, not computer science forces."

In the case where there is just the CLI and a list of programs spawned from a single input line, having a whole collection of tools that work well together is a must. But when you move to a graphical interface, so huge is the change in interface mechanics that the idea of the end-user setting up a chain of programs to run from one mouse click should be alien.

The UNIX mentality of small, modular programs doing one thing well can still be maintained while a graphical environment is running, but his criticism that "do everything kind of okay, interface with nobody" can't be taken as criticism: it's just the way that GUI stuff appears to the user*. The computer system may be organised so that the GUI program you're using shares a lot of libraries and calls a lot of helper programs to do its work, but the user should only see the graphical interface, making his point moot.

*: Maybe he means something else: that an environment where one program does only one thing, from ground to GUI, does not help people to tinker, develop and hack new features into the software.

Re:Do One Thing Well

[iGN97]

But when you move to a graphical interface, so huge is the change in interface mechanics that the idea of the end-user setting up a chain of programs to run from one mouse click should be alien.

I've been playing with this idea in an application. It's sort of similar to the automator in Tiger, I've been told, except my system allows many-to-many-connections:

http://www.ignosaurus.com/pf1337/

The URL contains some screenshots and a flash demo.

I think there's definitely room for this kind of inter-application chaining, but in a more refined form. It makes batching a breeze.

The "acquire"-module is similar to a simple "ls" in many ways.

Re:Do One Thing Well

[99BottlesOfBeerInMyF]

OK I think you're missing two important points. First, chaining together lots of small applications easily provides functionality that is considered impossible or very, very hard to the average Windows user and even to most GUI only users of other OS's. Want to change the phrase "World leader in Iron manufacturing" to "One of the World's top leaders in Iron manufacturing" an a few hundred unrelated web pages, text files, help systems, etc? Easy as pie with the CLI, but time consuming with a GUI. Apple's Automator is an attempt to fix this, but it really has a long way to go.

Second, small CLI programs are really really good at what they do, because they are so specialized. How many spell checkers does the average Windows user need? Well the one in Word, and one more for each other application they use to compose text. Maybe the one in Word is great, but the one in their text editor is only so-so and the one in their e-mail sucks. How many people do you know that paste text into Word to spell check it then back into the program they are using. The point is you only really need one good spell checker, but it needs to be available to all other applications. On the CLI you just call the spell checker for your text file and then move on to the next application you want to use on it. In OS X they have brought this functionality to the GUI as well with system services that allow spellchecking (and many other operations) on all text (or image or sound or whatever) in all applications everywhere. I think this is the sort of thing he was originally talking about.

Re:Do One Thing Well

[chris462]

In the case where there is just the CLI and a list of programs spawned from a single input line, having a whole collection of tools that work well together is a must. But when you move to a graphical interface, so huge is the change in interface mechanics that the idea of the end-user setting up a chain of programs to run from one mouse click should be alien.

This is already happening:
http://www.apple.com/ilife/
http://www.apple.com/macosx/features/automator/

Re:Do One Thing Well

[zkn]

it's just the way that GUI stuff appears to the user

Think of it this way. I want to run an automated photoshop process on 20.000 images. Since Photoshop was developed to the "Interface with noone"-ideology. I still have to open op the GUI(one that is useless to the process at hand) and it even has to show me each picture open up, show the changes and then be saved.
Had they been of the "interface it" mind, they could have made it possible for me to run it in a commandline, sending back information of where the changed pictures are saved and stored(so further changes can be made when photoshop is done with it).

Currently there is simply a great gap between "GUI programs" and "console programs" when it comes to playing nice with others. This is the gap Apple is trying(trying) to close with automater, asking developers to make make it posible to use their apps through it. And just as the commandline automater isn't for everyone but those who use it find it extremly helpfull.

First UNIX system brought to the masses...

[Zweideutig]

OpenBSD (or something else) running on firewalls. Then eventually home servers brought to the masses.

Re:First UNIX system brought to the masses...

[timbloom]

It's already for the masses. Mac OS X is a great GUI (an understatement of course) on BSD. This is really the way to go since people really need the security of Unix, but they also need a break from the frustrations of Windows.

It's all about value ....

[Anonymous Coward]

... to both the customer and the producer.

"Proprietory == profit, open == kudos"?

Bollocks

Let's get down to tin-tacks - accounting in other words.

If I'm BillG and I sell you proprietory software for $100, I have to amortise that revenue over 10 years (say). Profit? $10/year

If I'm me, I give you the software and sell you the services to install/configure/manage/whatever it at $100 for the first time installation.

Amortisation? Depreciation? What's that? I'm a service guy. Profit? $100/year

Don't laugh. This is really the way accounting works.

I'm a much more "profitable" proposition (albeit there's tax implications, cost of sales etc but never mind)

Ever wonder why those paragons of hardware - IBM - are now more or less entirely services? Now you know.

Failure

[kryogen1x]

Isn't there going to be some sort of Unix failure in like 2038? How can it be the future if that's true?

Re:Failure

[Redrover5545]

Yeah, because time in a unix system is stored as a number of seconds since January 1st 1970 in a 32-bit signed integer called c_time. In January 2038, the integer will overflow and the destruction of humanity will ensue.

Except, of course, if c_time gets bumped up to 64-bits, as it will probably be done now that 64-bit processors are going to become a lot more common.

Re:Failure

[JabberWokky]

It has nothing to do with Unix, it has to do with 32 bits. Most *nix style OSes store time as the number of seconds since 1970. If you store it as a 32 bit number, it runs out in 2038. Ironically, this is easy(ish) to fix in the actual OS (just use a 64 bit number)... it's the databases and such (which may run on some other OS like Windows) that *store* those numbers that will have the bigger problems, since they need to be converted. Filesystems will be the same way; there has been some talk about changing the time storage there (some may have already). All those USB keys and such that use FAT will run out in 2099.

Incidently, the 64 bit number will last until... uhh... somewhere around 280 billion AD? I could be off, but it's way out there.

--
Evan

Re:Failure

[256byteram]

try this perl script:

use POSIX;
$ENV{'TZ'} = "GMT";

for ($clock = 2147483641; $clock < 2147483651; $clock++) {
print ctime($clock);
}

cant remember where I found it.

Re:Failure

[bk4u]

Hmmmmm, must be based on the Mayan calander.

Well Known OS Hacker Leo Laporte says...

[Dink Paisy]

Hrm... I'm not going to say a whole bunch of mean things, but read the interview. Leo Laporte isn't an OS hacker, doesn't seem to know the details of operating systems, and doesn't seem to know the history of Windows or Unix.

Although this interview doesn't have the controversial tone of a John C Dvorak article, the content seems to be similarly well thought out.

so the 21st century, the future of computer use...

[cyclomedia]

...is command lines, meaningless three letter acronyms and pipe scripts with hundreds of flags and switches. well i for one am jumping straight into this revolution. i just smashed my mouse with a hammer and am going to track down a CGA monitor on ebay.

OS

[paithuk]

If there is anything that drives me insane, it's people dribling on about what OS they use. Dude, it doesn't matter like it used to 20-30 years ago, we're past the OS era and what Linux or Unix really needs is some good quality, easy to use applications that complement a great graphics engine. Changing the OS is highly unlikely to change the success of a particular system, but changing how you think will...

Re:OS

[Anonymous Coward]

• "...what Linux or Unix really needs is some good quality, easy to use applications that complement a great graphics engine."

Ummm, isn't this called OS X?

The technology does not matter

[defile]

Microsoft's platform is the standard because they focused on the business of the software products market. They promised something to independent software vendors and delivered it-- a single platform that any developer no matter how big or small can target. At the same time they pushed hard to get this platform on as many PCs as possible, breaking kneecaps along the way when necessary.

They achieved a form of write once run anywhere. In 1985.

It does not matter what's under the hood, it mattered that the ISV only had to write one binary and not have to spend the money supporting two dozen incompatible platforms. Even Java cannot match this (I know, because I have to deal with it).

Today there must be half a billion PCs that the ISV can generate one single binary for, and with that you've covered what, 90% of the market?

Linux needs to offer big marketshare (doesn't have) and good developer support (has, sorta) for ISVs to care about it, because Microsoft proved that most ISVs won't bother targetting more than one major platform.

Re:The technology does not matter

[SilentShriek]

That's it exactly. It's the marketing, not the actual product that has won people over. The fact that Windows is so common as a desktop platform, makes it even easier to market it as the best way to go; average America wants new things, but they don't want to have to learn anything new. If it's not intuitive, based on familiar things, it's very difficult to overtake what has already been established. The brand names alone can carry more weight with the public than the innovation.

Re:The technology does not matter

[ookaze]

Microsoft's platform is the standard because they focused on the business of the software products market.

But fact is it is not the standard platform. There is none actually, Linux is the closest one to have this title.
MS had Win9x, Win NT, WinCE, and now .NET.
MS is incapable to make a standard anyway. You have mistaken monopoly for standard.

They promised something to independent software vendors and delivered it-- a single platform that any developer no matter how big or small can target.

They didn't then. A lot of them were screwed and others are till this day. Especially when going from Win9x -> WinXP, VisualBasic/MFC -> .NET, ...
And Vista will be the worst one.

They achieved a form of write once run anywhere. In 1985.

That just isn't true. See above.

It does not matter what's under the hood, it mattered that the ISV only had to write one binary and not have to spend the money supporting two dozen incompatible platforms.

Like Win9x, WinNT, WinXP, .NET, Vista, WinCE ? So they are all the same platform ?
Game developers did not have to support every graphic card in Windows ? You got to be kidding.
Vista just promised what you say, meaning it was never delivered before on Windows.
However, Linux does.

Today there must be half a billion PCs that the ISV can generate one single binary for, and with that you've covered what, 90% of the market?

No, the last numbers indicate 60 % of WinXP in the Windows market. So it's more like 50 %, which is big already.

Linux needs to offer big marketshare (doesn't have) and good developer support (has, sorta) for ISVs to care about it, because Microsoft proved that most ISVs won't bother targetting more than one major platform.

You didn't use the correct verb. It's not that MS "proved" anything, but rather they "assured". And we are talking the home market here. In the server market, that is Linux that moved the PC platform in the server room.
ISV care about Linux where they can, where MS can "assure" that they won't bother. The only OS I see used in handheld, appliances and PC with the same standard is Linux. Windows just can't do it. You have at least WinNT and WinCE. And soon .NET. I don't see your unified platform there.

Re:The technology does not matter

[poot_rootbeer]

They achieved a form of write once run anywhere. In 1985.

Then how come the programs on my DOS floppies wouldn't run when I stuck them into a Macintosh drive? Or an Atari 800? Or an Amiga?

Seriously. What are you talking about? QBASIC?

Re:The technology does not matter

[Anonymous Coward]

Never mind that. How come the programs on my DOS floppies won't run when I stick them into my XP machine?

Re:The technology does not matter

[Anonymous Coward]

Like Win9x, WinNT, WinXP, .NET, Vista, WinCE ? So they are all the same platform ?

Of course not. But a single Windows PE or .NET binary can run on Win9x, WinNT, WinXP, or Vista, so effectively they're "the same".

Game developers did not have to support every graphic card in Windows ? You got to be kidding.

No, you have got to be ignorant. Game developers on Windows don't need to know anything at all about the graphics card, and never have: Microsoft have provided abstractions (first WinG, then DirectX) which mean that only the graphics card makers need to know how the graphics card does stuff, while the game makers just write to the API and it all works.

I suggest you try learning what you're talking about before spouting off - it'll save you looking such an idiot when you're completely wrong about everything you say.

Re:The technology does not matter

[defile]

No, the actual product has won people over -- they're called average people.

The day they release a product that wins me over, that is, someone who works in technology, is the day they go out of business, because I do not have the same standards that the average computer user does and there are a lot less of ME

Re:The technology does not matter

[ciggieposeur]

Of course not. But a single Windows PE or .NET binary can run on Win9x, WinNT, WinXP, or Vista, so effectively they're "the same".

Wrong. The Win32 API has changed a great deal between Win9x and WinXP, I can write a trivial program that will work fine on XP but fail on 95/98.

Game developers did not have to support every graphic card in Windows ? You got to be kidding.

No, you have got to be ignorant. Game developers on Windows don't need to know anything at all about the graphics card, and never have: Microsoft have provided abstractions (first WinG, then DirectX) which mean that only the graphics card makers need to know how the graphics card does stuff, while the game makers just write to the API and it all works.

That's not entirely true. DirectX is not a part of the underlying "OS", it's an add-on. (One can also use OpenGL or Glide to program the hardware.) Claiming that I can write a game using DirectX 8.x features and it will run on Win9x (no DirectX), WinNT (DirectX 3.x), etc. isn't true, unless I bundle it with my game (as many do). I could just as easily claim that Java is the write-once-run-everywhere platform because after downloading the JVM you can run a .jar file.

On a side note, professional games often have to go a bit beyond the vanilla DirectX API to make a playable game. Many many cards have bugs in their driver implementation so certain features have to be disabled to make things work, it's not at all a trivial "put it in and everything will be OK". (Remember the S3 Virge?)

I suggest you try learning what you're talking about before spouting off - it'll save you looking such an idiot when you're completely wrong about everything you say.

Geez, take a chill pill. You sound like frickin' Avril Lavigne or something. Go write some POSIX C code for a while and come back to Win32 and see just how "standard" it really is.

Leo's on the sauce again..

[Anonymous Coward]

But UNIX is such a well understood and smart to handle the issues that an operating system has to handle that it ultimately will prevail.'"

WTF?

I get the "just of it", but come on... Is this a bad quote, or just a bad sentence?

Re:Leo's on the sauce again..

[Anonymous Coward]

I think you mean the gist of it, you fucking tool. And it's a bad sentence.

Talk about your pipe dreams...

[suitepotato]

One need only look over this book and do six months of desktop end-user support on Windows to see how insane an idea it is that Unix of any kind is going to win in the market over Windows as long as the Unix community remains ruled by sadomasochistic techie dweebs who love things based on how hard they are which is the exact opposite of the attitude that has allowed Microsoft and AOL to prosper and thrive in the common end-user market.

I love my FC3, but once again, don't mistake my technical abilities and the chance to flex them each day on it for meaning that everyone is going to take to it like a fish to water.

Apple's OSX most definitely is the best Unix-ish distribution ever conceived, built, and sold to end-users without any doubt in my mind. But do the Linux geeks get it as to why? No, they try mightily to avoid the BSD-ish ancestry of it and sit there wishing this beautiful *nix-style OS with such wonderful design and construction were a Linux distro.

Won't happen. Linux is dominated by the sort of people on whom it is still lost that ease of use, administration, and support are paramount over everything else for end-users. Windows XP and Mac OSX give them what Linux never will as long as the current crop of leaders and movers and shakers controls the Linux scene.

Re:Talk about your pipe dreams...

[Robertatwork]

Back when I was supporting SCO and using Linux; I told my self that if anyone ever came out with a usable UNIX, I would buy it. OS X came out and I bought it. I am now on my third OS X machine and I still am happy with it.

Re:Talk about your pipe dreams...

[Anonymous Coward]

I downloaded the linux source today.
Do I "dominate" linux too now?

Re:Talk about your pipe dreams...

[Anonymous Coward]

I think what you are saying is somewhat specious.

Sure having a slick UI Tool allows most technically able people to administrate a computer but like anything else in life, you get out what you put in. The command line affords much more flexibility in administrative tasks, sure if all you want to do is add a new user to your PDC or ADTree the UI is fine, but what was your experience when you needed to change a configuration setting on a farm of machines, and lets say for argument that this particular setting does not have a handy dandy checkbox available.

I have no money so I can't bet but lets make a sportsmans bet that the unix admin will have scripted the solution and be updating the entire farm before you have updated the second computer using the glitzy UI.

I suppose the point is that not every administrative task falls neatly into the predetermined options offered by a user interface. In the aggregate it is these types of task that are most visible to users in terms of outage or response time.

The ability to script together a collection of generic tools to achieve a specific aim is not something WIMP is currentlyu able to do competitively.

Re:Talk about your pipe dreams...

[Anonymous Coward]

"Sure having a slick UI Tool allows most technically able people to administrate a computer but like anything else in life, you get out what you put in. The command line affords much more flexibility in administrative tasks, sure if all you want to do is add a new user to your PDC or ADTree the UI is fine, but what was your experience when you needed to change a configuration setting on a farm of machines, and lets say for argument that this particular setting does not have a handy dandy checkbox available.

I have no money so I can't bet but lets make a sportsmans bet that the unix admin will have scripted the solution and be updating the entire farm before you have updated the second computer using the glitzy UI."

It seems that the assumption is that: in order to have a "glitzy" UI, the CLI must be abandoned. Why is that?

I thought it was all about choice?

Is it impossible to be able to admin from a GUI (joe end-user) while at the same time keeping the CLI intact for those that would rather use it (slashdot geeks)?

Re:Talk about your pipe dreams...

[zsau]

Apple's OS X has one huge problem that many Linux geeks recognise. It mightn't be a problem for everyone in the short term, but in the long term it'll become obvious.

MacOS X is closed source and proprietry. Late last year, I bought myself a new iMac G5. I liked my ROX desktop, and disliked Windows, but I thought something a bit more solid and coherent would be nice. After all, I didn't really do much modifying software, and the fact that so much needs to be compiled was pissing me off. (I started using GNU/Linux in the first place because I couldn't afford a Windows licence, and didn't want to run unlicensed software for what I considered then to be ethical reasons.)

Boy, was I wrong. I'm a geek, but not a skilled developer. I might occasionally make a minor change to a piece of software (which given my ROX desktop and my limited abilities usually meant choosing 'Look inside' from the Filer's menu, then fiddling with the Python files in there). The fact I couldn't do that on MacOS X was actually painful. I was also unable to make basic customisations to the interface like moving the close icon away from the minimise/maximise ones in window decorations. These limitations probably won't hurt your average user, but they will hurt people who are used to GNU/Linux and other *nix-like OSes.

These limitations are intergral to MacOS X. Linux geeks do not yearn for a beautiful MacOS-X-on-GNU/Linux. They yearn for a system they like. Some of them would like a simpler-to-administer system, and they're writing it. Others get paid to do it. Fortunately unlike Apple they are not selfish and are allowing others to make changes (minor or major) to their software. This is what Ubuntu and Gnome people are explicitly working towards (even if they're still a ways away). Even if they fail, they create the market, and with it the money that will pay other developers to write better software.

The system is not currently suitable for all (or even most) users, probably. But as more and more people and companies recognise the benefits of free software, and a greater variety of inputs and requirements are asked of GNU/Linux, desktop systems based on GNU/Linux are becoming accessible to more and more people. It will be and is strengthened by its diversity. And although it may be that (say) ROX isn't for everyone, if someone decides to convert from the undetermined überenvironment that is able to replace Windows as the everyman's operating system, they can do so at absolutely no cost. No driver issues. All their data will be accessible. Even all their old software will be accessible! In fact, switching back to what they liked before will be a menu-option away.

Isn't that amazing? Isn't that how computing should be? Isn't that the true way to complete userfriendliness?

Most people don't want to shape their operating environment around them, but some do. Those that do want to do so to different extents. A system that prevents me from doing that isn't userfriendly, any more than a system which requires me to do it, or forces me to a console. Only a free software environment can cater to all tastes. Linux geeks get this. They may be pursuing it in the wrong order. But maybe they aren't, only unwritten history will tell us that.

That's why GNU/Linux will still be going strong when everyone's forgotten that Apple did anything but make portable music players.

More concisely, the strength of GNU/Linux is that paste there is no 'crop of leaders and movers and shakers controls the [GNU/]Linux scene'. But anyone (including technophobes by virtue of money) can add extra stuff on top of it.

(To the extent that this post wasn't directly relevant to yours, it was something I wanted to say. As this is an open forum, such bits may be considered an open letter to people who aren't so understanding of open software.)

Re:Talk about your pipe dreams...

[mcrbids]

Linux is dominated by the sort of people on whom it is still lost that ease of use, administration, and support are paramount over everything else for end-users. Windows XP and Mac OSX give them what Linux never will as long as the current crop of leaders and movers and shakers controls the Linux scene.

Which is so much HORSEPUCKY. Recently, my Dell Inspiron 600m laptop died. A hardware failure, I'm still waiting for my system back from Dell. (with a new MB)

So, I have a laptop HDD, with all my email, browser, data, documents, and everything else that matters, sitting on a 2.5 inch HDD in my hand.

I called around to ask about installing a laptop HDD into a desktop system, and am told by a qualified professional that it's simply NOT A GOOD IDEA to mount my HDD into another system because of all the driver issues that will arise.

Well, guess what? I installed said HDD into a desktop system (after buyting a $15 adapter) and in 10 minutes had everything working fine. Sound, video, X Windows, KDE, the works.

Other than a changed resolution, it's the same computer, and all my stuff is there. So, I spend a morning, and am 100% back up and running.

Can Windows do this?!?!?!

I didn't think so. Linux isn't for everybody - but it's definitely for ME. I can't afford to have a hardware failure take me out, and my filesystem is contiguous to 1998! (I have email, documents, etc. going back now 7 years in a continuous history, across many O/S upgrades and system changes, yet I can read it all just fine. Can Windows do that, too?)

Re:Talk about your pipe dreams...

[mikefe]

Well, guess what? I installed said HDD into a desktop system (after buyting a $15 adapter) and in 10 minutes had everything working fine. Sound, video, X Windows, KDE, the works.

Other than a changed resolution, it's the same computer, and all my stuff is there. So, I spend a morning, and am 100% back up and running.

Can Windows do this?!?!?!

Yes.

Newer windows do it better than the older ones, but they all add (and leave old) cruft to your registry each time there is a hardware change.

That said, I find moving drives from one computer to another and expecting the software to work in the new system (after a few driver setting changes) much more reliable in Linux.

Its all about ease...

[pickyouupatnine]

If you notice.. for desktop computing anyway - the easier the UI (not even necessarily prettier) is, the more successful the OS is likely to be. I think when it comes to desktops of the future.. they'll always evolve - always borrowing heavily from the norm and making small leaps forward. UIs will have to look pretty and work the same way - not cutting out the same mundane steps to do smaller tasks, because Joe and Jane can't be bothered to remember multiple paths to a solution. When it comes to high end OSes - be they number crunchers, art machines or big honkin' servers - this is where Linux has the greatest chance of beating everyone - the users of these OSes care about productivity. Yes the UI will be important - but it'll be to make things more productive than to look pretty. Continuous streamlining of the OS and the UI - even if it means not having the same OS for all things - thus making linux the best choice since it can be built like (as someone mentioned a little while back) like Old lego sets ;).

not one answer

[javiercr]

there is no reason why there will be one single answer, 'the answer'.

The future?

[heptapod]

A thirty year old operating system is the future of computing?

Of course Linux isn't ready for the desktop!

[bahwi]

Of course Linux isn't ready for the desktop! But what they don't tell you is that Windows isn't either!

C'mon, Spyware, Adware, Numerous Bugs(My Soundcard driver crashed the other Day. My Microsoft Certified Driver completely crashed. A reboot and it worked, but that's unacceptable as it never has any trouble in *Nix). Crazy Service Packs, bad to no real support.

Hell, you NEED an Anti-Virus just to browse the net and check your email, even if you don't download and open any attachments. Just to protect you from the wild internet. You have to combine XP + Norton + Ad-Aware/Spybot S&D just to get a near usable PC. That's quite a stone's throw away from a desktop.

The problem with it, is Windows IS used as the desktop, even though it isn't ready for it yet. That means it is the standard, however, how often has your mother had to call you over to fix it? Linux wouldn't require the same thing, especially if all they want is browsing and email. They're quite matched at that point. But no, Linux isn't ready. Neither is Windows.

(I can't speak of OSX, I don't actively use it)

Re:Of course Linux isn't ready for the desktop!

[Anonymous Coward]

Hell, you NEED an Anti-Virus just to browse the net and check your email, even if you don't download and open any attachments. Just to protect you from the wild internet. You have to combine XP + Norton + Ad-Aware/Spybot S&D just to get a near usable PC. That's quite a stone's throw away from a desktop.

Bullshit. I use Win2k; I don't have any anti-virus or firewall or anti-spyware software installed. And my PC is perfectly usable.

Once in a blue moon I download and run an anti-virus program or an anti-spyware program, just to check, so I can say for a fact that the total number of viruses or spyware to have made it onto my PC, despite the complete lack of any specific software protection, is zero.

It's amazing what keeping your OS patched, sitting behind a router for NAT, and avoiding Internet Explorer and Outlook Express can do.

So, no, your problems aren't with Windows... tell me, do you know what PEBKAC means...?

Re:Of course Linux isn't ready for the desktop!

[bahwi]

Haha! Again, how do you avoid IE and Outlook if you want Internet and Email? Firefox?? Opera?? You're on a crutch, yet again! You need third party software for it to be an internet enabled desktop. And how many of our mothers avoid IE and Outlook? (Since Linux is held against that criteria, Windows should be too).

Re:Of course Linux isn't ready for the desktop!

[Anonymous Coward]

I love it when posts are modded as funny when the poster is serious.

So long as Freedom Survives

[Zombie Ryushu]

What does it matter if Linux, FreeBSD or OSX winds up on top? The important thing, is that ideas like freedom, the GNU and GPL, Open Source, and individual liberty survives. Right now, that is the biggest threat.

What we need to worry about is stuff like TCPA destroying our freedoms of speech and expression. Worry about that before whether or not we worry about whether our favorite OS becomes the leader. If we don't take that threat seriously, it won't matter what OS Comes out on top bcause we won't be having this discussion.

My idea: the reverse-mullet approach

[British]

Unix in the backend, handling all computery stuff(services, servers, etc).

A nice, pretty GUI up front(Macintosh, Windows, whatever you like), that grandma can use.

IIRC OSX does this to an extent already.

Thus, the reverse mullet approach. Party in the front, business in the back.

Re:My idea: the reverse-mullet approach

[TheClam]

Reverse mullet? Is that bangs with a buzz cut, or what?

Re:My idea: the reverse-mullet approach

[onosendai]

'to an extent already'.. No extents about it, you just described OS X - take away Aqua, and OS X is a pretty hard-core and stable BSD.

Re:My idea: the reverse-mullet approach

[Anonymous Coward]

Like Flock of Seagulls?

Re:so the 21st century, the future of computer use

[freshman_a]

...is command lines, meaningless three letter acronyms and pipe scripts with hundreds of flags and switches.

yep, and here's a perfect example of what you're talking about.

I love all this endless squabbling

[gelfling]

It reminds me so much of the company I work for where endlessly arguing about how something is impossible and how change is absolutely the wrong thing to do is what we do in and of itself. Explaining to ourselves why we need to continue to safely fail is really what we spend almost all of our time and effort doing.

I celebrate mediocrity and I cheer that open source is finally in the boat with us!!!!!

Huzzah Huzzah!!

Fear the Hurd

[Mr. Underbridge]

the future is the HURD. Even in the future, the future will still be the HURD.

If the GNU kernel had been ready last spring, I'd not have bothered to even start my project: the fact is that it wasn't and still isn't.--Linus Torvalds, 1992.

Some things never change, eh?

Re: ** very idea of unmanaged code will disappear*

[Felinoid]

Video games current drive the hardware market and have always had some very tight code.
While the avrage user may not have noticed it back in the Commodore 64 days frame rate was an issue. However getting 30 fps was very unlikely. It was more of getting enough frames so the human eye didn't see jerky motion.

The faster the hardware the smarter the AI. The faster the video card the better the detail.
Thats just the games.

The scientific community is always updating hardware and even worked out that it's better to delay a project a week to get new hardware than be stuck on old hardware. The newer hardware will get the job done a few months faster than the old.

Then you have the movie industry. CG movies and CG specal effects were the domain of super computers and high budget films. The new Doctor Who series is using CG effects and PC farms are being used to render CG movies.

There is something else to consider,
Linux can run on 2 meg ram and 1 meg disk.
It's hard to imagin anyone making anything with requirements lower than 16 megs total (memory and storage)

Just fab a X86 clone with 64 meg ram and 2 gig flash and you have an imbeded Linux on a chip.
(Didn't somebody do this already?)

Linux is already populare in servers and gainning popularity in the imbeded market.
(Totally acing out Microsoft. Linux compeates with the far more practical PalmOs.)

Linux isn't the only OS for this. BSD works pritty well.
And if you want to drop to the 6502 then Lunix (not Linux) is your only choice.

GUI vs. Commandline

[Savage-Rabbit]

I wonder why MS is working on a new command line at all.

Try comparing the old CMD shell in Windows to Bash...

Do people buy Xserves so that they can use the OS X command line? Do people run linux because they love staring at those grey characters on a black screen? No one really likes the command line... plenty of people get by with it, but it's obviously the most primitive computer interface. So why is Microsoft developing it? Do they really believe that *NIX users like their OS because of the command line?

People buy Mac servers precisely because they marry Unix/Linux based commandline power with the advantages of the slicker and simpler Windows GUI interfaces. Most Unix/Linux systems have a very good commandline toolkit but the GUI toolkit often sucks, on Windows this is the other way around. Yes GUI management interfaces have their advantages just like the Command line has its advantages. Alot of computer Geeks seem to believe that GUI tools are for morons but that is crap. You have to choose your tools based on the task you have to perform and sometimes the GUI tools are just better. When you want oversigth and have to deal with alot of information GUI interfaces like the one for MS IIS are very good. The problem begins when you gets stuck with some huge project like, say... migrating 800 websites from IIS 5 (Win 2000) to IIS 6 (Win 2003) with as little downtime as possible. If you do not have a REALLY good set of commandline utilities and scripts you will go mad doing a project like this with GUI management interfaces only. Microsoft is just waking up to the fact that GUI tools nice as they are can not replace the commandline completely and it has taken them far to long to make that realization.

Ferget, HELL!

[Thud457]

CP/M will rise again!

Commodities, Sales, Illogical consumers

[bubbaD]

Consumer technology follows psychological factors, not engineer's logic. Hence we have iPods dominating the mp3 player market. It will also just be easier to just stick in well-known un*x and let managed code run on top of it. Not really secure, not the best way to do it, but its easy and time-saving and pointy haired managers will like it better than developing an in-house solution.

Re:so the 21st century, the future of computer use

[CrazyWingman]

May i ask which three-letter acronyms you find meaningless?

The reason that UNIX will win...

[agraupe]

Is that programmers like to develop for an open source system. It's easier that way, and if they release their code as OSS, it just keeps building. People always ask me, "How do I do X?" where X is a semi-difficult task. I always find myself saying, "Well, I'd do it with this program in Linux, it would take about 5 minutes. The windows equivilent, on the other hand, takes the afternoon to figure out and get right." If there are any moderately useful programs for windows, they are usually cheap payware or annoying shareware. The reason that UNIX/Linux/BSD/OS X will work is that you can do almost anything for free.

Re:The reason that UNIX will win...

[alecks]

Man, I don't know what you're smoking. programmers like to develop for an open source system. Oh yeah?? How do you know? Just because you do?

It's easier that way It's easier??? How is it easier? If I write a program in C++ it'll be easier to write if I make it Open Source, versus if I charge for it?

The windows equivilent, on the other hand, takes the afternoon I don't know about you, but for me, between WSH and WMI, i can write some pretty complex things in no time.... and don't get me started on the ease of developing on MS's IDE.

Re:The reason that UNIX will win...

[Clansman]

Hmmm - I think this is not really a solid position. Linux is equally as bad|normal at having functionality that takes a fair amount of time to get up to speed with in order to get this so called 5 minute pay back. LDAP, apache, sendmail, emacs, perl, bash, gimp all take time to learn and master before payback.

So - in fact, both platforms are the same then.

Except one OS can't play a huge (vast!) range of modern and popular games that are easy to install, perform well, and are sold in Virgin, HMV and Game.

Re:The reason that UNIX will win...

[agraupe]

Although I can't speak with certainty on the first point, I will correct you on the second. I was referring, at that point, to regular user tasks, not programming. And, yeah, I suppose Visual Studio might be nice if you can afford it.

Re:The reason that UNIX will win...

[agraupe]

It's true that those tools do take a while to master. However, I will provide a counterexample: ripping a DVD. I've done this on windows, and to end up with a DivX, you require at least two (possibly three) programs, a guide to show you what options to use, and the patience to find all these tools and carry out the process. On linux, it's as simple as apt-get/emerge dvdrip, and away you go. Sure there are some options, but the defaults are reasonable and it's all in one program: once you hit go, there's nothing more to do. The tools you mentioned do take a while to learn, but so do their windows equivilents. So linux is the same as or better than windows on this front.

LISP machines: Worse is Better

[bubbaD]

http://kde.sw.com.sg/food/worse_is_better.html
Un*x isn't the best solution. LISP machines like those from Symbolics were much better 20 years ago. I think VMS was more secure and reliable, also. Un*x will prevail because its good enough, not that its "the best"- New Jersey solution.

Re:LISP machines: Worse is Better

[sickofthisshit]

You can't say that totally different OS designs are "better" than one another, as if there were one dimension of "goodness" in computing.

What Symbolics really kicked ass at (I am told), was (essentially) single-user programming workstations with *ultimate* flexibility and customization. Genera lets you inspect and patch virtually every aspect of the system, down to the CPU microcode level.

Securely serving applications to hundreds or thousands of non-developer users simultaneously with fault-tolerance is a different dimension, in which other operating systems beat Symbolics, and mainframes can often beat ordinary UNIX. Raw computing or I/O power is another.

Symbolics machines still exist, and quite a few developers still use them regularly. For compute-intensive stuff, current boxes blow them away, but as the ultimately "pimped-out" developer system, they still can be more efficient to work on.

Re:LISP machines: Worse is Better

[bubbaD]

You're right, but did you read the "Worse is Better" article? Perhaps I should have left out the part about Lisp machines. The OP seemed to beleive that Unix's superior technology got it where it is today. That the MIT approach to tech worked. Even if you can think of areas where Unix works "the best," you would be wrong to think that its being best was The reason for un*x's success.

BTW Lisp machines also the nice feature of being able to debug programs at run-time: you could rewrite a program while it was running.

That's why it *will* be UNIX

[public transport]

As the OS becomes less visible to the users, the dominace of a specific OS is no longer a sales argument. It will be more important to the vendors of the devices to have a customisable, stable and cheap OS on which they can put their polished wonderware.

Of what fame????

[Dragon218]

...Of ZDTV Fame, not G4/TechTV

I miss the basement studio

Re:Of what fame????

[zerOnIne]

I miss the basement studio

I miss Kate Botello.

Re:Of what fame????

[Dragon218]

I miss internet tonight and the surf guru.

That's where I found out about Google.

Re:Of what fame????

[Anonymous Coward]

formerly of G4/TechTV fame) Why is 'formerly' there? He is still on techtv!

Re:Of what fame????

[Dragon218]

there is no more TechTV. It's gone the way of the do-do bird, and the communism scare.

Comment removed

[account_deleted]

Comment removed based on user account deletion

Poor LEO nees a Brain transfusion

[shareme]

Poor Leo confusing Unix falvours to the point of claiming that hey are not Unix.. Okay lets help the poor old geiser out: Unix Flavours: BSD FreeBSD MacOSX SOlaris HP-UNix Linux AIX Xenix BeOS .and many more.. Constrast this with how many up to date windows flavours..:)

Re:Poor LEO nees a Brain transfusion

[mbonnett]

BeOS is not a flavor of UNIX. Discussion at comp.sys.be.misc regarding Be and Unix.

You have got to admit he is 100% right

[tod_miller]

It might be BSD, it might be Linux, it might be some third thing

So, its either option A, or option B, or an option C which can be anything.

He has given himself quite a bit of leeway there.

If Marshmallows evolve into the dominant lifeform on this planet, his dying breath will be, I was right I tell ya!!! Its the third thing!!

(yes I RTFA and yes he really says that)

Re:You have got to admit he is 100% right

[aCC]

For the record, the whole quote is:

[...] I think that UNIX is the answer, in some form or fashion. It might be BSD, it might be Linux, it might be some third thing. [...]

So, no, he won't say "I was right I tell ya!!! Its the third thing!!" when Marshmallows evolve into the dominant lifeform of this planet. Unless they are a breed of UNIX by that time and that UNIX has transformed into a lifeform which I seriously doubt....

Re:You have got to admit he is 100% right

[Anonymous Coward]

Leo is an outspoken and known user of Apple Macintoshes so most assume he meant that.

WTF?

[bubbaD]

WTF?- meaningless 3-letter acronym. Sorry, I found that attempt at humor irresistable. Personally I'm bothered by two letter acronyms like MS, VB and IE.

What?

[mstyne]

UNIX is such a well understood and smart to handle the issues that an operating system has to handle that it ultimately will prevail.

Leo, what the fuck are you trying to say, dude?

Re:What?

[Anonymous Coward]

s/such a/so/

going backwards is the American way

[Anonymous Coward]

havent you been paying attention recently ?
we have the same politicians in goverment now as we did in the 70s/80s, oil is at record highs like in the 70s/80s, so it only makes sense that computers reflect the trend

Calling all fanbois!

[OxygenPenguin]

Why does every UNIX article/interview/water cooler discussion have to turn into a fanboi flame war, with people flaming about how OSX is UNIX, and how MacOS already has all the needs of the desktop, etc, etc, ad inifinitum. As much as I love operating systems and product evangelists, the discussions are here are really starting to get out of hand.

What would be nice to have people talk about would be the feasibility of the prediction that is being made, in reference to principles that are desired by the average computer user. Not how a current operating system already has all of these things, and thus should be considered the obvious future choice.

Someone already pointed out that the most technologically sound examples don't neccesarily make it in the industry, and this is illustrated time and again.
 
  Let's start talking about stuff that actually matters, or perhaps ::Gasp!:: even start doing it.

Screensavers Sequel

[eWarz]

Those that are fans of the screensavers can hear leo's new podcast at www.twit.tv.

Re:Screensavers Sequel

[Mantorp]

Until a few weeks ago I thought that Leo was the one who brought the least substance to the podcasts compared to the others. But listening to the podcasts that were done without him was just painful. There's a good reason that Leo is the only one among them who is still getting paid to be on the air.

Home gateway/server

[CdBee]

I agree with the points you make but I'm unconvinced by the idea of a home server, at least, as we understand it today. My guess is it will prove to be more like a combination of a wireless broadband router and a Network Attached Storage device.. to provide simply DHCP configuration, authentication, connectivity and storage in a locked-down environment.

Editor: Not "formerly", he's still at G4/TechTV...

[Anonymous Coward]

"Leo Laporte (formerly of G4/TechTV fame)"

No, he's still at G4/TechTV, just not airing in the United States. Leo simply moved to Toronto. Call for help is still being produced with new shows daily here in Canada.

Apparently, he still has an audience here.

Re:Editor: Not "formerly", he's still at G4/TechTV

[Kookus]

You Canadians really are glutons for punishment eh?

Leo is just the dopier version of Bob Villa, who knows how they still get on TV... Well, Bob is starting to know what he's doing atleast.

Can someone explain G4/TechTv to me? Please!

[Anonymous Coward]

Having only recently freed myself from the shackles of Comcast, I saw G4/TechTv for the first time. As a long-time gamer (my first game as an oscilloscope) and professional s/w dev, the subject matter is of great interest to me. However, I find in unwatchable. I guess I am a bit older and less "extreme to the max" than the 14 year old market that is their main demo, but something about seeing people IN THEIR lates 20s and 30s acting like teens makes me vomit in terror.

you definitely should have left that out.

[Anonymous Coward]

In comparison to C and Unix, Lisp and Lisp Machines were slow and expensive, which was enough of a factor to slow their adoption and eventually stifle the platform altogether. In light of this, calling it 'better' is deceptive--it manifestly wasn't, at least not in any sense that includes cost and efficiency in the picture.

That's like saying that using fourier transforms for multiplication is 'better'. Maybe it is (a) if you're multiplying *extremely* large numbers, or (b) if time/space considerations aren't considered. That is to say, in a general sense, and for most common applications, it isn't better.

However--you may see what the world of Lisp machines might have looked like yet--in this age of Java, .NET, Java Machines, embedded runtimes, etc., you have something that's in principle much closer to Lisp, and much more widely adopted as well. I wouldn't call it 'better' yet, though. :)

Unix is already everywhere

[NatteringNabob]

Sort of like Elvis. For better or worse, by virtue of POSIX compliance (sort of), even WinXP is 'just another Unix', just not a particularly good implementation. If the trade press didn't have pro-monopoly blinders on (thoughfully provided by the monopolists, and would be monopolists) they would have noticed that FOSS, through the mechanism of the BSD TCP/IP implemeention, not only out-innovated everbody else, it decimated the proprietary competion virtually extinguishing it.

Solaris 10

[Anonymous Coward]

I think Solaris is the new 600 pound gorilla on the block. It is now Open Source and is now in my opinion the most technically advanced UNIX-like OS.
(I'm not counting the GIU as part of the OS) People say they like BSD's stability and long history, Solaris has that today and other's like Linux's long list of native applications. Solaris will have that too once Janus is finished (Janus will allow linux binaries to run under Solaris with no speed penalty) Today Solaris scale better then the others and runs well on 64-way multiprocessor systems

Re:Solaris 10

[Anthony]

It is true that Sun seems to have made some improvements. Weren't they supposed to fix ufs with Solaris 10? DOn't get me wrong. Solaris pays a lot of bills.

Solaris scale better then the others and runs well on 64-way multiprocessor systems

This Linux box scales to 64 processors. So does This AIX/Linux box.

Boycott Laporte-Why do people still listen...

[abricko]

He's a living, breathing commerical and he's pratically in bed with Apple... you should've heard him crying on his radio show after apple annouced the intel switch... something like: 'Biggest mistake, i'm sad to hear this, i don't think this is right...'

(I know what you're about to say but my shower radio is set to KFI AM 640 so i occasionally hear part of his show)

He knows just enough to do commericals on his radio show and will step over his own toes to do them, yet he's regarded by the media as "The Tech Guru"

Boycott Laporte!

Re:Boycott Laporte-Why do people still listen...

[stilleon]

I heard that broadcast via the podcast version. He worried about the ability for Apple to keep the OS linked to the hardware. He explained that Apple always has considered itself hardware first, but the possibility in his mind is that the OS will be cracked, it will then be fixed to run on the average Dell and pirated. This would undermine Apple's apprach to the game. I like Leo a lot. He loves OS X because it is based on OpenBSD. He is a UNIX freak.

Re:Boycott Laporte-Why do people still listen...

[jaycontonio]

If Leo was truly as smart as he thinks he is, he wouldn't have had such a ridiculous prediction for the Intel move. He is constantly making the wrong predictions on his show and the only reason that twit is even getting downloaded is because of Patrick Norton and Kevin Rose.

Your "vision" is a nightmare...

[Anonymous Coward]

...but it is the way the world is moving.

Re:Editor: Not "formerly", he's still at G4/TechTV

[Anonymous Coward]

what made him move? I found him arrogant and a bit of an Ass to trash Linux,he ignores a great Linux distro like Xandros!

We are getting there...

[Nahor]

3. Components for Music, Television, Desktop, and Video Game consoles will (in many cases wirelessly) interact with this server.

We are getting there: Orb Networks

There is no ultimate and Windows of opportunity

[Bozovision]

Oh where to start? This intro is so misleading in so many ways.

Anyone who has noticed anything about computers has noticed that software is intrenched by the network effect; that is, you use a particular bit of software because your friends and colleagues do, and you want to be able to interoperate easily. Windows is a bit of software, and the network effect tells us that it's not going to go away unless Microsoft actively work hard to destroy their market. (This is possible - groups of people make mistakes all the time - but it's not likely.)

Excluding computer geeks, who are a tiny though influential minority, there are only two points where people change operating systems, and one is much less influential than the other.

Firstly, people will change OS when they change computers. BUT everyone wants to completely retain their existing investment in skills and software, so people make the minimum change for the maximum benefit. For most people this means that they are prepared to upgrade operating systems to a later version, but aren't about to change OS completely.

(To change completely the OS has to be 10 times better. That's a tall order since most OSes have incorporated the ideas of command line and GUI, and there aren't many fundemental breakthroughs floating about which can lead to 10x value. Anyone want to prove me wrong?)

The other point where there's the possibility of change is when there's a form-factor change. By that I mean computers physically change shape or function. It happened when mainframes gave way to minis, and when minis gave way to PCs. During the switchover period there's a lot of flux and it's anyone's guess who will come out tops.

But Microsoft recognise this as a serious threat to their business, and whenever a competitive model arises, they work overtime to crush it, whilst making sure that they can switch if need be. (Sensible strategy since it protects the value of their existing investments.)

When EO and GO where hot new ideas that looked like they were on an exponential upwards curve, MS used FUD, Pen Windows, and allegedly anti-competitive behaviour to bleed the idea dry. When thin clients threatened Windows, MS made sure that it was more expensive to use thin computing with Windows than it was to not. When MS figured that there were an awful lot of Playstations, each of which could easily be turned into a computer, they made sure that they had a competitive product so that they could take care of the market if it grew. And with set-top boxes, we have Media Centre edition.

There are two noteworthy cases where there's form-factor change. Firstly, there was a virtual form-factor change with the Internet where was a window of opportunity when the browser could have become the de-facto way of interacting with software, but Microsoft responded to the change fast enough and the opposition wasn't united. MS built IE which took enough share away from Netscape during the period of upheaval which meant that Netscape fell below critical mass for the network effect. This was as much Netscape's failre as MS taking it, because NS didn't build in mechanisms that re-inforced the network effect and they didn't keep providing developers with a better and better platform, whereas MS did, but at the same time made certain that it didn't get so good that it threatened the traditional software model.

The article touches on one area where there's a form factor change which threatens Windows - - cellphones. MS have followed part of their usual strategy - make sure that they own part of the market to fragment it, but also ensure that there's some retained value should there be a serious trend. Cellphone operators play a hefty part too - they don't really like the idea of having an open software platform, probably because they wouldn't/couldn't control it. So while every phone has an OS, only a small percentage of phones expose anything with any serious capability to developers and the user. And what is exposed is tightly controlled. And the platform

where they both went wrong

[emil]

cli and gui continue to make many mistakes. The OS to fix these first wins my vote (even if it's M$):

• The gui allows a "regex select" dialog which selects files based on either standard regex names or ksh syntax
• The main regex library can be replaced; pcre or training-wheel regex is available, depending upon user choice
• All bundled apps support regex (M$, it's great that you support unix regex in findstr, now put it in Word and Excel)
• ksh wildcards and regex should never have diverged; time to reunify (perhaps it takes M$ to do this to us; it would force us to abandon Bourne)
• M$, why do I need to download unxutils to create hard/soft links on NTFS? Why do you despise your own POSIX so?

In general, gui and cli have ignored one another for too long. Share some ideas!

Re:Editor: Not "formerly", he's still at G4/TechTV

[abricko]

You're so right! Laporte is exactly like Bob Villa... Who is far from Norm Abram!!

That is a fantastic comment, I love it!

UNIX IS the future

[Chanc_Gorkon]

More and more people use UNIX or Linux every day and don't know it. One example would be the Tivo. There are many many more examples of Linux being used in the embedded space. This is proof that UNIX or Linux CAN be made user friendly. There will ALWAYS be the Debian's, the Gentoo's and others for the real top end geeks, but lots of people CAN be comfortable with Ubuntu, Linspire and others. The sheer customizability of Linux is how this all can happen. Managed code can be run on top of UNIX and users will have to type a password or do some biometric security to install certain items. The modulairty of UNIX saves us from the hell of a integrated Windows. That does not mean that software will perform bad or every program will have a different interface...one CAN have a system that is LESS integrated, yet the componets work well. Just look at pipes, grep and regular expressions in UNIX. Things like these allow you to make it seem like desparate programs that were made by different people with different ideas can all work together. Small components working together hand in hand instead of giant monolithic api's will save the day.

I think Leo is right, but for different reasons

[tcampb01]

I recall the days when computers didn't network very well at all. Even your IBM mainframes couldn't talk to your IBM PCs without installing some proprietary special-purpose product to make communication possible -- and they were both made by the same company. Once standardized networking began to emerge (IP networking) the industry changed. I do recall the suggestions that the PC could probably run some of the applications that the mainframe was running and how absurd the IT folks thought those claims were... it would never happen. And yet... it did happen. Once computers could finally communicate, a reason for open standards, open systems, and open software emerged.

Zealots like me like the idea of "openness", but I don't think Joe Average Consumer really thinks a whole lot about it. But... I do think that Joe Average does want "interoperability".

A common comment you'll hear from Apple owners is that they perceive their computer spends more time helping them do what they want to do rather than them spending all their time supporting their computer. Taken differently, these (mostly non-technical) people are really just saying that they don't want to care about their computer... they just want it to work.

So far, consumers do still need to "support" their Linux installations. Linux doesn't "just work". Windows still needs too much care & feeding... it doesn't "just work" either -- especially where security is concerned. Mac OS X seems to be the best at just working -- even for people who don't know what they're doing. Yes it happens to be Unix. Yes Unix has a more stable foundation than Windows. But we only know this because we're largely a technical community. Joe Average doesn't know this and doesn't want to know this. Most of them probably have no idea that Firefox is considered to be "open" and that IE is considered to be "proprietary". They run what they run either because (a) it was there or (b) their computer buddy told them that's what they should use or (c) that's what they learned to use at work or school -- and I almost forgot... it works with the sites they need to visit.

Open software tends to be more interoperable - so I think it's tends to slowly erode away at the market share of non-interoperable systems. But I think the real difference is that interoperability wins over non-interoperability... not that open systems win over proprietary ones.

Though I want open software to win out, I must confess that I don't believe the average consumer really cares a whole lot about it.

Wood still catches fire and burns down...

[emil]

...and so does UNIX, from time to time.

If we mandated the use of fire-proof building materials, incredible amounts of damage to personal property might be avoided, but the costs of retrofitting the building industry are excessive.

But... if you absolutely, positively, cannot tolerate a fire, then you probably want a "VMS" house (if you don't mind writing your shell scripts in FORTRAN).

In the same way, object filesystems, system-wide GC, system-wide bounds checking, and permissions that support a more complex model than ugo would lead to fewer system cracks.

We don't do it because we're lazy.

CLI is a speech interface

[Anonymous Coward]

The command line interface is the closest we have to a real speech interface where you use words to tell your computer what you want it to do. If you look carefully at the UNIX command line you will see it is cryptic in precisely the same way that IRC and chat is cryptic. Everyone shortcuts the commonly used words into a shorthand representation. Beyond that and the keyboard (which a large number of people still fear) it isn't that complex. It's just verb noun syntax.

    Michael

Which platform created the concept of superuser?

[brokeninside]

Traditionally, typical users have very limited access in Unix while PC operating systems such as DOS, Windows (pre NT) and Mac OS (pre OS X) allowed the user to pretty much have complete control of the hardware. It seems to me that locking users out of the hardware is much more in line with the way that Unix traditionally works than the way that PC operatings systems have traditionally worked. The only difference is that instead of the root account being controlled by a BOFH that lives down in the ops center, it gets controlled by DRM chips.

POSIX

[Anonymous Coward]

I think that you have hit on an important point, we have been yammering on about this OS or another, but if we put the same effort into making sure the POSIX standards are afvanced and implemented it would appear to make the majority of the OS questions moot. User interfaces should be simply be interfaces to the underlying business logic and / or operating systme functionality, if we were generating code, be it Virtual machines, C++ libraries, foundation frameworks, LISP iengines or whatever, that are expressed in terms of the POSIX standards then generating the libraries for any platform is relatively simplistic. We are almost there right now but more needs to be done.

If we could only agree on a user interface markup language and a POSIX notification and event propogation system then we are done.

Each OS is then competing on the merits of how well it does these things. In which case Apple wins on the desktop, monolithic kernel UNIX systems or specifically tuned micro-kernel systems like L2 (HURD) win on the server side and windows, well windows will always tempt people from the path.

And for everyone who wants to reply and say what about Java. Well what about Java.

Thus far it has failed to deliver on its promise and is ever more entrenched in the server side where it faces increasing competition from simpler more focussed technologies.

As the article suggests the UNIX philospohy of chaining together gneral tools to achieve a specific aim is no different to component based development witht he shell providinf the glue.

Todays corporate environment is no different, Serives that perform one function well and that are orchestrated together to perform a discrete business function is how web services are being used today. People who are engaged in the development of monolithic software systems are struggling mightily with the restictions imposed by their implementation tool of choice and the confines of their programming paradigmn.

We are all working real hard to ve better engineers, that's why we disagree so much, its healthy and it leads us to a better place.

Keep up the debate.

Mod parent interesting / insightful

[spectre_240sx]

This is probably one of the best posts I've read on slashdot for a month.

This sums it up well:

[EmagGeek]

"Hi everyone, I'm not a technical expert, but I play one on TV."

As it was in the beginning, so shall it always be.

[jmorris42]

> No one really likes the command line... plenty of people get by with
> it, but it's obviously the most primitive computer interface.

Speak for yourself, MCSE.

The command line is the most natural interface possible if you are computer literate. Think of it as comparing books to TV. If you are a literate person you might still watch TV to veg out and because it is a totally different medium it can do some things better. But even though seeing the Battle of Helm's Deep was hella cool, the books tell a much more detailed and better story.

Yes, graphical tools are handy for new users and even us old timers can use them for really simple tasks, but dependence on them should be avoided by those seeking mastery. UNIX is a language and you won't ever understand it until you reach a conversational level at the command prompt.

> Do they really believe that *NIX users like their OS because of the
> command line?

Yes, and you will pry it from my cold dead fingers. Command prompts, everything is a file, pipes, redirection. These things are what make *NIX what it is, any attempt to change that will be met with fierce resistence. See resource forks for an example of an idea the graphical dweebs try unsuccessfully to foist off on us every year or so for an example.

Hic Jacet Unix

[hey!]

Rex quondam, Rexque futurum.

In short thers's simply not, a more congenial OS.

Unix is like a fine cheese

[Anonymous Coward]

Unix is a little like math, if done correctly math is entirely logically correct. It doesn't age or corrode. In Calculus, although new methods arise, the basic techniques don't change much - if it ain't broke, dont fix it.

In a similar way, the maturity in Unix means that it is very stable and efficient platform where much of the junk has been gradually removed over time (Something that MS and many Linux distros could benefit from).

You also seem to imply that unix is dead in terms of development - IT THRIVES! Take a look at Mac OS X for an example.

Apple have utilised unix under pinnings to create an entirely modern OS (and compare this to OS 9 et al. OS 9 is not 40 years old but still remains somewhat barnicle encrusted and is in no way modern, unlike OS X).

Unix may be an old concept, but this does not make it any less modern than other OS's.

Laporte's oppinion

[Anonymous Coward]

is about as usefull as Don King's. He's just a figurehead.

somebody

[akhomerun]

somebody read the recent article about the future of the net and copied all of their opinions from the article.

the radical changes that are predicted for the future don't always work out. we were supposed to drive electric cars by 2000, we're still driving ford pickups.

Keep it down...

[tod_miller]

Else SCO will start asking for licenses from confectioners.

Licences

[tod_miller]

I think (but then I realised yanks write both with -se, so, as The Rock would say - it doesn't matter).

Unix maybe, but not as you know it..

[bored]

More clueless crap, for Unix to really be the future, it needs to get rid of its legacy bagage and truely become "well understood". Frankly a lot of people think they understand unix because they are stuck in a single process/text based enviroment mindset. In reality the "extensions" made to unix to support current programming models are full of holes.

When RAS, threads, async io, multiple processors, and may other things that really are the "future" (or rather the current state of the art) are well understood by the unix community they will understand what needs to be changed in the model from the 1970's the people claim is Unix. When that happens unix will be the future, but it won't be "Unix" as you know it.

Now for some more concrete examples. Lets start with a simple one. What does the system call "close()" do? Thats right, did you know it can fail? Whats the solution? Try again. Now think about what happens in a multithreaded enviroment with open() happening in other threads. I can't find a link to Linus's comments on this but they are ammusing. The bottom line is that in a threaded POSIX enviroment you have to write code that looks like (in psudo code to remove the specifics):

app_open(filename,...)
{
      lockmutex(globalopenlock)
      rc=open(filename,...)
      unlockmutex(globalopenlock)
      return rc
}

app_close(filehandle)
{
      lockmutex(globalopenlock)
      while (close(filehandle)!=EBADF);
      unlockmutex(globalopenlock)
}

If such a simple unix concept as open/close is screwed up by threads, just imagine what happens when you write code to trap percise floating point exceptions, deal with async filesystem IO over an unreliable network, the list goes on. Basically unix is good for certain kinds of applications and absolutly blows chunks for other kinds. Everyone doing a lot of these things has tied themselves to a particular Unix implementation and uses system specific knowledge to solve the problem.

Still you don't see the point!

[bubbaD]

Damnit! The original poster was trying to make a sarcastic comment about unix! My whole point is that UNIX IS NOT A SUPERIOR TECHNOLOGY. It has gained acceptance, and will continue to be a major factor in the future because IT IS GOOD ENOUGH. I also was hoping people would READ THE CLASSIC "Worse is Better" ARTICLE!!! UN*X WILL NOT BE SUPPLANTED BY SUPERIOR TECHNOLOGY IN THE FORESEEABLE FUTURE, and it has never been GODDAMMIT!

The AC is also just wrong about what killed LISP machines. That's made me angry.

Re:Still you don't see the point!

[sickofthisshit]

The whole problem with your approach is that you use the word "better" with an "MIT-style" bias. The New Jersey folks simply have a different definition for the word.

I've read "Worse is Better" multiple times over the past decade. But nobody is going to read it when the link is coupled with gratuitous Lisp-machine fanboyism.

Yes, Genera is an amazing piece of software, yet to be matched in many ways, despite today's hardware UNDENIABLY being *much* faster (even running Lisp) than even the fastest Symbolics machine ever made. But that can't be summed up in the one word "better" without becoming a troll or flamebait.

Perhaps before blowing up at an AC, you should elaborate on what *you* think killed Lisp machines. Kent Pitman has said, as I recall, that Symbolics was mainly killed by bad business decisions, unrelated to the technology. But that doesn't mean that there is a different set of decisions which would have left Symbolics systems on top of the heap *today*. Intel still would have beaten their processors soundly in general performance. Without hardware tagging, Genera relies on a virtual machine.

Re:Unix should not be the Future

[bubbaD]

Have you read the old "Worse is Better" essay?
I posted a link to it and rambled a bit a bout LISP machines, and I got replies from people who just didn't "get it." I hope more people in the future get a taste of using lisp, but I'm cynical about it ever getting out of a niche in the programming world.

Re:Unix should not be the Future

[Bob Uhl]

Yup, 'Worse is Better' is a real classic. Although to tell the truth I'm not convinced that it's evocative so much of the Lisp/C divide as the LispM/Unix divide. And part of its point was that in a Darwinian sense worse really is better.

I think that Common Lisp and Scheme are probably more popular now than they ever have been. A great deal of this is due to the essays of Paul Graham, who despite his faults has been an excellent evangelist for the Lisp family of languages.

Of course, what really interests me is the next step: the language which offers features no Lisp can. Is that even possible?

Re:Editor: Not "formerly", he's still at G4/TechTV

[Jardine]

Leo simply moved to Toronto.

No he didn't. He flys up to Toronto and films a bunch of episodes all at once. Then he flys back home.

I believe he's in Toronto for about 1 week per month.

Re:Editor: Not "formerly", he's still at G4/TechTV

[^_^x]

Yeah... it's worse now than when he was on USA TechTV...

It's kind of like all the irrelevance and inaccuracy of Dvorak, and all the excitement factor of Dave Chalk's old show, but with a very small set. :/

Leo is Baghdad Bob?

[saskboy]

"But UNIX is such a well understood and smart to handle the issues that an operating system has to handle that it ultimately will prevail.'"

I think not, Leo. UNIX is of the past, and with its decreasing market share, and lack of mainstream marketing, it's not going anywhere but down.

Re:Leo is Baghdad Bob?

[goMac2500]

UNIX is gaining marketshare. Especially this one little known UNIX distribution called Mac OS X.

It's not the GUI

[jesterzog]

The UNIX mentality of small, modular programs doing one thing well can still be maintained while a graphical environment is running, but his criticism that "do everything kind of okay, interface with nobody" can't be taken as criticism: it's just the way that GUI stuff appears to the user*.

I don't think the GUI-ness of an interface has much to do with it. There are already several programming environments (often aimed at children) that are based around the organisation and connection of graphically-presented blocks, each of which does some function. Some of them are designed to give lots of graphical and visual feedback about what's happening, and hints about why something is or isn't working. Command Line Interfaces are a neat and efficient way to do similar things, but they're not the only way.

If there's a problem, I think it's that most people who are general users of computers don't want to have to learn or understand how to chain things like this together. They're after much more immediate results, and usually more intent on dealing directly with their problem rather than wanting to learn how to construct a program in detail. I can sympathise with that -- often if I buy something as a necessity, without being really interested in it, I'll settle for something that's recommended instead of taking the time and effort to get or build one suited exactly to me.

Re:Which platform created the concept of superuser

[quarkzone]

Brilliantly insightful! Wish I had mod points (hint-hint for any who do).

This can't be done in UNIX, why?

[lokedhs]

no support for automatic versioning in the filesystem

SCO OpenServer (yes, that operating system) had versioning in the file system back in the 90's. I suppose other UNIX versions does too. You don't need to replace the entire OS just to add such a feature.

The same goes for all the other features you requested. All of those fits very well into UNIX if you really want them. The reason they aren't there is not because it can't be done, it's because people hasn't cared enough to implement it.