LWN - Letters to the editor

Sections: Main page Security Kernel Distributions Development Commerce Linux in the news Announcements Letters All in one big page See also: last week's Letters page.	Letters to the editor Letters to the editor should be sent to letters@lwn.net. Preference will be given to letters which are short, to the point, and well written. If you want your email address "anti-spammed" in some way please be sure to let us know. We do not have a policy against anonymous letters, but we will be reluctant to include them.	May 23, 2002

From: "Roberson, Kyle" <Kyle_Roberson@duratekinc.com> To: "'letters@lwn.net'" <letters@lwn.net> Subject: too far Date: Tue, 21 May 2002 08:52:05 -0700 I think RMS has gone too far in his grasping for "mind share". How can I tell he's gone too far? I'm writing a letter to you, that's how I know. Normal, everyday grasping for credit wouldn't get me to write a public letter. In my view, FSF lost the marketing battle a long time ago by their own deeds. FSF had framed the issue: early announcements about what the system would be like, how the kernel would work, etc, etc, etc. I was there; I was waiting for it. The early Linux kernel specifically disavowed being that "pre-marketed" thing- GNU or the HURD. The "GNU System" was still late and being pre-marketed in techno-mystic terms when the first Linux "brands" came out. RedHat Linux could have chosen any brand name. Maybe because FSF had their brand clearly marked out and they didn't want to associate with it for whatever reason, RedHat didn't take the GNU/Linux "brand". FSF and GNU actually distanced itself from Linux initially. Now it wants to be named first in the same slashed word. I use RedHat Linux and Mandrake Linux. Those are brand names just like any other brand name. I don't have the right to rename them and neither does RMS. I know from experience in another industry, once a brand name is lost it is hard to make up. Big brewers and contract brewers stole "Microbrew" and there is no way to get it back. GNU and HURD have a long way to make up in the kernel "market". GNU has Debian GNU/Linux for a brand name. That's progress for RMS and FSF; stubborn bullying is working. Linux is successful and makes a good brand. GNU doesn't have a record of success in kernels and does not yet make a good brand for entire systems. I don't think there is a problem with other markets like compilers (although they may be getting a boost there from you know who) and cloned utilities. It is certainly an excellent brand for EMACS on the console. Distributors like RedHat do the work of putting together the system, Group Torvalds does the work of putting together the kernel, and FSF/GNU/RMS wants credit for that work. Just what RMS so easily accuses others of doing: getting "free" credit. Not that irony is an effective weapon against ideologues, especially one with a new "omagio" biography. Recent RMS missives seem to be getting a bit more harsh, more desperate, more grasping. Now he's calling into question whether the Linux kernel is GPL and whether Torvalds is a menace because he likes fine tools for himself and those who feed him bits. RMS is famous for being stubborn and I have personally never read anything whereby he admits significant fault. Considering FSF and GNU management failures in the past, this is a danger sign. Heed it. Kyle Trademarks are owned by whoever owns them... not me.

From: Jonathan Walther <krooger@debian.org> To: rms@gnu.org, letters@lwn.net Subject: it's not GNU/Linux; it's GNU Date: Tue, 21 May 2002 13:38:57 -0700 Dear Mr. Stallman: After reading your recent interview where you discussed SIGLINUX and BitKeeper it hit me: Linux is just the kernel. Hurd is also a kernel. Yet we never really talk about GNU/Hurd. We just talk about GNU. GNU is the free system built by the FSF, regardless of the kernel. I think a lot of peoples bitching and confusion would go away if it was a simple battle between GNU and Linux. Nobody felt right saying Gnu/Linux. I know I don't. The name of the kernel has nothing to do with the name of the system. To take an analogous case, one might as well say FreeBSD/Mach to refer to Apples new proprietary version of Unix. But do they? No, they call it Mac OS/X. It boils down to this: the kernel is NOT the system. Why don't we include Xfree86 in the name of our Free systems? It is as important a component as the kernel. I can see it now, GNU/Linux/Xfree86. A rather unwieldy mouthful. And maybe if somone adds nmap to their GNU system, they would name it GNU/Linux/Xfree86/nmap. And so on for each important component. Down that path lies madness. It took me 7 years of using "Linux" to get what you were saying, but I am satisfied I understand the issue properly now. I will refer to all my GNU systems as GNU, without any reference to "Linux" unless I am specifically talking about the kernel written by Linus Torvalds. It would be nice to see Redhat change it's name from "Redhat Linux" to "Redhat GNU, now with the new, improved Linux kernel!" Cheers! Jonathan -- Geek House Productions, Ltd. Providing Unix & Internet Contracting and Consulting, QA Testing, Technical Documentation, Systems Design & Implementation, General Programming, E-commerce, Web & Mail Services since 1998 Phone: 604-435-1205 Email: djw@reactor-core.org Webpage: http://reactor-core.org Address: 2459 E 41st Ave, Vancouver, BC V5R2W2

From: Richard Stallman <rms@gnu.org> To: krooger@debian.org Subject: Re: it's not GNU/Linux; it's GNU Date: Wed, 22 May 2002 16:29:24 -0600 (MDT) Cc: letters@lwn.net Yet we never really talk about GNU/Hurd. We just talk about GNU. Actually we do sometimes say "GNU/Hurd", when we want to emphasize that it is GNU, using the Hurd as kernel. I think a lot of peoples bitching and confusion would go away if it was a simple battle between GNU and Linux. It isn't a battle, just a competition. It isn't between GNU and Linux. Linux is used with the GNU system, and so is the Hurd. It is not wrong to shorten the name GNU/Linux to GNU. The system is basically GNU. There are three reasons why I say "GNU/Linux": * To distinguish it from GNU properly speaking, which uses the Hurd. * To give Linus a share of the credit. It would be ungentlemanly to ask people to stop giving him credit. * To help people connect it with what they have heard about "Linux" (which is about this system).

From: Leon Brooks <leon@cyberknights.com.au> To: rms@gnu.org Subject: GNU/Linux Date: Wed, 22 May 2002 10:17:17 +0800 Cc: letters@lwn.net I'm CC'ing this to LWN because I don't get stuff broadcast around the net simply for being Leon Brooks. (-: I'm sure LWN readers would be delighted to see your response there as well. > The use of Bitkeeper for the Linux sources has a grave effect on the free > software community, because anyone who wants to closely track patches to > Linux can only do it by installing that non-free program. > One solution is to set up another repository for the Linux sources, using > CVS or another free version control system, and arranging to load new > versions into it automatically. This could use Bitkeeper to access the > latest revisions, then install the new revisions into CVS. A better solution would be the traditional approach: to reverse-engineer BitKeeper, at least as far as being able to extract enough information to CVS it. There would be additional justice in there from your POV because the proprietariness of BitKeeper would have been diluted by its contact with Linux, not the other way around. > if they'd like someone from the GNU Project to give a speech for them, they > ought to treat the GNU Project right, and call the system "GNU/Linux". The > system is a variant of GNU, and the GNU Project is its principal developer, You could make that case for an entire distribution, but I don't believe that it holds for the Linux kernel itself. I would cheerfully accept GNU+Linux (even though this now sounds like the Wars of the Roses) since it connotes one of either Linux-platformed GNU tools or a GNU-enhanced Linux kernel - but GNU/Linux, minor though the difference is, implies that GNU wrote or owns the Linux kernel, which is not true. To take the argument further, consider the example of Mandrake Linux 8.2, which includes 3151 packages. Mandrake is pretty good about licences; they've even got to tossing out the old Netscape this time around for not being even BSD. The vast majority of these three thousand packages have nothing to do with the GNU organisation, per se. If you consider a package tally improper representation, count the binaries or the sources. I'm sure KDE alone outweighs the GNU utilties (it certainly does in file count or sheer bytes). Reductio ad absurdum says that it should be GNU/BSD/TrollTech/MySQL/etc/Linux which aside from being unreasonable would be bad because it also highlights non-Free and less Free licences. Another set of edge cases are Debian's FreeBSD and Hurd `ports'. Debian uses `GNU/Linux' throughout, but would you expect them to use `Debian GNU/FreeBSD' if they bundle some GNU utilities with their FreeBSD `port'? On the other edge, `Debian GNU/Hurd' is pretty much optimal for their Hurd-based `port'. I think a better line of argument would be based around the extensive use of GPLed software in such distributions. GNU/FSF (is a two-headed Gnu logo in order?) _did_ make and publicise the GPL, at first single-handedly. Meanwhile, your arguments about choosing on usability vs choosing on principle are much better put IMESHO, but there are still cases (fewer each month, hurrah!) where closed software is the only realistically useable candidate. Even though an Open or even Free implementation exists, it is sometimes completely unreasonable to justify their choice in mission-critical positions. As above, my choice is usually the fastest effective solution, reverse engineering to make a Free product. Do it fast first so that you have an answer, any answer (Linux) and then do it properly later (Hurd) so that the final result will be best. In analogy: put your finger in the dike now, replace it with concrete later. Always there must be a dependable plan for and action towards the concrete replacement. Microsoft, for example, would have you put your finger in the dike, only to feel it being grasped and firmly held by a licencing agreement while the dike erodes and you eventually drown. The only-on-technical-merit argument would often stick there. The only-on-principle argument would come back with the perfect concrete replacement, only to find the land flooded and the populace drowned or departed. For the record, my own systems and as many others as I can make such decisions for run 100% Open Source and the vast majority of that 100% is also Free as in GPL or close derivative. Perhaps I'm being too pragmatic in my reasoning. What do you think? -- CyberKnights Modern tools, traditional dedication. +61-409-655-359 http://www.cyberknights.com.au/ linux.conf.au 2003 The Australian Linux Technical Conference http://conf.linux.org.au/ 22-25 January 2003 in Perth, Western Australia

From: Leandro Guimarães Faria Corsetti Dutra <leandrod@mac.com> To: info@digitalconsumer.org, LWN Editor <letters@lwn.net> Subject: Open Systems and the Consumer Technology Bill of Rights Date: Thu, 16 May 2002 18:29:31 +0200 I would advance that the Consumer Technology Bill of Rights should include: 7. Any digital product that encodes users' content should be based on open, fully documented standards. This right guarantees that the user can always get access to his data even if the vendor of the system which encoded it flounders or looses interest in maintaining the product. Just imagine the disaster which would ensue if Microsoft floundered or lost interest in MS Office, given how much data is encoded in its proprietary, hard-to-reverse-engineer formats. -- _ / \ Leandro Guimarães Faria Corsetti Dutra +41 (21) 216 15 93 \ / http://homepage.mac.com./leandrod/ fax +41 (21) 216 19 04 X http://tutoriald.sf.net./ Orange Communications CH / \ ASCII Ribbon Campaign against HTML email +41 (21) 216 15 93

From: "Hethcoat-III, Charles L" <Charles.L.Hethcoat-III@boeing.com> To: "'letters@lwn.net'" <letters@lwn.net> Subject: A brief comment on Judge Whyte's statement Date: Thu, 16 May 2002 12:02:50 -0500 "Meanwhile, back in the real world, here is a release from the EFF on the latest ruling in the Elcomsoft case. Judge Whyte has refused all of the defense's motions for dismissal. The DMCA, he says, is entirely clear: it means to ban all "circumvention devices" regardless of their legal uses.[...]" Johnny Holmes, the famous former Harris County (Texas) District Attorney, once made this statement concerning a controversy about whether some local law was right or wrong, and should or should not be enforced: "The best way to repeal a bad law is to enforce it vigorously."

From: Bill Bogstad <bogstad@pobox.com> To: letters@lwn.net Subject: Banning magic markers the next step? Date: Thu, 16 May 2002 14:25:48 -0400 LWN suggests that we should now expect the recording industry to bring marker manufacturers to court under the DMCA because some of the CD copy protection technologies can be circumvented using a magic marker. Obviously, that isn't going to happen. There is nothing that the recording industry could possibly gain by doing this. If they lose, then at least some aspects of the DMCA would have been struck down. If they win, then the pressure on Congress to do something about this ludicrous law would be too strong to stop. Either way they lose. That's why WE will have to be the ones to bring such a suit. Burn some CDs with the same (or a similar) copy protection scheme and then sue the marker manufacturers ourselves. Whether we win or lose in court doesn't matter. Either way we win to at least some extent. Unlike the situation with ludicrous patents, we don't even have to nominally violate the law to get standing to bring a suit. The only downside that I can see is that we'ld have to drag the marker industry into a situation that really isn't of their making. Unfortunately, I see no way around this. Bill Bogstad bogstad@pobox.com

From: dave mallery <dmallery@cia-g.com> To: letters@lwn.net Subject: outlaws all Date: Thu, 16 May 2002 15:37:33 -0600 hi guys not only must we outlaw felt pens, but felt, alcohol and and ink since they can be combined by outlaws into subversive weapons of mass circumvention. i believe we have achieved a classic "reductio ad absurdum." we should all lobby for immediate and strict enforcement. when pens are outlawed, only outlaws will have pens! -- Dave Mallery, K5EN (r/h 7.2 krud; debian testing) PO Box 520 .~. Ramah, NM 87321 /V\ no gates... /( )\ running GNU/Linux no windows! ^^ ^^ free at last!

From: Dylan Griffiths <Dylan_G@bigfoot.com> To: letters@lwn.net Subject: Copy-protection story. Date: Thu, 16 May 2002 18:06:26 -0600 "Essentially, all you have to do is cover that track. This can be done with a Post-It note, a piece of electrical tape, or a carefully-drawn line with a heavy marker. All it takes is a few seconds of effort, and the "rip protection" is no more. " While true, one should never place tape or a post-it (tm) note on a disc which is to be inserted into a player of any kind. Such objects affect spin of a disc, especially at high speed, and may cause damage or destruction of both the disc and the player. This is why rental places generally write in marker on DVDs they rent (or put a small ring sticker around the central hole, where it is least likely to cause problems). Please don't reccomend the tape approach as it is likely to lead to lasting damage. -- www.kuro5hin.org -- technology and culture, from the trenches. -=-=-=-=-=- "This chart is a visual representation of amici's understanding of the decline of the growth of public domain as a result of repeated copyright term extensions." http://eon.law.harvard.edu/openlaw/eldredvashcroft/pubdomain.html -=-=-=-=-=-

From: Paul Komarek <komarek@andrew.cmu.edu> To: lwn@lwn.net Subject: Per-driver filesystems -- easy for sysadmins Date: Thu, 16 May 2002 16:57:43 -0400 (EDT) As a grad student/sysadmin/chief-bottle-washer, I have no qualms about learning to mount mini filesystems to allow user-space communication. In your coverage, you ask "How are VIA motherboard users to know that they can mount a devvia filesystem somewhere to read their temperature sensors?". This is a legitimate question, but not a show-stopper. Having fought with lm_sensors a number of times, I wish I could simply execute "mount -t devvia none /floppy" to get sensor info instead of fiddling with dozens of modules and discovering where in /proc the data came out. The same goes for getting scsi info, etc. I don't think that learning how to mount the devscsi filesystem will be any harder than discovering that /proc/scsi/scsi has a wealth of useful info. I've been using GNU/Linux systems quite a while now, and the information-rich, human-readable /proc is among my favorite features of linux. That said, I'm familiar with less than half of the stuff in /proc. Your question about devvia is very similar asking how one discovers that vm tweaking is possible in /proc/sys/vm. Configuring user-space communication in user-space seems downright natural. One has to learn about files and directories at some point (though Microsoft disagrees), and eventually one discovers /etc/fstab. Having to add the "right" line to /etc/fstab is no worse than having to add the right line to /etc/modules.conf (of course, we can't get rid of modules.conf because of fancy tricks like conditionals and module stack ordering). Union mounts (if supported) could allow recreation /proc. Other mount options could allow leverage of existing fs knowledge to achieve just about any configuration one desired. I don't think there is a system management problem with Al Viro's mini filesystems for user-space communication. It's different than the status quo, but certainly not harder. -Paul Komarek

From: mikeraz@patch.com To: letters@lwn.net Subject: Banks and Browsers Date: Fri, 17 May 2002 13:30:21 -0700 You folks said: > If AOL really does deploy Mozilla-based browsers to its customers, > one can expect things to improve quite a bit more. That may not be the case. I use Mozilla on Linux and Win platforms. While I cannot access Fleet from Linux with Mozilla, it works flawlessly under Windows. Recently release 0.9.9 of Mozilla receives an error accessing ATT Universal Card site, yet it still works when access through Mozilla on a Windows machine. AOL/Mozilla will yeild gains by encouraging banks to support the browser. The combination may have no effect on improved Linux support. -- Michael Rasmussen aka mikeraz Be appropriate && Follow your curiosity "They that give up essential liberty to obtain temporary safety, deserve neither liberty nor safety." -- Benjamin Franklin and the fortune cookie says: Anger is momentary madness. -- Horace

From: spamtroll@primary.seanmcpherson.com To: letters@lwn.net Subject: Re: Subsidizing the development of non-free software Date: Fri, 17 May 2002 09:47:09 -0400 (EDT) >Ronald Cole <ronald@forte-intl.com> writes: >dm@chrononaut.org (David Moles) writes: >> Let me put the question another way: Is it acceptable for private >> interests to take free software developed with the public's money >> and make it into software that is not available to the public? > >If it's licensed under the GPL, then the answer is yes! I am free >to take GPL's software, make enhancements and sell it. Can you get a >copy if you want one? Only if you pay my exorbitant fee (the GPL >doesn't require me to distribute to just anyone who asks). Bonus for >me if I sell binary-only with the written offer for sources and the >three years (the minimum) that the GPL requires me to make them >available passes without anyone taking me up on that offer! >Essentially, I will have taken GPL'd code and made proprietary >enhancements for which I won't have had to distribute the source (and >it's not entirely clear to me whether the GPL forbids the binary-only >recipients from further redistribution if they don't have the source, >but I would think so). The GPL requires that you provide "equivalent access" to the source; this means that any fee for access to the source can't be greater than the fee to download or purchase the binary. So, if your fee is so exorbitant that no one thinks the code is worth downloading, the odds of there being many purchases of the binary-only product get a lot lower. And it only takes one person/company to purchase the product to then request (and pay the fee for) the code and release it. Is there a possibility for abuse, as Mr. Cole so 'helpfully' points out? Yes, but the GPL makes the cost for such an abuse pretty high, as well! Sean McPherson