Slashdot Mirror
The State of GNU/Linux in 2002: It was Good.
An anonymous reader writes "This year has proven most interesting for GNU/Linux. While there was not any amazing surprises, there were numerous events that are noteworthy for review. The upshot to all of this is that most of what happened was good overall for the Free Software community. Read the full story."
It rocked! Then again: I'm using Debian/Woody which is about a year old so I wouldn't know...
You can't judge a book by the way it wears its hair.
The State of GNU/Linux in 2002: It was Good.
By Timothy R. Butler
Editor-in-Chief, Open for Business
December 31, 2002, 20:55:12 EST
Free Software
This year has proven most interesting for GNU/Linux. While there were not any amazing surprises, there were numerous events that are noteworthy for review. The upshot to all of this is that most of what happened was good overall for the Free Software community.
One of the most notable events for Linux this year came early in March when Sharp Electronics started marketing the first mainstream GNU/Linux PDA. The Zaurus, which premiered at $500, started off by appearing in electronics stores across the United States, including Best Buy and Office Depot. While by the end of August, Best Buy had cleared out their stock; Office Depot seems to be sticking around for the long haul.
The Zaurus seems to have fulfilled the promise of a powerful GNU/Linux PDA, attracting interest from a wide array of pundits and purchasers alike. In fact, yours truly is writing on the said device right now. On a side note, the Zaurus SL-5500, which will be replaced with the 5600 in early 2003, won the best Innovation award in the Open for Business Open Choice awards in July.
Linux got another round of mainstream attention in June, when, through a series of announcements, mega-retailer Wal-Mart announced plans to offer not one, but three different Linux distributions on their line of Microtel PC's. Since then, the company has also started to offer custom configuration of the systems.
2002 was also a year of uniting, starting with the formation of UnitedLinux in late May. The alliance, a motley group of two has-been GNU/Linux contenders (Caldera and TurboLinux) and two fairly strong players (SuSE and Conectiva), left most people wondering what exactly the group hoped to do in its quixotic attack on Red Hat. More puzzling was why anyone would want to ally with a company with as little to offer as Caldera, or would let Ransom Love - whose name does not represent the emotion most often expressed towards him - speak as a "spokesman" for the new consortium.
It does seem that the group did finally get the message that Love, if anywhere in the organization, shouldn't be the one speaking for the group - perhaps the muffled sounds of him speaking with a sneaker jammed into his mouth was what finally got that change accomplished (needless to say, Love's comments caused lots of fire towards UnitedLinux for its first few weeks). Whatever the case, the group did manage to get UnitedLinux 1.0 out by year's end, just as they promised, which shows at least the development is better managed than the publicity.
UnitedLinux wasn't the only game in town for various groups joining forces this year, either. We (that is, Open for Business) were part of the alliance that created the popular new GNU/Linux news network Linux Daily News on July 2. It should be said that talks between OfB, LinuxandMain, DesktopLinux.com/LinuxDevices.com, and KernelTrap.org began months before the UnitedLinux announcement, so this was a strange coincidence and not a sign that your GNU/Linux media has taken to playing follow-the-leader with the companies we cover.
Not particularly surprisingly, Red Hat did not fail to provide some of the most interesting (and controversial) news items of the year. It all started in the spring, when Red Hat announced a first ever program offering a rebate to those who "upgraded" from select other distributions to Red Hat Linux 7.3. Beyond drawing a lot of attention in commentaries, it is not apparent whether this program was successful or not.
Red Hat continued plowing forward with controversial moves when, in late July, it became public knowledge that Red Hat was asking a major project it had often ignored to promote Red Hat, while in exchange, the company would do nothing more than mention the project's name on some signage. That might not have sounded like a bad deal had it not been one of the most recognized projects in the community: KDE. Considering that KDE receives most of its support from Red Hat competitors MandrakeSoft and SuSE, and none from Red Hat itself, the project's developers weren't too keen on the offer. After Open for Business' exclusive coverage of the conflict, however, Red Hat made good on its mistake, and KDE showed off their handiwork on Red Hat, using a nice machine provided by the company.
Still, Red Hat, perhaps on a quest for the controversial, hit two more big items as fall approached. The first, BlueCurve, has been a rather polarizing issue; those using GNOME seem to like it, whereas those using KDE do not. The issue at stake is the fact that most of "KDE" on Red Hat systems has been modified to look and act like GNOME. In fact, the vendor went so far as to make GNOME applications default over their KDE counterparts in most situations, even when using KDE. The other issue we mentioned promised to show perhaps what the "red" in Red Hat really meant, as the company surprisingly removed the Taiwanese flag from the localization dialogs in favor of the communist mainland's flag. Worse, at least to those who don't look up to Mao, Red Hat labeled the change a "bug fix." Ouch.
Still, it might be most appropriate to rename Red Hat to Midnight Grey Hat (Black Hat is taken), as the company announced a first ever profit in its latest quarter. That's right, the company best known for the mysterious guy in the red Fedora is no longer in the red.
Unfortunately, things didn't go quite as well for those in Paris, France. MandrakeSoft, the popular GNU/Linux developer that almost wasn't until the company's founders managed to take back the helm in 2001, was forced to issue two pleas for financial assistance as it continued to reel from the debt built up during the previous management's one year reign. While the company's "MandrakeClub" was at first looked on as nothing more than a donation system, by the time that the March 2002 financial communiqué caught everyone's attention, the membership service was finally starting to shape up into something useful.
MandrakeSoft also stirred up the dust of controversy earlier this month with the announcement of their new Multiple Network Firewall product. While the package's availability as a proprietary product (as well as a GPL'ed download), and the associated lock-down on intellectual property, caused somewhat of a skirmish, Duval cleared up much of the concern in an interview with me two weeks ago.
Certainly, MandrakeSoft isn't alone in the financial woes department, with two other long-time GNU/Linux names, Lineo and TurboLinux, going up on the auction block over the year. A third, Caldera, didn't go up for sale, but it did change its identity to a company it bought the year before, renaming itself the SCO Group, and seemingly refocusing on the proprietary UNIXes that the original SCO had developed.
The year was also a great period for software releases, with the highly anticipated GNOME 2.0 finally arriving and bringing that desktop back into the ring with KDE, which had been significantly ahead since the 2000 release of KDE 2.0. Still, due to a decision that is hoped to increase the latter project's next generation platform longevity, KDE 3.0 was already out the door before GNOME's June release.
Other exciting software releases for the year most certainly include the first non-development release of a full-fledged Free Software office suite, OpenOffice.org 1.0. The suite, while still tied to a molasses slow startup time, offers nearly perfect compatibility with Microsoft Office 97/2000/XP, making a non-Microsoft office product finally a reality for many businesses. Likewise, while it really offers only a few major advantages (and some disadvantages) over KDE's Konqueror, Mozilla 1.0 finally did arrive on the scene attracting attention from many mainstream sources. While its impact on the "browser wars" may be minimal, it does promise a real alternative to Internet Explorer on pretty much any platform.
Last, and certainly least, if one goes by media attention, was the arrival of popular "proprietary GNU/Linux" distributions. The duo of Xandros Desktop and LindowsOS (incidentally, both based on the same never released Corel Linux OS 3.0) both sport much more restrictive licenses than any other popular distribution, requiring per-seat licensing for businesses. As our review of Xandros revealed, the package is a pretty good deal -- but mostly for those looking to run Windows applications under GNU/Linux. You'll have to wait a few more weeks to hear from us about LindowsOS.
In all, it may not have been a perfect year for GNU/Linux, but in my opinion, the successes far outweigh the failures. All indications seem to suggest that 2003 might just be even better.
Timothy R. Butler is Editor-in-Chief of Open for Business. You can reach him at tbutler@uninetsolutions.com.
The Sharp Zaurus!
200+mhz in my pocket along with 64mb of ram, and Debian GNU/Linux as soon as that damn SD card I ordered comes in!
You can't judge a book by the way it wears its hair.
To my mind, the best thing, and it's a biggie, is that we finally have a distribution (Redhat 8.0 -- perhaps others?) that, out of the box, renders fonts so that they look good to non-nerds. This is the first step towards bringing Linux to the masses!
Next we need to radically cut the number of choices that the average user needs to make at install-time (Gee, which of the following 87 libraries should I install? And what the hell is a library anyway?)
If some entity (Redhat? IBM?) just grabs the bull by the horns, we'll have a good Windows replacement in a few months! Pleasepleaseplease somebody do it!
Many Linux users have been waiting for Linux to break out and start converting more users. Walmart certainly helped supporting Lindows, which i hope succedes as a desktop replacement. I think It's demize is the generally high price of the Subscription. In other light I know of schools and many other instutions switching to MS bassed mail systems due to ease of maintence and webacces they offer (Yes Many Linux solutions exist I like them myself). But a switch to MS Products is very bad for Linux on the server side...espically considering security issues as Windows is insecure.
I agree ith the PDA article. I found the Sharp to be just as usefull as the Palm software and almost as easy as WinCE. I think the Small evices market could easioly be dominated by Linux because software for those devices needs to be customized by a manufacturer and the cost quickly becomes cheaper for manufacturers due to little to no cost for the Linux and abou tthe same cost to customize it as any other OS (ie Drivers for the hardware and customicing software).
I hope the economy gets better
Happy New Year
1. Write lots of PR about GNU/Linux
2. ???
3. Claim that the year was good for GNU/Linux
Yes it's all been replaced with binary now.
Well if not for Linux, then for the users....
Key projects are starting to mature and become more 'user friendly' which is important for the desktop usage Linux is missing.
a few examples are:
Desktop+ performance features have made it into the 2.6 kernel, things like pre-emption, lower larency, higher frequency clock, async-io, better threading, alsa.. All great news for desktop users.
Support is under development for older video cards (often build into modern chipsets) in the DRI project, giving decient X performance to most users.
KDE and GNOME had a lot of new features in the latest releases, the next couple of revisions should see the features maturing, with performance gains, more universality/intergration &co...
CDRecord has added support for IDE drives (without having to run ide scsi) application like arson are making cd burning easy... all good news for the home user.
Wine is comming along in leaps and bounds....
Xine and Mplayer now work for more-or-less every format out there, hopefully next year will see Moz plugins.
This year was good, next year will be better...
thank God the internet isn't a human right.
I've only been using Linux for about 1.5 years now, and it amazes me how fast things get better in the OSS world. I mean sure Linux has been around for 10 years so maybe that's not "so fast", but in the last year I've noticed huge strides.
The first time I installed linux (redhat 7.1) it took me a few tries to get it to see my mouse, my laptop video card didn't play nice, my desktop sound card didn't get found and took like 3 weeks of teaching myself kernel compilation stuff to get it up and running, my desktop NIC was a hassle, and I thought the desktop choices were attrocious (KDE 2.2 and gnome 1.4 I Think...)
Not to mention any software to do real work (Office apps, decent browser) or to have any fun (IM, Decent mail client) had to be installed after the fact requiring more compilations, and messing with the system...
More recently I installed RedHat 8 on my desktop and laptop... Oh the beauty... Gnome 2 is a truly nice system if you ask me. the new theme is easy to look at (finally!!) All the apps I need (OpenOffice, Gaim, Evolution, Mozilla) are the defaults and are already installed. All of my hardware was perfectly and flawlessly recognized, even my wireless network card was setup during the installation (Shake a stick at that WindowsXP!).
All in all, night and day, in 1 year its gone from taking 1-3 days to get a desktop linux system really ready for production to about 30 minutes... If the next year holds as many leaps and bounds of usability MS will be in dire straights soon.
I have still done WindowsXP installs during the last few months that don't recognize all of the hardware in a box, especially wireless network cards (the linksys wpc11 most notably). Besides the fact that from a clean install of WindowsXP you still have to install all of the software (office, developement environment), it still takes at least 2 hours to get a windowXP box really ready for use, then another 4 to do all the updates it needs... (granted, it takes about 2 hours to download and install all of the redhat updates since the 8.0 release.. but it all happens in the background and doesn't require a reboot, while with WindowsXP and windows update, there are at least 4 updates that you have to download *alone* and then reboot after each one, meaning to do the updates, you are going to reboot 5 times and you have to babysit the box while the updates are happening, times reflect downloading on 1mbps DSL).
In this users opinion, its been a GREAT year for OSS and Linux, and I hope it just keeps getting better.
A: Bill Gates doesn't get it.
...can't you just pick a profile from the list of install options, and it takes care of the packages for you? I could be wrong.
;-)
Besides, some of us like the look of the sans-serif non-anti-aliased fonts.
Note to M1-ers: a curt but otherwise insightful message is not "Flamebait" or "Troll".
Program at school without attracting too much attention as my iBook tends to do.
You can't judge a book by the way it wears its hair.
Doesn't look like free software will have a bad year for the forseeable future. More and more fortune 500 companies are converting to or seriously considering converting to Linux (in whole or part). And it's even more common in Europe. Not to mention the fact that many governments are in the process of or considering switching to Linux.
Sorry to burst your bubble, but free software is a growth area in IT and it will be until MS is ground to a shard of it's former self.
In 20 years, the only people using MS will be home users. Games are MS's biggest advantage over Linux. That's the one area in which it is not making headway. MS PC's will essentially be X-Box v 5's by then. They will not be used for serious business applications.
First off, bad sex can be just bad. Trust me on this one.
Secondly, I'm pretty sure the world's richest man has an easier time getting laid than someone who's posting on slashdot.
--
the strongest word is still the word "free"
Other reasions why 2002 was great:
Phoenix 0.5 - http://mozilla.org/
Chimera 0.6 - http://mozilla.org/
The Open CD - http://www.TheopenCD.org
GNU Win II - http://gnuwin.epfl.ch/en/index.html
yEnc - http://www.yenc.org/
Karma: The shiznight, mostly because I am the Drizzle.
To see how rapidly GNU's alternative to the Linux kernel is moving along, look at the Initial GNU HURD announcent in 1991 and the last announcement. Note the following phrase in the last announcement:"Popular PC devices are generally supported." What a fantastic 12 years it has been for GNU!
My step-brother (more than "computer literate" but not someone who likes to spend *all* his time tweaking his computers) told me at Thanksgiving that he's finally been mostly converted from Windows to Linux by Red Hat 8.0. He's a good example of how good 2002 has been, even with the crunch that some companies are going through.
... well, that is what would impress me most about the coming year in software. And I'd certainly count such an app being developed and made available via apt / apt4rpm ;)
;)
:) This is not hypothetical -- I'd like to start converting family videos to digital format, editing down to reasonable / watchable lengths. Right now, this means I'm thinking of spending more than I'd ideally want to on a large external drive for my iBook, just for that one reason.
Red Hat and Mandrake (which I name only because they seem to be the most visible in non-specialist stores) both produce distros which are relatively sane to install, and come with far (*far*) more extra-Operating System software included than the obvious conventional competitors do.
Which brings on an optimstic rant:
The included software with the usual distros varies a lot, by category and in quality. Saying that RH8 has x-jillion packages, though, is nice for exploring, but not helpful when the one thing a potential user would like to use is not among the x-jillion.
If Red Hat, or debian, or lycoris, or *any* distribution of [Linux + GNU utilities] were to come with an video editing app with ease of use approaching iMovie
I think that Windows now comes with a video editing app as well. When will plugging in a firewire video camera for dumping footage in order to do simple, cut-and-paste scene rearrangment be as easy as it is under Mac OS? Hats off to the developers of Cinelerra, Kino, etc, but what I'd like to see is something like "Cinelerra Lite" , or perhaps "Cinelerra Ultralight"
Such a beast would have to be simple, reliable, fast, pleasant, and with the ability to save to VCD/SVCD for DVD-player compatibility, and to DiVX;) or other free video format for long play. Wouldn't it be nice to have a complete toolkit for making low-budget video production using all free software?
This would also be a cool way to show off / play with the capabilities of Xiph's in-progress video format, eh? Eh, eh?!
(iMovie, though a well-made app and IMO sufficient reason for non Mac users to try out the Mac OS, does not make it easy -- and is it even possible? -- to create VCDs or DiVX;) disks.) I'd love for someone to point me to a tutorial indicating otherwise
All in all, thanks to the distro makers and application developers who have made GNU/Linux so much more accessible and friendly. Hope you have a good year!
timothy
jrnl: http://tinyurl.com/c2l8yr / foes: http://tinyurl.com/ckjno5
Microsoft has $40B in the bank and is still making money like they have a license to print it.
The Linux vendors have fallen to beg mode, "please give us money or we will vanish" (Mandrake). VA Linux Labs, now VA Software (stock symbol: LNUX) says "We are in no way a Linux company - we are a proprietary software company". Red Hat made a $300K profit last quarter, first ever, on a market cap of about $1B, what a complete joke.
Yes, yes, we have Apache, we have MySQL, we have numerous charity cases, but there is no way in hell that this has been a "great year" for Linux. If you can't make a buck, you can't eat, and sooner or later, you will stop breathing.
In the meantime, Borg-like entities like IBM (for Christ's sake) are adopting Linux (should I say "swallowing up Linux"?) and this is somehow a twisted victory for "the cause".
I want to throw up.
happened for Linux in 2002 was that our boss put himself into this situation:
PHB: "It's a well known fact that Linux is developed by a bunch of ameutars, a toy.", in a meeting with big Boss and many others, "There's no proof in saying that Windows server is unstable! Look at our file server, it hasn't had a single downtime since it started!"
another non-PHB: "but sir, but your staffs told me that it's actually a Linux running Samba service."
PHB: "Is it?!...."
He should have talked to us more.
to put a brick back into someone's pocket without their knowledge
You can't judge a book by the way it wears its hair.
But afterwards too. I've recently installed Mandrake 9.0 and it installs *seven* terminal programs. Seven? What on earth does the geekiest geek on God's green earth need *seven* terminal programs for?
Here's the deal, either you don't give a damn and will use whatever default shows up in your prefered enviroment, or you have a fave that you just can't live without for some reason and you'll manually install it from the CD anyway. If you're that picky you're sophisticated enough already to handle this.
If you're *not* that sophisticated the plethora of choices of terminal programs is at best confusing, and getting rid of the unwanted ones ( if you can even figure out which ones are unwanted, and why) may well be a somewhat daunting task.
Because free software is free as in beer to the distro makers they can throw in everything including seven "kitchen sinks," so they do. This doesn't mean it's a Good Idea.
I've got something of a rep as an Ubergeek in meatspace, but even I don't want a distro that just dumps the entire universe of software (including some pretty alpha stuff) on my HD just to prove it can.
Here's what I want to see in a default desktop install. A choice of KDE or Gnome ( I use a couple of others as well, but I'm perfectly content to install those seperately after I'm up and running for a bit), ONE terminal, preferably the default for the enviroment. ONE office package, preferably the default for the enviroment. A basic collection of utilities and, well, that's about it.
Clean, simple, and covering about 99.9% of all typical desktop funtions in one go, with no cruft.
For a newb throw in a special section in the manual explaining that one of the things free software is about is choice, how the CD's offer them many extras to play around with if they want, and clear, simple directions on how to install, and *UN*install, them.
Kinda like installing Windows, only better.
Installing a system should be an additive process, not like hacking away at a mighty oak with a chainsaw to release the inner OS.
Small is Beautiful.
KFG
"player 4 hit player 1 with 0 stroms"
However, I don't think it's true. First of all, owning a Zaurus myself, I find its use of QPE the biggest problem with the device--it means I can't use it for what I primarily want to use a Linux PDA for: running regular Linux software. Almost any software that uses a GUI needs to get ported. I can't script with my favorite scripting environments (Tcl/Tk, wxPython, fltk-lua), I can't use my favorite image display programs, etc.
Fortunately, the folks at handhelds.org have been working busily on putting together a high-quality X11-based handheld distribution. And the Opie versions of the Sharp/QPE applications have been recompiled for X11.
To me, Sharp will be a success story when it really does run the entire Linux environment: command line and graphical. Let's hope that in 2003, Sharp will base their Linux distribution on X11. Because of Qt/X11, the user experience and applications will remain unchanged (well, things may actually get a little faster with X11, but that's not going to be that important on a 400MHz XScale).
That means that palms finally broke the "33mhz barrier"
You can't judge a book by the way it wears its hair.
A commercially backed OS like MacOS or OS/2 or NextStep will die without corporate success, but Linux already has more developers working for it for free than MacOS or OS/2 or NextStep ever did for pay.
The only way Linux and its free software friends will ever die is if laws like the SSSCA are passed to make it illegal.
If we want Linux to be as slick as it should be, we can't expect beginners to have to run "make install" and "configure ./". Often these things don't work, and they are just so damn ugly.
So, to all developers: Please, please, please make all your software available in RPM (or .deb) format! Have tarred, gzipped sources as an extra for geeks by all means, but make package manager files the main format.
Software installation ticks of new UNIX users more than you realise.
What makes a man want to be a mouse? (Python's Flying Circus)
Yes, Gnome is sponsored commercially, as are many other projects. There is nothing wrong with commercial sponsorship.
There is a big difference, however: Troll Tech has a dual licensing model, and they claim that the ability to license their software commercially is necessary to finance their open source efforts. In contrast, when other companies sponsor open source efforts, the sponsors do not retain any special rights to the sponsored software.
That has numerous implications. For example, contributing to Qt and Gtk+ has very different implications and beneficiaries.
And with their dual licensing model, Troll Tech's motivations are different as well. On the Zaurus, if you want to develop commercial apps, you have no choice but to pay them; that is a very strong incentive for them to push Qt/Embedded over Qt/X11. If Troll Tech were sponsoring Qt like IBM is sponsoring the Linux kernel, they would have no special economic incentive to prefer Qt/Embedded over Qt/X11--their choice would be driven only by technical considerations. Those are the dangers of a Troll Tech-like model.
"If that is true, I think it really calls into question the entire open source effort." No it doesn't. Not in any way, shape or form. Open source and commercial can co-exist fine.
I think they can co-exist, like IBM and Linux do. Troll Tech is a different model, and I think it's dangerous. Sun, in fact, is trying to do something similar with Java, and I think it suffers from similar problems.
In other light I know of schools and many other instutions switching to MS bassed mail systems due to ease of maintence
Bwaaaaaaaaaahahahahah! Hah!
They'll be back once Exchange hoses up all their data. They will be back.
Just as a bit of info, here is why I use GNU/Linux in my articles covering Open Source/Free Software (including the one these comments are about). It's simply because I see GNU and Linux as two parts to our "success." If you take away GNU from Linux, you no longer have a working operating system. Likewise, if you take Linux away from GNU, you don't have a way to use all of your tools.
Thus, we don't just have GNU or Linux, to have a real competition for other operating systems, we depend on GNU and Linux (GNU/Linux). Why not celebrate both major core components to our wonderful operating software?
-Tim
-------------
"You would not get a high grade for such a design" -- Andy Tanenbaum on Linus' Linux design.
Moderators, please moderate the parent down. This is (at least) the third time that friendly_fire (whose Joyce-esque style -- or lack thereof -- is instantly recognisable to anyone who's spent time on the NYT Forums in the last two years) has posted this today.
friendly, plase grow a pair and create an account so I can add you to my Foes list.
Il n'y a pas de Planet B.
True.
:-)
There is some argument that it's a bit burdensome to say exactly what you mean anymore when specifying a group of OSes on Slashdot, though.
Let me give an example. I might say "Unix had a good year with mplayer", except someone would be sure to pop up and "Linux isn't Unix...it's a Unix-like OS". Okay, that we can deal with. Then we run the dilemma of whether "Unix-like OSes had a good year with mplayer" means "all operating systems that are like Unix, but are not, in fact, Unix", or whether it means "all operating systems that are or are not Unix, as long as they resemble Unix". You could say "Linux had a good year with GNOME 2", but then you get nailed by a FreeBSD guy that says that GNOME 2 works *fine* on FreeBSD as well, and another Stallmanite or Debianite who says that what you *really* mean to say is "GNU/Linux", not "Linux", which refers only to a kernel. "Red Hat Linux" would refer to a whole operating system, but only one distribution of such. Now, one must be sure not to say only "Unix and GNU/Linux have excellent text processing tools", as someone will be sure to mention that GNU/Hurd can handle those same tools as well, and is being shortchanged in your original comment. One could say that "Sun's make is only supported under Solaris" -- is that in fact true, or is it also supported under SunOS?
At this point, we may think we see a clever loophole. "Free operating systems achieved an enormous boom in the last year." However, that would be sure to get Stallmanites pointing out that you do not mean "Free", since BSD is not Free (or perhaps it is -- even with an almost 24/7 tech habit I can't keep up with what the FSF believes). Instead, perhaps you mean "free". Also, Red Hat may or may not be Free, based on their previous inclusion of Netscape Navigator. But you aren't talking about "free" operating systems -- that would include BeOS and Apple's System 6.0.8. You might change to "open source", whereupon you are informed that several companies consider their operating systems to be "open source" but only to some people or under some restrictions. Instead, you must mean "Open Source" operating systems. Well, even assuming you're familiar with ESR's exact rhetoric and can tell what falls under the "Open Source" moniker, at this point you're probably a bit bewildered.
I've reached the point where I just transpose the proper term, the one someone meant, whenever I see "Linux" or "Unix" or "open source operating systems" on Slashdot. It just isn't worth trying to be perfectly accurate, since a term to properly define the set you're talking about is probably at least two sentences long.
May we never see th
At least as far as the corporate desktop, people will use whatever the company they work for *tells* them to use..........and for that reason, and that reason alone, Linux can beome a serious desktop competitor without having to lower any conversion cost.....the (l)users will be TOLD to convert.
Microsoft can have the home game/personal finance/entertainment machine market...because if they only have that, they are going out of business because the price of a home PC is lowering to the point the "Microsoft Tax" is too much!
A few weeks ago I had a conversation with a salesperson in a Walmart-like store in a developing country which shall remain nameless. They were selling no-name machines with Linux installed; had 'em right up in the prime space at the front of the electronics area. About half the price of the cheapest XP machine (maybe US$300 including small CRT monitor).
He said that as far as he knew, most people just bought them to save money and then illegally installed W2K or XP over top. This seems plausible to me, as Linux awareness in said country was not high.
So maybe those Lindows sales can't all go in the Linux column.
"Patriotism is your conviction that this country is superior to all other countries because you were born in it." -- GBS
Comment removed based on user account deletion
The second big problem Linux faces is that its written by the OS-infatuated for the OS-infatuated. It very clearly lacks the "common touch". All I want is an OS that does what it's supposed to, then stays the hell out of my way. With Linux, I'm constantly tripping over piss-ant details and indiosyncratic quirks. The control is nice, and so is the ultimate reliability. But at what price? There's a line in one of the many HOW-TOs I've waded through that goes something like this:
God help us all!its ready. its just not ready for everybody.
I know you are psychotic, but please make an effort.
It takes the time to initialize graphics mode (like 0.1 secs to 2 secs, depends on your graphics device but it's the same for all graphical systems using the device, Windows included) plus a few milliseconds overhead even with old machines to get X up and running. Hardly slow for me, but perhaps you've got something better to offer.
What you're calling SLOW is the desktop environment running on top of X, GNOME or KDE or something. And I certainly can't argue with you that there wouldn't be lots of room for improvements here. But it's not fault of X, it's the bloat in applications.
So one thing that could improve all of Linux in terms of speed would be removal of GNOME and/or KDE, X is irrelevant here. Oh, wait a second. Actually that sentence doesn't make sense at all, you can't remove X or GNOME or KDE from Linux since it's not there. Linux is just the kernel. Let's try once more. One thing that could improve all of Linux distribution somedist in terms of speed would be removal of GNOME and/or KDE. But then again, I wouldn't use mydist if they took my GNOME desktop away.
Software installation can be a nightmare.
./configure and make install are simple and beautiful when they work, however lately when I have upgraded some core system software as well as installed some new projects I run into some problems.
/usr/lib/bleh.so. Or you upgrade the library and the upgrade is installed to /usr/local/lib but the older version is in /usr/lib or /lib.
./configure and make install with it working, you're an IDIOT. But this method is not perfect and the errors mentioned above do occur. Distributing binaries, NOT the best solution to the installation problems. The supported systems by linux are just too different. A binary compiled for your system is more efficient, additionally linkage problems can come into play with pre-compiled binaries. But with binaries, the real convenience in not providing them is for the poor developers. Maybe they do their code on a PowerPC or something, why should they have to go through the trouble of cross compiling for you. I will say one thing though, automake has certainly improved tremendously. A few years ago there were many things it didn't detect and it was necessary to go into the source code and make modifications to get MANY programs to work. Most of my installations once I have the correct libraries work perfectly.
./configure script which will check your system as it does now. Library not found comes up, you are presented with a set of information about the missing option required to get the program to function, and a list of options(Exit setup and install it yourself, Attempt to auto download and install the library online, Adjust path to the library, etc.). This way there is no web searching for the library, you will see the URL where you can get it in the info. Furthermore it could download and install it right away. If that library has depenencies they could be installed. This method would be extremely convenient, even more so than windows. And it shouldn't be too hard, except that authors of software would have to keep track of where they got the libs from and make sure their URLS are still valid. Additionally the test would give detailed information, not only on the missing package but why the test failed. Hacking an auto generated configure script is NOT FUN.
The project depends on other libraries which either
A) Are not mentioned in the software documentation at all, so need to look at messages and do some searches to find the libraries.
B) Are linked to on the web page but the link is bad so need to search them randomly, then most of the links from the search engine are bad, but eventually get to the right page.
C) And the worst is when there is an entire tree of dependencies, and most of them have to be installed in a specific order. Installing one package results in a few hours of compiling all the required libraries in the specified orders.
D) some of the libraries you need to install/upgrade are installed to directories other than the program searches. The libraries are in the path, but the program has hardcoded into it
Software installation can be a nightmare. It's just the way it is. In windows more often then not, when a program needs a library, the library is bundled with the program and installed by the installation program. Errors like required DLL's missing are less and less. This is because many of the windows programming environments include package programs to put all the required dll's with the software and to install them if they are not there. Many games that use additional programs(ie DirectX or what nto) will offer to install it if it is not already there. Whereas with Linux I must go cruising the web, downloading the libs I need, compiling them, then the software.
However, distributing the source code is a better option than the binaries in most cases, because of the multitude of systems linux works on. Assuming the program is written without machine dependent parts, if a binary is made for a different system than yours, you have to emulate it and it is much slower than if it was made in your native format. Furthermore code optimized for a 386 does not work as well on a Pentium IV as code optimized for a Pentium IV in many cases. Additionally in the past I used some software in binaries, and many libraries had newer versions made without regard to backwards compatibility, so that because my distribution was newer my old binaries stopped working. This hasn't happened for a long time, probably because libraries are mostly backwards compatible, but still there are no guarantees, so compiling the source and relinking libraries is the best option.
So to sum things up, if you are whining because you can't
A vision for the future:
Source code comes with a
But, even without these changes installation the way it is, is not too hard to learn. For some library heavy software, it can be a pain, but hang in there and you'll survive. Recently I installed a program which required me to update about 15 libraries, it was a pain, it wasn't mentally challenging, the program is installed and works now. If you stick in there you will succeed with your installations too. There is a higher learning curve than with windows since if you don't know what a directory is you'll have trouble in linux, whereas in windows pop cd's in and many autorun and install themselves. If that's the type of user you are, maybe linux is not for you.
But overall the free software programs have made leaps bounds. Autoconf is working better than ever, the compilers are faster than ever, configuring the linux kernel has gotten much simpler, Gnome and KDE are very convenient and come with a slew of applications that are well functional. Office software for Linux has made many advances since the days of Applixware. Even installing the linux kernel source has gotten simpler. make xconfig is a dream come true. The kernel this year has made tremendous advances. The new pre-empting package alone makes 2.5 worth downloading. ALSA support is absolutely wonderful, configuration was almost a breeze. Compare to the old days where I needed much more specific info and addresses for my sound cards, along with DOS driver hacks to initialize them and that is improvement. The community has had many great years in the past, and will continue to have them in the future!!!!!
like prounouncing Linux Lienux
I call him Lienus, I call his program Lienux
I don't say Linnus, so I don't say Linnux
If you wanted me to say it differently you should have spelled it properly.
As for alternatives, http://plan9.bell-labs.com/plan9 will do just fine tnx
There are places where the networks are not touching,and there are places where they are-Boeing's Lori Gunter
What can we look forward to this year? Off the top of my head:
Mmmm, toys :)
You see I'm from England and we talk a bit different.
So yes, I do say kill-o-meter
I'd prefer my 20 years in the computer industry to say more about me than how I pronounce the name of an obscure operating system.
and oh btw, the reason tunelling over HTTP works is because many desktops only have Internet access via an HTTP proxy rather than every desk in the company being NATed to the internet.
There are places where the networks are not touching,and there are places where they are-Boeing's Lori Gunter
I don't want to sound like a MS apologist, but we (the Linux community) really have no lesson to make to MS about server security this years. While 2001 was the year of shame for MS (Code Red|Blue|Red II, Nimba et cie), 2002 was a really bad year for OSS server software : OpenSSH exploit, Apache chunk encoding bug (Ramen), OpenSSL, etc.
:wq
Likewise, while it really offers only a few major advantages (and some disadvantages) over KDE's Konqueror, Mozilla 1.0 finally did arrive on the scene attracting attention from many mainstream sources. While its impact on the "browser wars" may be minimal, it does promise a real alternative to Internet Explorer on pretty much any platform.
I think the author downplayed the importance of Mozilla 1.0 :
For me, Mozilla 1.0 is THE event of the year 2002 for OSS.
:wq
No, it is not up to sysadmins or IT to make that choice, and it will not be me who forces anyone to use it. I am talking about more & more OWNERS & EXECUTIVES of companies making a choice not to use Microsoft's products (and they are the ones who have the RIGHT to do so), and their employees will use what system they are told to use. In a business setting the phrases "Linux is about freedom and choice" have absolutely no relevance whatever. A company exists to make money, not to give its employees the OS they would like.
The "cost" mentioned in original post was effort by end users to learn, not dollar cost. And "conversion" meant user switching, not the effort of translating data files, training, etc. Certainly monetary cost, data conversion, & new procedures need to be justified. I know of companies that are doing desktop change to Linux...they've done the justifications and are doing it. The C?0's are the ones driving it, and that's the only way it will happen in most traditional (non-computer software/hardware/serverices) corporations
I think perceived "harsh tone" mostly came from my use of "(l)user", which is only humorous way of referring to end user...I've been the end (l)user & (ab)user myself on systems adminned by others for many years. Also, when I do turbotax and photo work on my windows 98se box and it seizes up I am (l)user.
Hmm, and you think Windows development is an egoless place of peace and harmony where all 5000 developers agree and think in tune with each other? I can tell you, it's not. Ego abounds.
I have a Redhat/Gnome box. I daily suffer at least six different graphical file selection paradigms of various capabilities and presentations. One lets me scroll with the mouse wheel; another doesn't. One lists files in strict ASCII order; another ignores capitalization when sorting by name. And none approaches the ease and uniformity of the Windows standard.
6? I could probably get up to 3 if I tried hard and included really old apps like Motif Emacs (which you don't use file selectors in anyway) or RealPlayer. The GTK filepicker does suck, it's a known problem. When there is a better one, there will be 2 pickers, both of which are quite good. If somebody wanted to standardise common dialogs so KDE apps could be set to use GTK print/file dialogs and vice-versa, I guess that could be done, it's just a case of somebody sitting down and hammering out a spec.
But at least Microsoft has the luxury of dictating to their programmers that they SHALL adhere to a common look and feel.
Try telling that to the developers of Trillian, Roxio CD Creator, hell, even Office doesn't use the Windows look and feel. Apple is just as bad, they give their own apps completely made up UIs seemingly on a whim. QuickTime 4 had such an abomination for a GUI, that wasn't at all standard on any platform (not even the mac), it earned a place in the UI hall of shame. The idea that only Linux has fractured UI standards is a fallacy. It's a "problem" that affects all platforms, and really isn't all that much of a problem. The lightwave GUI is utterly non standard, but once you get used to it quite productive for instance. Better a productive GUI for a pro tool than a "standard" windows one that acts how you expect but is slow as molasses to use.
There's a line in one of the many HOW-TOs I've waded through that goes something like this:
I think what that means is that if you're an admin and you can't fix/understand this problem, you probably shouldn't be an admin. Arrogant attitude? Yeah, maybe. Probably true though.
no impunging taken
the real reason I persist in say Lienux is that for my accent Linnux is a foreign pronunciation which means my mouth muscles don't get much practice saying it.
Plus being corrected by someone says something about their attitude that might take a few weeks of working out by other linguistic means. Never underestimate your conversational partner.
Did you like the obscure bit? I liked that one. Can't get much more obscure than my dying BSD network. I'm sat here installing a FAMP setup into a 486 overdrive to see how it will cope.
NAT doesn't cache and makes it slightly tougher for malicious code to get route out. Not that I've done it myself, was just a snippet of knowledge from another mailing list.
okayokay enough already
There are places where the networks are not touching,and there are places where they are-Boeing's Lori Gunter
i imagine we could rip the sun desktop out but they still service the machine.
its on solaris 8.2
what are you running?
'There is a Light that never goes out.'
What I was *saying* is that more & more people will *have* to use Linux on the desktop, and not by their choice. And more & more "silly ignorant business types" as you put it are the ones who are/will push it. Maybe not the happiest situation, but the real live work world is like that. I didn't say it was good, just that it is happening
And I like OpenBSD much more than I like Linux.
Also, people may not know what is the *best* tool for their job if they haven't tried alternatives. The designers and drafters who worked for me found Pro/E on an UltraSparc with Solaris to be much better for the work they needed to do than running it on NT / Wintel, once they had a chance to try it. And people running MASS-11 word processing on a VAX found they liked the Mac IIci and WYSIWYG word processors much better, once they had a chance to try it, and so I had to requisition a couple dozen of those.