Slashdot Mirror

2 hours ago, the US and UK started an attack on Iraq. Under international law, this attack is illegal as it was not approved by the UN security council. In fact the UN security council has rejected repeated US requests to authorize it. Under international law, this is a war-crime. Under US law it is illegal, since Congress was not asked to approve it. (CT:Sengan is wrong here. Under Executive Order, Clinton can do this) The Pentagon estimates that 10,000 people will be killed. There are disturbing reports that Richard Butler, chairman of UNSCOM, has been in consultation with the US to draft his reports. Moreover, Cofi Annan, UN-head did not instruct Butler to evacuate the arms inspectors -- the US did. France, Russia and China have voiced opposition. Of relevance is that tomorrow, Clinton was to be impeached since most of the congress members wanted him impeached. Indeed, the Senate told the Whitehouse that he would be impeached tomorrow unless he bombed Iraq. Comments are disabled, as I expect a lot of people will believe the US/UK side of the story and will consider this story flame-bait or political. Lots of readers have submitted the story, so some of you are interested. I personally am disgusted at my country's behaviour (the UK) which always follows the US lead, because it believes it has some sort of special relationship with the US... yeah, right. Also, notice that the attack (5pm EST, 1am Iraq time) happened to be timed for prime-time US-TV, just like the Libya attack. Remember that Iraq was an ally of the US, and the US shipped Iraq weapons of mass-destruction, such as chemical weapons. Indeed, after the Gulf War, when the US was supposed to be enforcing a no-fly zone, Iraq used chemical weapons against the southern Shia Muslims.

Update: 12/16 08:53 by CT : I enabled comments. I didn't think it fair that a gag be placed here, especially not considering Sengan's fairly inflammatory statements. I won't bother harping on my political beliefs here, you guys can state your own. Feel free to refute the huge number of flaws in Sengans of writeup- including the fact that Legally a president can, under executive order declare military operations without congressional approval.

Update: 12/16 10:20 by CT : I'm getting a boatload of email about this thing, and I have no choice but to post this form response and to beg people to lay off until tomorrow at noon when I'll be done with my exams. I really don't wanna retake History next semester.

1. It was inappropriate for Sengan to disable comment posting and use Slashdot as his personal soapbox. I reenabled comment posting as soon as I noticed. I'm really sorry about this, it was inexcusable.
2. This story was inappropriate for Slashdot. We focus on technical news here, and geek humor, and free software. Not on international politics. That said, I give authors free reign to post stories on what they see appropriate. In this case, that free reign bit me in the ass.
3. The article has several mistakes, errors, and inaccuracies. The comments list many of them over and over again.
4. I really need to continue studying or I'm gonna bomb this exam, and I really don't want to retake a class next semester.
5. I'm sorry.

Update: 12/16 11:45 by S: I disabled comments in order to avoid /. going down. Not censorship. I apologize about that, since apparently /. can now support the load. The news source for the above is Pacifica Radio's Iraq coverage, and in particular Prof Chomsky. I'm sorry for the spelling mistakes. This morning Pacifica claimed it was illegal again according to US law, but I am no lawyer. However, the US did ask for the right to take military action in November, and the Security Council refused them. Why ask if they did not need it? I got a lot of email about this, about a 1/3 supportive, mainly non-US, and 2/3s against, mainly US. To the extent that I believe democracy is about questioning government motives, something I have not seen on US-media, I do not apologize for this post. I do think it is stuff that matters, and if I were not ill I would have come up with better sources. I am sorry though, about the reaction it caused. Further info is in Liberation and Democracy Now. World reaction is not all favorable.

ikky writes "This problably falls into the "We've-had-enough-of-Java-in-the-past-few-days", but here we go anyways. Seems like SUN has decided that the new JDK (officially launched today) will be named Java 2. Also the link provides some insight into SUN's OSS plans for Java. Great news for OSS! And it comes just after the Jikes release... " As an aside, thanks to Jason Y. Sproul for our spiffy new Java Icon. We've been running with that crappy blob thing ever since that whole Invaders/Java thing quite some time ago. Update! S: Here's Tim O'reilly's (and others) reaction(s) to the opening (but not open-sourcing) of Java.

Courtesy the uberinformed David Lesher. Sun has won their Java suit against Microsoft. The most obvious ramification is that Microsoft must stop shipping Windows 98 and IE immediatley. Hopefully, more information will come in later. H:For those of you who don't want to go to the NY Times, read about the victory over here. J: To clarify - my original post was very poorly worded.They did not win in the legal sense, but they did win in the sense they have succeeded in making MS either comply or lose. S: Hey you're all famous! we're mentionned on news.com.

Christian Reiniger of the new Linux Game Development Project has written up a nice piece that you might want to read if you want to see more games on Linux, and how this new project will aid that. The way I see it, the apps are coming, and in many cases, already here. We just need the games. The following was written by Slashdot Reader Christian Reiniger The Linux Game Development Center Rationale

Linux is gaining much attention these days. People who were anti-Linux for a long time suddenly discover that it has changed much the past few years, ultraconservative magazines feature positive stories about Linux at prominent places and The Big Ones in the computer business are almost crowding to support the former "hacker OS".

Good press is always welcome - but can Linux live up to its new image? Can it avoid to dissapoint the people finally giving it a try?

Well, the "It doesn't have a nice, easy to use desktop" and "There are no applications for it" arguments are vanishing in a puff of colorful smoke and the "It's too hard to install" problem is quietly dissolving. But there's still that nasty "But I can't play my favourite games in Linux!" thing.

Linux has games. Linux has good games. But that other operating system has several orders of magnitude more good games than Linux. That's bad. And difficult to overcome, as it's not only because of technical reasons. But we, the free software community, have have a long history of solving But we, the free software community, have have a long history of solving problems and shipping around obstacles. There is no reason why we should not be able to solve this issue, too.

So what's the current situation, what needs to be done and what can be done? Here is a short overview of the major issues:

• Despite Linux's rapid growth - both in terms of user base and existing software - it still is not generally perceived as viable platform for high quality games. Some of the often cited problems are without doubt true, but most of these are already at the verge of being solved and the others mainly need more public discussion.
• While many game-related SDKs and applications exist or are in the make, there is no comprehensive overview of them available.
• As all of these SDKs have their strengths and weaknesses, much can be gained by making them as modular and interoperable as possible, so that game developers can combine them to an almost optimal solution.
• For both commercial game developers wanting to port games to Linux and yet-inexperienced Open Source® developers aspiring to write free games, easy to read documentation and online help via mailing lists and/or irc are very valuable.

Quick Summary

In essence we are suggesting that this new Linux Game Development Center be a kind of meta-project. It would be dedicated to advocating Linux as gaming platform, collecting knowledge about Linux game development and using it to help all interested people, providing facilities for discussion to Linux game developers and, last but not least, encouraging and helping existing free (Open Source®) game SDK projects coordinate with one another.

Please note that this is not an attempt to impose standards or rules on anyone. We just want to do what we can to help everybody coordinate their project with the others and to encourage all game SDK developers to develop compatible libraries.

This is also a call for developers, users and game SDK projects to join our efforts.

History

In the beginning ... there were many unrelated games SDK projects started by many different groups with little or no inter-group communication or coordination.

The initial initiative of starting the Linux Game Development site came from Ian Crawford (you can read his announcement of the site here).

It was first meant as a meeting and coordination point for people developing native and free Linux games, but its scope was soon widened to support Linux game development in general - the phrase "This site aspires to be the headquarters for all Linux game development" is from that time.

Cut - Switch to the PenguinPlay mailing list. Shortly after Ian's announcement of the site, Sam Lantiga suggested on the PenguinPlay mailing list that people could get together on IRC to discuss the future of Linux game development. His idea was considered as "really good" and after the first meeting the thing was extended to all people involved in pushing game development for Linux. Here are the archives of past meetings and the plans for future ones.

Well, the irc meetings became a regular event (each Saturday) and the possibility to have a real-time discussion through irc gave a big push to our work. We started discussing on how we could coordinate our efforts better, how to make Linux more appealing to professional game developers etc. After a few meetings we came to the conclusion that it would be best to merge the SDK projects (ClanLib, CrystalSpace, GAMES and PenguinPlay) to one, giving it the full support. It seemed to be the right thing, but we were a bit uneasy with it, as merging projects is a very, very difficult task.

Then Charles Durst threw in an proposal for a clearing house project, i.e. a project that would give developers from different game SDK projects a good way to communicate with each other, remind these developers to keep the different SDKs compatible to each other etc. He first proposed that PenguinPlay could become this "meta-project", but we found Ian Crawford's "Linux Game Development Center" much more fitting.

We started working on the homepage for this and Charles wrote an announcement text we wanted to post on Slashdot or Freshmeat and several newsgroups. However, as we assembled material for the homepage, discussed its structure etc it slowly mutated from the "Linux Game SDK Coordination Center" to a site for Linux game development in general - the "Linux Game Development Center" or LGDC for short. Ian's original site laid the foundation for this (as it was aimed at helping people to develop actual games) and the transformation was completed when the "Linux Game Breeding (LGB)" (aimed at creation of new projects around Linux GameDev) and "Linux Gaming Awareness (LGA)" (aimed at advocating Linux to commercial game developers) projects joined in.

So here we are. The Linux Game Development Center is a project from Open Source® game developers, maintained by them and dedicated to all people interested in the subject. Located at www.linuxgames.org, it serves as a sister site to www.linuxgames.com, the already well-established site targeted towards game players.

The Proposal

The new Linux Game Development Center would:

• Maintain a collection of links to various game SDK projects and a "news page" of the current status and functionality of each.
• Help coordinate efforts to increase compatibility and perhaps create "glue" software between the libraries produced by different game SDK projects.
• Help game SDK developers coordinate with one another (via mailing lists and perhaps IRC get-togethers), and share algorithms and code. This could even help SDK developers abstract out new layers of common or overlapping functionality between projects.
• Help to fill the functionality gaps that are currently preventing any combination of game SDK libraries from being comprehensive enough for many professional game developers to use.
• Help to direct game developers to the right tools for their particular tasks. Making it easy to find software for a particular purpose, within certain platform, language or license requirements. We are considering using existing web-based knowledge base tools such as WikiWikiWeb or faq-o-matic, as well as tables of the features and limitations of each available package.
• Collect the general feedback that game developers might want to give the Linux community about any porting problems they might have. And helping to start, extend or fix projects to meet those needs.
• If neccessary initiate and host "please port this to Linux" petitions and mane the commercial game developers aware of the demand.
• Find volunteers willing to port commercial games to Linux and act as mediator between them and commercial game houses.
• Provide facilities for discussion between commercial game developers and Linux users on how support for Linux can be increased in the future.
• Help rally game SDK development efforts to port existing game libraries to needed, unsupported platforms.
• It could help direct interested people to other projects as needed to help with bugfixing, porting, and documentation (especially with respect to interoperability between projects).
• It could even have a relationship to game SDK projects and Open Source® games somewhat similar to the relationship Debian has with the packages that it collects. It could collect easy-to-find and easy-to-install packages of game SDKs and try to make it easy for a new developer to choose the one(s) that best meets their needs. It could even help develop policies to ensure clean interaction between libraries wanting to be added to the collection.

While game development for Linux would be an important goal of the web site, the most important goal would be the development of quality cross-platform game libraries. For that reason, developers of games and game SDKs for platforms other than Linux would be more than welcome to join us. Especially if they are interested in porting software to or from Linux.

In the end, there would still be multiple, competing game SDK packages, but that should be OK as long as at least one comprehensive open-source solution can be cobbled together from the pieces. As we have seen with multiple distributions, and even the KDE/GNOME projects, competition can sometimes be a very good thing ... if you can see past the flame wars.

The biggest problem with having multiple, competing projects is the resultant (developer and user) confusion. What we are proposing is a Linux Game Development Center that is aimed simply at reducing that confusion by helping people to find, evaluate, combine and use the available tools, or to develop new, missing ones.

Request

At this point, we are mainly looking for:

• More people to work on the web-site (in particular people who have ideas for ways we should do it with existing or new web server and/or database technologies).
• Other game SDK related projects that should be added, or who want to help, or who should at least join the linuxgames mailing list(s).
• Other Game or Game SDK developers who want to be in on the discussions, prioritizing, development, or who just want to influence the direction of the Linux Games project in one way or another.

All interested people are invited to join the linuxgames mailing list and participate in the discussions (send a blank message to linuxgames-subscribe@sunsite.auc.dk)

Current Linux Game Development Projects

These are the current Linux Game Development projects we have been able to locate and invite to participate. If your favorite project is not included, let us know and please join us.

• 3dfx HowTo
• ALSA - Advanced Linux Sound Architecture
• ClanLib
• CrystalSpace
• Daryll Strauss' Linux 3D page
• DUMB
• GAMES - GNU Animation Multimedia Entertain ment System
• GGI - General Graphics Interface
• GSI - General Sound Interface
• Linux game development webring
• Linux Game Programming HowTo
• Linux Game Programming Megasite
• Linux Game Tome
• LinuxGames.Com
• Mesa
• MGL
• PenguinPlay
• SDL - Simple DirectMedia Layer

Charles Durst, Christian Reiniger

Emmanuel Marty writes "Apparently Robert Collins' famous "Intel secrets" website (www.x86.org) is unavailable as his domain is on hold, according to whois. I hope it gets back up soon."

belial writes "The Chicago Tribune has an interesting piece on 'buying into' Microsoft Standards." S: This is of particular relevance given the Halloween Paper's suggestion of "enhancing standards" to block competition. This should make it a much more powerful a statement to any drones out there.

Found in LWN. Apparently, Microsoft has begun its dissection of the open source community. Eric Raymond has been leaked an internal Microsoft memo regarding open source software. Quite the interesting read. Turns out they even mention Slashdot. This could be beneficial to us in another way, as well: it points out, in no uncertain words, our deficiencies. Consider it the master TODO list, guys. Update!(S) Vorx writes "ZDNet just posted an article about the MS Halloween document. They mention ESR and his role in publicizing it. Seems like more attention is being paid to this article now, maybe some PHB's will start to wonder about MS and Fear :)" Update! (S) David Rysdam wrote to tell us: "I wrote to ESR myself and told him about the brouhaha over here at Slashdot over the "Halloween Paper". I explained that many thought it was a hoax perpetrated on or by him, but that I thought even if he was the hoaxster I didn't believe he'd keep up an outright lie. So I asked him straight: "To the best of your knowledge, is this memo what it purports to be: a leaked, internal, MS communication?" He responded: "Absolutely. I believe it is authentic." So I think we can drop the "ESR is lying" hypothesis." Also, Paul Victor Novarese has mirrored the Halloween Document (with ESR's permission).

Found in LWN. Apparently, Microsoft has begun its dissection of the open source community. Eric Raymond has been leaked an internal Microsoft memo regarding open source software. Quite the interesting read. Turns out they even mention Slashdot. This could be beneficial to us in another way, as well: it points out, in no uncertain words, our deficiencies. Consider it the master TODO list, guys. Update!(S) Vorx writes "ZDNet just posted an article about the MS Halloween document. They mention ESR and his role in publicizing it. Seems like more attention is being paid to this article now, maybe some PHB's will start to wonder about MS and Fear :)" Update! (S) David Rysdam wrote to tell us: "I wrote to ESR myself and told him about the brouhaha over here at Slashdot over the "Halloween Paper". I explained that many thought it was a hoax perpetrated on or by him, but that I thought even if he was the hoaxster I didn't believe he'd keep up an outright lie. So I asked him straight: "To the best of your knowledge, is this memo what it purports to be: a leaked, internal, MS communication?" He responded: "Absolutely. I believe it is authentic." So I think we can drop the "ESR is lying" hypothesis." Also, Paul Victor Novarese has mirrored the Halloween Document (with ESR's permission).

frokost writes "There is an editorial at freshmeat written by Jim Gettys, one of the original authors of X. Very interesting, indeed." S: Jim's idea of cross-toolkit themes should be extended to Wine: many companies will use Wine to port to Linux, simply because there are a lot of Win32 capable programmers out there. Wine will therefore be one of the important GUI toolkits out there. I also agree with his point that WIMP is step 1. Voice recognition is step 2. Right now, Linux GUI standards are in very rapid development, which makes it the right time for a voice recognition standard to be built and added so that apps may be designed/reworked early to include voice-recognition. Some work similar to what he's talking about has been done by IBM, such as a browser to help blind users surf the web.

In a U-turn that will be welcomed by all members of the Linux community who still have to use a Windows box to access Lotus Notes at work, Lotus has announced that it is looking into porting its SmartSuite and Notes/Domino suites of applications to Linux. It would also make SmartSuite open-source. This link was found on Linux Today. Unlike the analysts of the story, I believe this is a low-cost (type make) move which will be very lucrative to Lotus (many Linux users in Fortune 500 companies, have 2 PCs, one for the company standards of Notes/Office (Windows) and one to do real work on (Linux). For those that do not know SmartSuite, it's a pretty awesome suite of office applications, including Voice Recognition -- I really hope this too will be open-sourced: Lotus Organizer time and contact manager, Lotus 1-2-3 spreadsheet, Lotus Freelance Graphics presentation graphics, Lotus Word Pro word processor, Lotus Approach database, Lotus ScreenCam multimedia, IBM ViaVoice for Word Pro and 1-2-3, Lotus Mail Internet e-mail. update: "will port" to "looking into porting". However, given that the Strategy vice president said it was just a question of a recompile, if enough people want it it should happen.

Justin Bradford writes "Corel announced on comp.emulators.ms-windows.wine that they plan to use WineLib (the library side of WINE, the open source Windows emulator) to port all of their products. Here's what Gavriel State (gavriels@corel.com), Linux porting team leader, had to say: "We are actively committing engineering resources to the Wine project. Once our exploration and evaluation has settled down a bit (read: we've cleaned up our hacks), we'll begin submitting patches to Alexandre to review and commit to the CVS tree. We'll be spending some effort on getting the WordPerfect Suite 8 applications running under Wine, but our primary focus will be on porting ou WordPerfect Suite 9 applications using WineLib. We'll also begin participating publicly on the WINE newsgroup and mailing lists, so that we can contribute our ideas as well as our code. You should be seeing a lot more of us in the future."" S: What is of particular interest to me is that they have already coded a Win32-on-MacOS library when they ported CorelDraw to the Mac. This means Corel has very valuable expertise in the Win32-emulation area.

echo writes "On C-NET's Project Heresy yesterday Dan Shafer and Brian Cooley talked about where they think Linux needs to go in the future. In particular, they seem to think that the command line should be totally hidden, that the notion of multi-user should go away, and that programs shouldn't have "modes". At least for "regular" users. Here's the site. What do you think? " I agree that these things should be hidden from the novices, but making multi user go away? Yeah, right. update: (S) Apparently the guys at Project Heresy tried to contribute to the discussion but failed. This is what they had to say.

Clark Shishido writes " MacWeek has rumors on Gateway's plans for AmigaOS 4.0. sounds like the good old days when Amigas could run Mac software faster than a Mac. Also note the irony that it might boot BeOS which is Gasse's Mac-Amiga succesor. Maybe this will knock some sense into Apple that it has to innovate and compete instead of killing clones. I'm not an Amiga head, and I'll probably be editting makefiles for compiling under MacOS X, but I think that it's good that Apple can't be complacent. " Update (S) Apparently this information may not have been completely true, but a practical joke. Oh well.

Vidar Hokstad writes "Apparently, IBM will team up with SCO and Sequent to create a new Unix for IBMs entry-level to large enterprise servers. I expect that means mostly hardware that runs Aix or AS/400 today, but why don't they just port Linux or *BSD instead? This article at news.com has more details" S: IBM's website has further details on this Project Monterey, in which they spell out that the partnership is to deliver only one variant of Unix running on PowerPC, IA32 and IA64 platforms. Can anyone make any sense of this? A cynical first analysis would be that IBM has realised that there is still money to be made in Unix in particular by using its position of a single large vendor to reduce support-FUD; IBM would use SCO to gain market share very quickly while SCO benefits from a shield from Linux' penetration into the "enterprise market". I guess I just don't get what technical reason there could be to yet another variant of Unix. Do you?

Andy Tai writes "Printer makers are working together to define a single universal printer driver. Called the Universal Printer Driver Format, it will replace the 3000 printer drivers in existence. UPDF will allow the characteristic of each printer to be described in a text file, residing in the printer itself." S: This effort is 2 years old (not related to Intel's initiative) if you look on their website. It looks pretty open from their IPP pages, including specs and apparently joining is free. Looks to me like an area where Linux participation may be beneficial for Linux and the PWG by providing early widespread testing (if printer manufacturers release descriptions for existing printers in file format before actually including them in their newer printers).

Microsoft is not going to wait for the outcome of its trial with the DOJ before attacking Linux. The battle scene is France, where Microsoft's new (expendable?) regional director Marc Chardon has just issued an open letter to his clients. Click below to read the translation of the Linux-section (it's in French) and some commentary. The new director of Microsoft France (MF) has just issued an open letter to his clients. Most of the letter says essentially "We're right. Critics aren't. Millions of people use our products." and other fascinating insights. However it does contain two interesting revelations: Microsoft's OS is an integral part of your PC, and Linux is of very limited interest to anyone but fringe groups (students and researchers).

As previously reported, it is illegal in the EU to tie the sale of a product with another in the EU. However Microsoft France (MF) argues that because a computer can be argued non-functional without an operating system (or a hard-drive), what is being sold is a single product, not two. A little later, MF's director contradicts himself by stating that "assemblers" (basically small computing shops that will assemble a computer from standard components according to your desires) will sell you a computer without an OS -- is that a broken computer? Indeed, since the same letter claims there are more OS's for PCs than any other computer, we must be dealing with a product tied to another precisely because no other OS is sold (according to the letter) with a computer.

Further down, you'll see a whole section devoted to Linux. While today Microsoft's lawyers declared Linux to be developed by a single person, MF's director claims it to be a movement. I've translated the rest of what he had to say about Linux since it's interesting...

"It would seem that Linux does not satisfy the requirements of most companies, let alone the general public.

Linux presents limits that will slow down its widespread distribution, particularly in companies and the general public

In terms of the system's stability, Linux has problems of general coordination, and one feels strongly the lack of a project leader. Linus Torvald ( S: Microsoft can't spell!) left university last year to join a Californian company. The development of Linux has since considerably slowed down. Similarly, the maintenance of Linux' functionality depends on the mobilization of its teams. Thus, certain of its functions have not been updated in the last two years.

The installation of Linux is delicate. For optimal system performance, each version must be tuned on each computer at each installation, by a competent computer-technician. Setting it up and its administration are therefore not within the reach of a computer-technician used to more friendly operating systems, let alone basic users. Using Linux is complex, its programs generally take text commands.

Finally most Linux application software has limited functionality. Word processors rarely have functionality common on today's PC or Macintosh: dynamic spelling correction, graphic input of tables, integration of imported graphics. Most Linux word processors bear more resemblance to Microsoft Write written in 1985.

Linux' advantages of zero-cost and open source are not relevant criteria for most users.

The zero-cost of Linux is a non decisive advantage: the cost of an OS is minor in comparison to the other costs of a company. The price of the OS is only one of the elements of computers in companies. Putting a traditional OS into place costs many times the price of the OS itself, and the same goes for the creation of an application program and its maintenance. So, by installing Linux, one saves the cost of the OS, but one increases the cost of installation, and one takes risks for the maintenance of the applications and the system itself.

If certain people consider the permanent availability of source code to be an absolute guaranty of independence from software editors, we fail to see the benefit for a company or a person to have access to the source of his OS.

However, the distribution of the OS source code is very useful for students and researchers, to understand the inner-workings of the OS and eventually to change it. Linux will therefore probably stay for a long time a good subject of study for computer-scientist, rather than an OS destined for widespread distribution. "

I must apologize for any mistakes in my translation. Use babelfish to get an alternative translation. I'm open to any corrections. update! Linux Weekly News has just published a de-babelfished translation of the whole letter.

S: This letter obviously contains many "inaccuracies":

If I buy a car, is the fuel a separate product or not? What about electricity? Computers are non-functional without it. I hope the EU lawyers will notice this wriggling.

His attacks on Linus stating that Linux development has slowed down are not credible with the inflow of new developers and the number of ports now in the standard kernel. The letter very much ignores the fact that Transmeta lets Linus hack on Linux during office hours and has other employees that contribute, underplays Red Hat employing kernel developers, and completely ignores the current tidal surge of major corporations towards Linux: Intel, Compaq, Oracle, Sun, to name but a few.

Anyone who has had to install Windows from scratch, as I do every 3-6 months when it has corrupted its hard-drive beyond repair at work, knows that installing Windows is a royal pain. It takes over an hour, requires minding (yes... I'm just here to click OK)... Linux takes me 20 minutes on an 100 Mhz system. The only argument here is: when Windows is pre-installed (i.e. when you buy your machine) it's easier than it is to install Linux. Duh!

I, and many others don't view GUI's as necessarily friendly. Until they are a substitute for true understanding, I prefer to have control over my system and to be able to repair it. The usual answer for Windows to reinstall everything, and then try eliminating various components until you've found the "culprit". The same applies for source code. Funny that the latest C'T has devoted 19 pages to "Hacks & Bugs & Workarounds: Large Projects with Word, and how one survives them". To me, and many others, this is an unacceptable hit on my productivity.

Indeed, the whole notion that Linux is too hard for the average user is nonsense to me. I gave my mother a Linux box -- I'm living 8 timezones away so I cannot help her fix an unreliable OS. She cannot rely on computer-savvy neighbours either, since she's in a very rural area. But, with Linux as her first computer, she is happy using it laying ridicule on Microsoft's claims about the difficulties that the average layman will experience.

Complaints about Word processors are unfair since most Windows word-processors are also not very advanced. Percentage-wise (if you count all the free, shareware, and old ones), I expect Windows/DOS have a worse ratio. Only a few products provide the features Marc discusses. Similarly, on Linux, we have WordPerfect 7 (hey Corel, port WordPerfect 8!), Applixware, and StarOffice (which I sometimes use), Angoss, Dtop, and Axene's Xclamation, On the free front we have Emacs which is also used by a very large number of people under NT and which will soon have a WYSIWYG interface, Thot, EZ, Papyrus, Cicero, Doc, Maxwell, and new promising upstarts such as Glue. And let's not forget TeX: I and my fellow PhD students wrote their theses in it because it copes well with 700 page documents. Most academic papers must be written, and many books are written in it. It also accepts any graphics as encapsulated postscript. TeX is still the only format which is guaranteed to come out looking the same on any computer, and still looks better to me and many others than the output of any other product. As to dynamic spell-checking, I turn it off: I think, I write, I reread, I spell-check. Dynamic spell-checking just breaks the flow of my thoughts.

The attack on zero-cost software is a pretty obvious diversion, and tries to draw the reader's attention away from the fact people use Linux because of its stability and features rather than its cost.

Finally, Microsoft's attempt to make academics and students irrelevant is interesting, since they are the ones pushing Linux, but also very dangerous. France values intelligence and high education more than most other societies, as Marc Chardon's own CV shows.

So what do you think of all this?

I'd like to thank A Dark Elf, Jacky Liu, and Linux Weekly News Daily for some of the material I used here.

Microsoft is not going to wait for the outcome of its trial with the DOJ before attacking Linux. The battle scene is France, where Microsoft's new (expendable?) regional director Marc Chardon has just issued an open letter to his clients. Click below to read the translation of the Linux-section (it's in French) and some commentary. The new director of Microsoft France (MF) has just issued an open letter to his clients. Most of the letter says essentially "We're right. Critics aren't. Millions of people use our products." and other fascinating insights. However it does contain two interesting revelations: Microsoft's OS is an integral part of your PC, and Linux is of very limited interest to anyone but fringe groups (students and researchers).

As previously reported, it is illegal in the EU to tie the sale of a product with another in the EU. However Microsoft France (MF) argues that because a computer can be argued non-functional without an operating system (or a hard-drive), what is being sold is a single product, not two. A little later, MF's director contradicts himself by stating that "assemblers" (basically small computing shops that will assemble a computer from standard components according to your desires) will sell you a computer without an OS -- is that a broken computer? Indeed, since the same letter claims there are more OS's for PCs than any other computer, we must be dealing with a product tied to another precisely because no other OS is sold (according to the letter) with a computer.

Further down, you'll see a whole section devoted to Linux. While today Microsoft's lawyers declared Linux to be developed by a single person, MF's director claims it to be a movement. I've translated the rest of what he had to say about Linux since it's interesting...

"It would seem that Linux does not satisfy the requirements of most companies, let alone the general public.

Linux presents limits that will slow down its widespread distribution, particularly in companies and the general public

In terms of the system's stability, Linux has problems of general coordination, and one feels strongly the lack of a project leader. Linus Torvald ( S: Microsoft can't spell!) left university last year to join a Californian company. The development of Linux has since considerably slowed down. Similarly, the maintenance of Linux' functionality depends on the mobilization of its teams. Thus, certain of its functions have not been updated in the last two years.

The installation of Linux is delicate. For optimal system performance, each version must be tuned on each computer at each installation, by a competent computer-technician. Setting it up and its administration are therefore not within the reach of a computer-technician used to more friendly operating systems, let alone basic users. Using Linux is complex, its programs generally take text commands.

Finally most Linux application software has limited functionality. Word processors rarely have functionality common on today's PC or Macintosh: dynamic spelling correction, graphic input of tables, integration of imported graphics. Most Linux word processors bear more resemblance to Microsoft Write written in 1985.

Linux' advantages of zero-cost and open source are not relevant criteria for most users.

The zero-cost of Linux is a non decisive advantage: the cost of an OS is minor in comparison to the other costs of a company. The price of the OS is only one of the elements of computers in companies. Putting a traditional OS into place costs many times the price of the OS itself, and the same goes for the creation of an application program and its maintenance. So, by installing Linux, one saves the cost of the OS, but one increases the cost of installation, and one takes risks for the maintenance of the applications and the system itself.

If certain people consider the permanent availability of source code to be an absolute guaranty of independence from software editors, we fail to see the benefit for a company or a person to have access to the source of his OS.

However, the distribution of the OS source code is very useful for students and researchers, to understand the inner-workings of the OS and eventually to change it. Linux will therefore probably stay for a long time a good subject of study for computer-scientist, rather than an OS destined for widespread distribution. "

I must apologize for any mistakes in my translation. Use babelfish to get an alternative translation. I'm open to any corrections. update! Linux Weekly News has just published a de-babelfished translation of the whole letter.

S: This letter obviously contains many "inaccuracies":

If I buy a car, is the fuel a separate product or not? What about electricity? Computers are non-functional without it. I hope the EU lawyers will notice this wriggling.

His attacks on Linus stating that Linux development has slowed down are not credible with the inflow of new developers and the number of ports now in the standard kernel. The letter very much ignores the fact that Transmeta lets Linus hack on Linux during office hours and has other employees that contribute, underplays Red Hat employing kernel developers, and completely ignores the current tidal surge of major corporations towards Linux: Intel, Compaq, Oracle, Sun, to name but a few.

Anyone who has had to install Windows from scratch, as I do every 3-6 months when it has corrupted its hard-drive beyond repair at work, knows that installing Windows is a royal pain. It takes over an hour, requires minding (yes... I'm just here to click OK)... Linux takes me 20 minutes on an 100 Mhz system. The only argument here is: when Windows is pre-installed (i.e. when you buy your machine) it's easier than it is to install Linux. Duh!

I, and many others don't view GUI's as necessarily friendly. Until they are a substitute for true understanding, I prefer to have control over my system and to be able to repair it. The usual answer for Windows to reinstall everything, and then try eliminating various components until you've found the "culprit". The same applies for source code. Funny that the latest C'T has devoted 19 pages to "Hacks & Bugs & Workarounds: Large Projects with Word, and how one survives them". To me, and many others, this is an unacceptable hit on my productivity.

Indeed, the whole notion that Linux is too hard for the average user is nonsense to me. I gave my mother a Linux box -- I'm living 8 timezones away so I cannot help her fix an unreliable OS. She cannot rely on computer-savvy neighbours either, since she's in a very rural area. But, with Linux as her first computer, she is happy using it laying ridicule on Microsoft's claims about the difficulties that the average layman will experience.

Complaints about Word processors are unfair since most Windows word-processors are also not very advanced. Percentage-wise (if you count all the free, shareware, and old ones), I expect Windows/DOS have a worse ratio. Only a few products provide the features Marc discusses. Similarly, on Linux, we have WordPerfect 7 (hey Corel, port WordPerfect 8!), Applixware, and StarOffice (which I sometimes use), Angoss, Dtop, and Axene's Xclamation, On the free front we have Emacs which is also used by a very large number of people under NT and which will soon have a WYSIWYG interface, Thot, EZ, Papyrus, Cicero, Doc, Maxwell, and new promising upstarts such as Glue. And let's not forget TeX: I and my fellow PhD students wrote their theses in it because it copes well with 700 page documents. Most academic papers must be written, and many books are written in it. It also accepts any graphics as encapsulated postscript. TeX is still the only format which is guaranteed to come out looking the same on any computer, and still looks better to me and many others than the output of any other product. As to dynamic spell-checking, I turn it off: I think, I write, I reread, I spell-check. Dynamic spell-checking just breaks the flow of my thoughts.

The attack on zero-cost software is a pretty obvious diversion, and tries to draw the reader's attention away from the fact people use Linux because of its stability and features rather than its cost.

Finally, Microsoft's attempt to make academics and students irrelevant is interesting, since they are the ones pushing Linux, but also very dangerous. France values intelligence and high education more than most other societies, as Marc Chardon's own CV shows.

So what do you think of all this?

I'd like to thank A Dark Elf, Jacky Liu, and Linux Weekly News Daily for some of the material I used here.

My eagerly awaited copy of C'T has arrived, and in it there is a short interview with Martin Sieler, multimedia software lead at the Fraunhofer Institute. The topic was Fraunhofer's demand for patent license fees on free MP3 encoders. In it Sieler disputes that the encoders were free because the internet sites distributing them made money on the banner advertisements. He also discusses the newest MPEG standard (MPEG-2 Advanced Audio Coding -- also to be in MPEG-4): a refinement of the MPEG-1 layer 3 technology (MP3), it will provide even higher compression rates for the same audio quality. As with MP3 unlicensed encoders will be illegal, but unlike MP3 no free decoders will be tolerated... S: I find the ISO's current trend of accepting standards which cannot be implemented without violating patents worrisome. While I agree with some readers that a lot of work goes into this type of research I disagree that the key elements are inventions: they are discoveries about how human perception works. As a result, an unrestricted alternative (like gzip was to pkzip) should not only be possible, but encouraged. What do you think? I've corrected MPEG-3 to MPEG-1 Layer 3. Thanks manuka for your correction. I've also corrected the "like gzip" statement to "like gzip was to pkzip" since it seems to be causing some confusion. Finally thanks to Christian who pointed out I forgot the u in Fraunhofer. Oops.

William Hood writes " I wanted to submit this story from CNN about the tendency for IT workers to wind up putting in 60 hour work weeks, that are paid as only 40. This article was enough for me to wonder if it's time to finally bring unions into the tech industry. It is well known that most of the people who work at Microsoft essentially have no life outside of work. I heard a rumor that Bill Gates and friends once called a large meeting on Easter Sunday. While it is true that some techies may put in an entire month of 16 hour days to put some video game on the web to show their friends (I did this once), it is unreasonable that employers expect this of us all the time. We don't need the divorce rate amoung IT professionals to keep going up." S: I notice that my productivity falls and my bug-rate goes up if I overwork. I end up spending even more time trying to fix the bugs I've just introduced -- a nasty vicious cycle. What about you? Also do you think unions would improve matters?

CMiYC writes " ZDnet has a helpdesk for bugs in software (specifically Windows and Office)... It lists there are like 100 well-known bugs in WinNT alone. I think some of the bugs listed are unbelievable. For a office 98 (on the mac) it says "Wby does my hard drive run constantly as I type." Microsoft reports this will happen if you use True Type fonts that CAME WITH office 98." S: It seems surprising that the media are providing support in this way, given that Microsoft is widely reported as providing corporate support, unlike Linux which is claimed to rely on third-party support. update! In response to recent criticism by slashdot.org, Microsoft will anounce tomorrow free fast 24x7 support for resellers, a practice common in the Linux world. This is only ever so slightly tongue-in-cheek. It looks like they're feeling the heat. Note the support is still only given by "senior support engineers" and not the people who wrote the code.

In potentially one of the first direct attacks against the GPL, we received the following from yacc: "I've been wondering about INITIO 9400UW support for Linux (the idea is cool: 4 UW-SCSI channels on one PCI card). In looking around, I've found the following page, which offers a 2.1.x Linux kernel zImage binary only, requiring you to sign an NDA. Just wondering how this can happen, and whom to contact to correct this problem? (I'mean, a binary only module would be not much better, but at least legal, right?)" S: Unlike a company using GPL'd software internally this qualifies as external distribution and must be accompanied by source as I understand it. It is unclear whether they intend to release source code eventually or not given that they have released source code previously, but never-the-less they would still be in violation of the GPL. update: Just got an e-mail from a Paul Peissner at Initio who will force the issue to be addressed at a meeting at 1:30pm Pacific Standard Time. Update 2: Apparently this was a mistake due to a new web admin. Source should be posted shortly, and Initio would like you to report any bugs/patches to them as they are the maintainers. Business as usual. Final Update Source code is now posted. Thanks, Initio.

A while back there was an uproar when Dave Perry of Shiny Entertainment said he would patent the 3D algorithms used in his next game Messiah. Well... he got slammed by the games developing community, as he richly deserved. Many pointed out that not only were his techniques NOT novel, but also that his claim that "Patents are more universal and more respected than copyrights" did not hold water: only a few countries (US and Japan) allow the patenting of software. Anyway, the July print edition of Game Developer's Mag now reports that Dave Perry will not after all seek a patent.

On a more open note, 47-tek is releasing the source code to its game TEAM 47 GoMAN. It is Direct-X based, but there are tons of people trying to port Direct-X, for instance to GGI. From the features description, there seems to be some nifty code in there.

Per Lewau writes "I was scanning the GGI mailing list archives and appearently SciTech have decided to Free their graphics library MGL. Appearantly this lib was used to create Hexen II and WinQuake." Here is the post, here is the license, and don't forget to visit The MGL Site.

Redherring has a historical review of the markets that Microsoft tried to dominate... but failed. Unix and NetWare outnumber NT by 5 to 1, MSN did not become the dominant Internet provider, WebTV is somewhat of a flop, and Java (the other technology NASA actually uses) is extremely popular despite Microsoft's disparaging it. Meanwhile Alex St John proud architect of that abomination DirectX, explains why Microsoft chose it as API... The article in the print edition says: "What happens when folks such as ATI and nVidia, who never manage to make a working driver -- even when they are supplied with a DDK, sample code, and a testing procedure -- take it upon themselves to make up their own OGL drivers? BOOM! That's what." Perhaps it's got something to do with the whole driver running at ring 0? I wonder why those clever folks at the GGI-project decided to put the least stuff possible in kernel mode... perhaps security? Ah, yes, I forgot, they did not implement the first idea that gelled in their station-wagons. The article goes on to state that "The day Microsoft encounters a major competitor whose strength arises from the community of game developers is the day game developers will get Microsoft's full, undivided attention.": divide and conquer!

Finally, when confronted with a new threat, the best strategy to keep your sheep^H^H^H^H^Hcustomers is to frighten them. Somehow I'm not losing any sleep about security issues that others might add...

Quake 2 is vulnerable to an attack posted on rootshell.com Zoid says in his .plan:

rootshell.com has posted an exploit for Quake2 under Linux. This exploit was talked about on the buqtraq list a day or two ago.

I've known about this one due to loading shared libs. I forgot to specifically mention in the readme that Quake2 should not be setuid. If you want to use the ref_soft and ref_gl renderers, you should run Quake2 as root. Don't make the binary setuid. You can only run both those renderers at the console only, so being root isn't that much of an issue. The X11 render doesn't need any root permissions (if /dev/dsp is writable by others for sound).

The dedicated server mode (+set dedicated 1) doesn't need to be root either.

I will look at solutions to this problem in the next release. Problems such as root requirements for games has been sort of a sore spot in Linux for a number of years now. This is one of the goals that GGI is targetting to fix. I plan on supporting a ref_ggi in the near future.

Sengan Baring-Gould sent me This Article talking about Netscape, it's roll now that it's code is free, and the business behind it. Not much for the code jockey nerd, but very interesting from the perspective of the web in general.