OpenBSD's De Raadt sings Stallman blues

By   |  May 1, 2008

The OpenBSD team traditionally makes available a new song with each new release of its software. The songs, though often of dubious lyrical quality, typically embody the current thinking of the development team, in particular that of team leader Theo de Raadt. In the past the songs have railed against proprietary software developers and corporates and the challenges they throw in the way of free software. This time, however, the target is the Free Software Foundation’s Richard Stallman.

Late last year a very public fight broke out between Stallman, a man known for his unbending opinions and Theo de Raadt, and equally opinionated free software advocate. The spat follows an email by Stallman to an OpenBSD mailing list in which he accuses OpenBSD of including non-free software in its distribution: “The fact that OpenBSD is not a variant of GNU is not ethically important. If OpenBSD did not suggest non-free programs, I would recommend it along with the free GNU/Linux distros.”

De Raadt, along with many other OpenBSD supporters responded with equal vitriol. De Raadt wrote:

Richard, you are wrong. You said very clearly in your interview that
the ports tree contains non-free software. It does not. It is just a
scaffold of Makefiles containing URLs, and an occasional patch here or
there.

You are just plain wrong. And you are not enough of a man to admit
that you are wrong.

I may be unfriendly at times, but you are a power-misusing
hypocritical liar who attacks projects that try harder than any
others to only make free software available.

Shame on you.

And today, as OpenBSD releases 4.3 of its distribution, De Raadt has also released a new theme tune titled Home of Hypocrisy in which the OpenBSD team characterises Stallman as trying to control the flow of free software. In the song, the Stallmanesque character says there are two rules: Rule one is that no-one can give away their software. Rule two is that “you must give it to me, so I can give it away properly for free”.

More telling is the description of the song in which De Raadt calls Stallman a “false leader”. He goes on to say: “[Stallman] prefers actions which he thinks are best for him — and him alone — and then lies to the public.”

Ironically De Raadt and Stallman are two of the primary leaders in the free software movement and have both dedicated themselves to advancing the values of free software. Indeed, De Raadt was the recipient of the FSF’s 2005 award for his contribution to free software.

Neither De Raadt nor Stallman is likely to back down quietly in this battle so look out for future missives on the topic.

Tags: ,

Comments

23 Responses to “OpenBSD's De Raadt sings Stallman blues”

  1. Flugtag
    May 1st, 2008 @ 3:24 pm

    Time for a fork!

  2. 0x000
    May 1st, 2008 @ 5:34 pm

    ^lol :)

  3. Kostelanetz
    May 1st, 2008 @ 6:15 pm

    > Time for a fork!

    I presume, then, you would be willing to independently maintain a fork of OpenSSH — an adoptive element of *every* free OS — under the GPL, with all of the highly advanced security expertise that would involve.

    Be our guest, Flugtag!

  4. Craig
    May 1st, 2008 @ 6:45 pm

    “You said very clearly in your interview that the ports tree contains non-free software. It does not. It is just a scaffold of Makefiles containing URLs, and an occasional patch here or there.”

    I don’t use OpenBSD (or any ports-based system) so I may be misunderstanding this, but it sounds like de Raadt is saying that there is no software as such in the portage tree, only URLs to projects hosted elsewhere, plus the occasional patch. So if what RMS really means is that some of the projects that OpenBSD links to through its ports tree are non-free software, then if that’s true, I’d say de Raadt is quibbling. It’s a distinction without a meaningful difference.

  5. Shagbag
    May 1st, 2008 @ 6:59 pm

    Never mind the fork. Put the snags on the barbie, break open a few tinnies, sit back and enjoy the show!

  6. Jaro
    May 1st, 2008 @ 7:07 pm

    What interview is De Raadt referring to? I don’t’ see the linked email containing anything that would suggest that ports “contains” nonfree software. All I see Stallman say is he “is unsure” if it contains any “non-free firmware blobs”, and “was told” ports “suggests” installation of non-free software. That’s quite contrary to what De Raadt claims Stallman to have said. So could someone point me to the interview De Raadt is talking about and/or explain to me why it hasn’t been linked to or otherwise mentioned in this whole article?

    I can see there’s a lot of hot air between these guys, but if there are any claims made about the other persons arguments the least one could do is to provide the proof to his claims. Now I’m inclined to think De Raadt is just twisting words and trying to see a problem where none exist.

  7. Somebody
    May 1st, 2008 @ 7:23 pm

    Recently RS wrote in his blog why he changend from his Thinkpad T23 to the “functionally equivalent” OLPC. He reasoned about it because of its FreeBios.
    I mean, that’s ok, but why didn’t he instead try to gain more support for http://coreboot.org ? The OLPCs Bios runs on exactly one system, which is the OLPC. Should we all twiddle along on OLPCs now, or what? I mean, he could have explicitely said that it is indeed coreboot what is running there. Which for now runs only on a limited amount of other systems. If he chooses his words like that, i think it will stay this way for lack of publicity, mindshare, interest, whatever.

  8. Jaro
    May 1st, 2008 @ 7:46 pm

    …I’d like to add… I also find it weird that Stallman emphasizes the fact that his decision is based on hearsay. ;)

  9. dimitri
    May 2nd, 2008 @ 6:39 am

    Stallman is a stupid!!!.

    fuck.

  10. jax
    May 3rd, 2008 @ 12:57 am

    This seems to me as if de raat is upset that stallman wont plug openBSD to the public because openBSD “suggests” non-free software, and De Raadt has not commented on the nature of that software. He has only stated that openBSD does not directly contain non-free software, which is not the issue at hand.

    So De Raadt is angry just because RMS wont tell people to use openBSD even though he stated quite clearly “If OpenBSD did not suggest non-free programs, I would
    recommend it along with the free GNU/Linux distros.” Seems like De Raadt is being a jerk and a half.

  11. kmashraf
    May 3rd, 2008 @ 4:59 am

    I don’t think we would have come this far without the ‘stupidity’ of Richard M Stallman!
    Please! Nothing in this world is perfect, least of all being a person who is liked by one and all without any reservations whatsoever. It is what you have done that counts for something. Period.

  12. Sysadm1n
    May 4th, 2008 @ 12:11 am

    Theo de Raadt, now a master at nailing BSD’s coffin shut from the inside.

  13. Tectonic » New releases: OpenSUSE, Mandriva Xfce, Slackware
    May 4th, 2008 @ 8:07 am

    [...] If you’re new here you may want to subscribe to our RSS feed or our weekly email newsletter. Thanks for visiting!As many of us enjoy the last hours of an extended long weekend the May Day holidays have been a busy period for free software developers. Among the releases of the past week are a new beta for OpenSUSE 11.0, Slackware 12.1 makes a showing, Mandriva adds an Xfce version, gNewSense 2.0 peeks out and, of course, OpenBSD 4.3 is released along with a new song that is causing much debate. [...]

  14. Kostelanetz
    May 4th, 2008 @ 1:04 pm

    > What interview is De Raadt referring to?

    The interview he gave for bsdtalk:
    http://bsdtalk.blogspot.com/2007/10/bsdtalk132-richard-stallman.html

    > I can see there’s a lot of hot air between these guys,
    > but if there are any claims made about the other persons
    > arguments the least one could do is to provide the proof
    > to his claims. Now I’m inclined to think De Raadt is just
    > twisting words and trying to see a problem where none exist.

    The nub of the issue is Stallman’s hypocrisy. He has a tendency in
    interviews to erroneously approve certain OSs as “free” on the basis of
    scant, second-hand research, but disapproves of OpenBSD, which
    systematically purged itself of non-free components seven years ago and
    only provides installation scripts for proprietary apps within a
    strictly unsupported Ports Tree.

    Stallman maintains that the existence of non-free applications in a
    repository/Ports system’s install scripts is evil, because it implicitly
    endorses the use of non-free software, but the GNU project *explicitly*
    advocates GPL-licensed software for use in conjunction with non-free OSs
    and extends *official* support to the Windows port of Emacs and gcc.

    Their excuse is that by creating Win32 builds, it will persuade Windows
    users — initially attracted by gcc and Emacs’s technical strengths — to
    move over wholesale to a free OS platform on the strength of FLOSS’s
    moral and political virtues. Stallman himself admits that the case for
    this is at best equivocal: “The practical effects are mixed. Making free
    apps run on non-free systems paves the way for some users to migrate to
    free systems, and for some users eliminates a motivation to migrate…
    Most users are reluctant to change operating systems at all.”

  15. jax
    May 4th, 2008 @ 5:36 pm

    Kostelanetz,

    I think either you or I have missed part of this equation. As I see this whole debacle it is over whether or not Stallman endorses OpenBSD as a free OS. If that is the matter at hand then there is no hypocrisy in any sense by him stating that he would not endorse an OS that suggests non-free software. It would be hypocrisy if Stallman was endorsing Ubuntu(or similar non-free software promoting OSes) and not OpenBSD but he hasn’t. The only systems that GNU endorse are those that are entirely free themselves and only promoted and endorse free software. A list which is very, very short actually (if I recall only about five rather obscure distributions).

    As to the endorsement of free software to a non-free software user (i.e. emacs on windows) vs endorsing non-free software to a free software user (i.e. non-free ports to OpenBSD) I think that the former is making a step forward in the promotion of free software and the latter a step backwards. Some things are relative and in a relative sense promoting free software to a windows user is good, while promoting proprietary software to an OpenBSD user is not.

    So as I said as I read it, this is not a matter of hypocrisy at all, and it seems to me that de Raadt is being the jerk here by dealing with this through insults rather than constructive dialogue. Can you shed a different light on this?

  16. Kostelanetz
    May 4th, 2008 @ 8:56 pm

    > I think either you or I have missed part of this equation.
    > As I see this whole debacle it is over whether or not Stallman endorses
    > OpenBSD as a free OS.

    You’re sorely mistaken. OpenBSD draws upon a quite different philosophy of software freedom (i.e. freedom as a minimum of legal/contractual encumbrances) that’s ultimately antithetical to Stallman’s, and has *never* looked for Stallman’s endorsement.

    > As to the endorsement of free software to a non-free software user (i.e.
    > emacs on windows) vs endorsing non-free software to a free software user
    > (i.e. non-free ports to OpenBSD) I think that the former is making a
    > step forward in the promotion of free software and the latter a step
    > backwards. Some things are relative and in a relative sense promoting
    > free software to a windows user is good, while promoting proprietary
    > software to an OpenBSD user is not.

    Well unfortunately, by RMS’s own reasoning, the GNU Project’s Win32 builds of gcc and Emacs can’t be excused/rationalized in that way. Why? Because Stallman’s on record as saying that the group least liable to be persuaded into adopting an unequivocally free software platform by Windows ports of GPL software are the highly tech-savvy “power users”. The *very* group that would use gcc….

    > The only systems that GNU endorse are those that are entirely free
    > themselves and only promoted and endorse free software. A list which is
    > very, very short actually (if I recall only about five rather obscure
    > distributions).

    Distros whose kernels are by no means free of binary-only firmware images. You can tell because, as Alan Cox is apt to point out, if you were to remove them entirely, Linux wouldn’t be able to run on modern hardware.

    Incidentally, I see RMS’s spat with Theo has had an effect on the GNU Project’s list of approved operating systems. This used to include ReactOS, an OS specifically designed to run Win32 binaries. Given that the vast majority of Windows applications are closed-source, how would that not — just as WINE does — also “implicitly suggest” the use of non-free software?

    Besides, in interviews he’s lent his endorsement to OpenSolaris, an OS actually very far from being “free”.

    > Can you shed a different light on this?

    Well the first thing to point out is that this isn’t a fight the developers had in any way sought out. OpenBSD developers normally couldn’t give a damn whether or not Stallman, the FSF or GNU Project endorse them, and weren’t overly interested in taking Stallman up on comments regarding a service users are strongly *dissuaded* from using and which does *not* enjoy official support.

    It was only when Stallman gatecrashed a mailing list strictly for technical discussions, claiming that he’d been misrepresented (in respect of which he wasn’t specific), that he started to raise the developers’ ire.

  17. Jax
    May 5th, 2008 @ 2:18 pm

    >It was only when Stallman gatecrashed a mailing list strictly for technical discussions, >claiming that he’d been misrepresented (in respect of which he wasn’t specific), that he >started to raise the developers’ ire.

    Well if that’s the case then it seems unfortunate, but it was in the light of a misunderstanding by Stallman thinking that people were misstating him, and then attacking him over statements that he didn’t make. It seems reasonable on Stallmans behalf to attempt to clarify his position to people that are supposedly allies. From what I’ve read Stallman has taken a lot of punishment on that list, (and from OpenBSD development by releasing this song), yet has not stooped to do the same. So whether or not it began with Stallman being mislead, it is the OpenBSD community that comes out unable to refute the disagreements that Stallman has with their OS, so they resort to vicious attacks.

    >Well unfortunately, by RMS’s own reasoning, the GNU Project’s Win32 builds of gcc and >Emacs can’t be excused/rationalized in that way. Why? Because Stallman’s on record as >saying that the group least liable to be persuaded into adopting an unequivocally free >software platform by Windows ports of GPL software are the highly tech-savvy “power users”. >The *very* group that would use gcc….

    So you’re saying that RMS is against people that refuse to use a completely free OS using GNU software? I don’t think that he ever said that, in fact GNU started out, and was fully supported, on Unix. So it seems like Windows would be the next likely progression in todays world. Even if he does believe the users unlikely to switch to totally free platforms, I doubt he’s ever advocated against pushing them toward it.

    >Distros whose kernels are by no means free of binary-only firmware images. You can tell >because, as Alan Cox is apt to point out, if you were to remove them entirely, Linux wouldn’t >be able to run on modern hardware.

    This is not true. Simply because Alan Cox may have said it, once upon a time, doesn’t make it true. I know that Blag and GNewSense use the same kernel that is deblobed, called linux libre, that the developers at Blag built and deserve a good deal of credit for. Ututo uses the same or similar kernel as well. It is even in discussion in the Fedora development whether to abandon the blobed kernel for linux-libre.

    >Incidentally, I see RMS’s spat with Theo has had an effect on the GNU Project’s list of >approved operating systems. This used to include ReactOS, an OS specifically designed to >run Win32 binaries. Given that the vast majority of Windows applications are closed-source, >how would that not — just as WINE does — also “implicitly suggest” the use of non-free >software?

    I don’t think that this is a very worthwhile assumption. I don’t know if GNU has ever endorsed ReactOS, so I can’t speak to that, however I do know that there was a time when only Blag and Ututo were endorsed, so it is doubtful that if ReactOS was endorsed that it was removed because of this spat. Perhaps it was removed because GNU realised that they are endorsing non-free software?

    >Besides, in interviews he’s lent his endorsement to OpenSolaris, an OS actually very far >from being “free”.

    I don’t know too much about OpenSolaris, or whether RMS has endorsed it, but I do know that GNU has endorsed the CDDL (Open Solaris’ license) as free software even though it is not compatible with the GPL. Perhaps that is what you are referring to?

    After reading a good deal of this debate it seems to me as if the OpenBSD community is trying to burn bridges that once seemed very strong. Now, rather than engage in discourse, or accept that GNU views their practices as unfree. They have chosen to keep their unfree ports and simply attack GNU with insults. The fact that de Raadt would stoop to something so childish as this is bizarre and I think may well ruin a potentially powerful alliance.

  18. Kostelanetz
    May 5th, 2008 @ 8:20 pm

    > I don’t know too much about OpenSolaris, or whether RMS has endorsed it,
    > but I do know that GNU has endorsed the CDDL (Open Solaris’ license) as
    > free software even though it is not compatible with the GPL. Perhaps
    > that is what you are referring to?

    No, Stallman was specifically endorsing OpenSolaris.

    > I don’t know if GNU has ever endorsed ReactOS, so I can’t speak to that,
    > however I do know that there was a time when only Blag and Ututo were
    > endorsed, so it is doubtful that if ReactOS was endorsed that it was
    > removed because of this spat.

    ReactOS was certainly there in January, together with GNU/Darwin. Strange that it should have disappeared from the list after the OpenBSD developers so forcefully pressed home the hypocrisy of its inclusion!

    > So you’re saying that RMS is against people that refuse to use
    > a completely free OS using GNU software?

    You don’t seem to have followed my argument here, so I shan’t pursue it further.

    > This is not true. Simply because Alan Cox may have said it,
    > once upon a time, doesn’t make it true.

    A senior kernel maintainer writing in March of this year, specifically in connection with the kernel libre project — yeah, what does *he* know? ;)

    > They have chosen to keep their unfree ports

    They’ve chosen to keep them, believing that software projects shouldn’t arrogate individuals’ moral choices, even where a choice in favor of installing proprietary code isn’t the one that they themselves would take (and as RMS concedes, OpenBSD ships with entirely free code). To call this an endorsement of proprietary code, as Stallman does, seem rather strange given that OpenBSD users are constantly and actively being dissuaded from installing the Ports Tree.

    As I’ve explained, BSD developers take it as axiomatic that licenses should be as permissive as possible, and as such Stallman and his nostrums hold absolutely no philosophical weight with them.

    > and simply attack GNU with insults

    Valid arguments *and* insults ;)

  19. Jax
    May 6th, 2008 @ 11:42 am

    >No, Stallman was specifically endorsing OpenSolaris.

    I don’t know anything about this, however I’m certain that if that is the case that there was context to the situation. Not knowing that context, and having none provided, there is little further that it can be discussed.

    >ReactOS was certainly there in January, together with GNU/Darwin. Strange that it should >have disappeared from the list after the OpenBSD developers so forcefully pressed home the >hypocrisy of its inclusion!

    I’m not certain about that, although if that’s so then they should not have been there, and if it was the OpenBSD community that alerted GNU to those OSes being non-free then good. Certainly GNU should have cleaned their own houses before commenting on others, however I find it neither strange nor hypocritical that they’d remove an OS that they were informed did not adhere to their requirements. Actually I find it quite vindicating of their position that they would lend credence to the worlds of a community that had been so virulent toward them.

    >You don’t seem to have followed my argument here, so I shan’t pursue it further.

    ok

    >A senior kernel maintainer writing in March of this year, specifically in connection with the >kernel libre project — yeah, what does *he* know? ;)

    Like I said, just because alan cox says it doesn’t make it true. We aren’t really discussing theory at this point, people are using these kernels as we speak, so alan cox’ belief, whether he still holds this position or not, is immaterial. Correct me if I am wrong but I believe that you’re referring to the red hat list discussion about the inclusion of the kernel into fedora, which I am quite familiar with and though alan cox may have implied that it was a potential disaster I don’t belief he ever disputes the fact that people are currently using the kernel and doing so on modern hardware. If you know of a different discussion please clue me in. Also if you know of more non-free blobs in linux-libre I’m sure that the devs at blag would be more than happy to remove them.

    >They’ve chosen to keep them, believing that software projects shouldn’t arrogate >individuals’ moral choices, even where a choice in favor of installing proprietary code isn’t >the one that they themselves would take (and as RMS concedes, OpenBSD ships with >entirely free code).

    Well I actually hold with the OpenBSD folks on this issue however that’s not the problem here as much as how the OpenBSD community responded in poor taste to someone that believed them friends that had been mistaking his position.

    >Valid arguments *and* insults ;)

    The latter is the problem and tends to make the former moot. When people act as the OpenBSD community did in this situation toward allies it just makes them look full of themselves and disenfranchises people that are of similar opinions.

  20. Kostelanetz
    May 7th, 2008 @ 1:19 am

    >> No, Stallman was specifically endorsing OpenSolaris.

    > I don’t know anything about this, however I’m certain that if that is
    > the case that there was context to the situation.

    The context was that Stallman had used his position as self-styled leader of the Free Software Movement to endorse an operating system cursory research would have revealed as unfree.

    >> ReactOS was certainly there in January, together with GNU/Darwin.

    > I’m not certain about that

    I looked at the relevant page in January, and ReactOS was certainly there.

    > however I find it neither strange nor hypocritical that they’d remove
    > an OS that they were informed did not adhere to their requirements.

    But in the eyes of the OpenBSD developers, indeed in the eyes of an impartial observer, it was surely hypocritical of RMS to endorse a project that by his own lights implicitly “suggests” the use of non-free software, and then proceed to criticize another project for doing so.

    > Certainly GNU should have cleaned their own houses before commenting
    > on others

    So it *was* hypocritical, then ;)

    >> A senior kernel maintainer writing in March of this year,
    >> specifically in connection with the kernel libre project — yeah, what
    >> does *he* know? ;)

    > Like I said, just because alan cox says it doesn’t make it true. We
    > aren’t really discussing theory at this point, people are using these
    > kernels as we speak, so alan cox’ belief, whether he still holds this
    > position or not, is immaterial.

    At best, what you’re talking about is anecdotal evidence drawn from a relatively limited number of hardware[/firmware] configurations — it would be a quite illegitimate step to then propose that this binary-only firmware image-free kernel is suitable for use in conjunction with the vast plethora of hardware in use today (and thankfully, kernel libre proponents don’t make that case).

    The irony, as Cox himself observes, is that the hardware configurations which have been found to work with this kernel — the very cause of the moral self-congratulation — do so partially by dint of unmodifiable firmware in the devices themselves!

    > Correct me if I am wrong but I believe that you’re referring to the
    > red hat list discussion about the inclusion of the kernel into fedora

    The very one — and quite fascinating reading it makes, too, from the point of view of the FSF’s political pretensions. If using a hardware device with unmodifiable firmware is in principle no less morally egregious than using a binary-only firmware image-encumbered kernel, then, logically speaking, the political stakes are far higher than would be the case if it were a simple matter of excluding this or that sequence of GPL-violating hex. Rather, a decisive proportion of modern hardware is itself “unfree” — the information that might allow one to write drivers, compose firmware or otherwise explore and extend a device’s functionality remains fatally subject to proprietary restrictions, upheld by the laws regulating property relations within a capitalist society. So software freedom presupposes hardware freedom, and hardware freedom presupposes/awaits a revolutionary transformation in society’s power relations. But the free software movement as embodied by Stallman — unless it was intended all along as a kind of parodic reductio ad absurdum of bourgeois property relations — seeks nothing of the kind, since that’s precisely what its licenses logically and materially presuppose.

    > The latter is the problem and tends to make the former moot. When
    > people act as the OpenBSD community did in this situation toward allies
    > it just makes them look full of themselves

    Well if it’s a question of credibility, RMS isn’t exactly blessed with a surfeit of the stuff amongst *BSD devs. A *BSD developer might place equally stringent demands upon his or her code, and pursue a concept and collective practice of software freedom older than copyleft, only to be told by Stallman — as OpenBSD’s Marco Peereboom was quite recently — that the movement for free software began with his efforts in 1983!

  21. Jax
    May 7th, 2008 @ 9:17 am

    >The context was that Stallman had used his position as self-styled leader of the Free >Software Movement to endorse an operating system cursory research would have revealed >as unfree.

    That’s not context, that’s the way you’ve chosen to frame it. Since I’ve never read anywhere that stallman did endorse OpenSolaris I can only assume that, if ever he did, the context would be in line with GNU philosophy.

    >But in the eyes of the OpenBSD developers, indeed in the eyes of an impartial observer, it >was surely hypocritical of RMS to endorse a project that by his own lights implicitly >suggests” the use of non-free software, and then proceed to criticize another project for >doing so.

    Well no observer is impartial. I think that it is important to point out however that GNU is larger than stallman, and while his endorsements go out to OSes that promise a certain standard that by no means implies that he or GNU personally inspect each aspect of the system before endorsing it. It is going to be largely on the word of the OS developers that the OS will adhere to GNU’s philosophy. An oversight would not be hypocritical. It would have been hypocritical if after condemning OpenBSD and once informed of the issue, they chose to continue endorsing the non-free OS, as well as continue condemning OpenBSD. That is by no account what happened however.

    >At best, what you’re talking about is anecdotal evidence drawn from a relatively limited >number of hardware[/firmware] configurations — it would be a quite illegitimate step to then >propose that this binary-only firmware image-free kernel is suitable for use in conjunction >with the vast plethora of hardware in use today (and thankfully, kernel libre proponents >don’t make that case).

    No, this is as anecdotal as basic arithmetic. It works, people are using it, I’ve used it, I’ve seen others use it, at least 3 OSes are based on it, if it were really that unrealistic people would be complaining in large numbers, which as yet, they are not. If you want to disprove it try running it and submit your findings to the respective developers if you have a problem. There’s no point continuing to discuss it’s potential viability here when it has moved out of the realm of theory into praxis.

    >The irony, as Cox himself observes, is that the hardware configurations which have been >found to work with this kernel — the very cause of the moral self-congratulation — do so >partially by dint of unmodifiable firmware in the devices themselves!

    That’s not irony that’s a problem that has to be worked out by struggling against closed software in hardware devices. That’s a very old battle and the fight continues… Where is the irony in that?

    >The very one — and quite fascinating reading it makes, too, from the point of view of the >FSF’s political pretensions. If using a hardware device with unmodifiable firmware is in >principle no less morally egregious than using a binary-only firmware image-encumbered >kernel, then, logically speaking, the political stakes are far higher than would be the case if >it were a simple matter of excluding this or that sequence of GPL-violating hex. Rather, a >decisive proportion of modern hardware is itself “unfree” — the information that might allow >one to write drivers, compose firmware or otherwise explore and extend a device’s >functionality remains fatally subject to proprietary restrictions, upheld by the laws regulating >property relations within a capitalist society. So software freedom presupposes hardware >freedom, and hardware freedom presupposes/awaits a revolutionary transformation in >society’s power relations. But the free software movement as embodied by Stallman — >unless it was intended all along as a kind of parodic reductio ad absurdum of bourgeois >property relations — seeks nothing of the kind, since that’s precisely what its licenses >logically and materially presuppose.

    Well I’m not certain of the FSF’s position on devices, my opinion is that it should all be free, of course it would be quite interesting to know their thought on it (though it would not change my opinion in the least), but since we don’t have the FSF here, and that discussion was not amongst them, we don’t know. However I believe that their position is somewhere along the lines of not using proprietary when free alternatives exist. That’s not really the case with a lot of devices right now.

    >Well if it’s a question of credibility, RMS isn’t exactly blessed with a surfeit of the stuff >amongst *BSD devs. A *BSD developer might place equally stringent demands upon his or >her code, and pursue a concept and collective practice of software freedom older than >copyleft, only to be told by Stallman — as OpenBSD’s Marco Peereboom was quite recently >— that the movement for free software began with his efforts in 1983!

    Well this get’s back into what is “free software” which is an endless discussion, that’s a bit far afield I think. However if I remember the original BSD was not so free in the mind of the FSF, and in fact was used in proprietary UNIXes. The freer BSD’s, which are considered free by FSF, came along later.

    Either way, doesn’t really excuse the behavior of OpenBSD here.

  22. Kostelanetz
    May 7th, 2008 @ 2:02 pm

    >> The context was that Stallman had used his position as self-styled leader of the
    >> Free Software Movement to endorse an operating system cursory research
    >> would have revealed as unfree.

    > That’s not context

    Damn right, it’s incontestable fact.

    > Since I’ve never read anywhere that stallman did endorse OpenSolaris I can only
    > assume that, if ever he did, the context would be in line with GNU philosophy.

    ?

    >> But in the eyes of the OpenBSD developers, indeed in the eyes of an impartial
    >> observer, it was surely hypocritical of RMS to endorse a project that by his own
    >> lights implicitly suggests” the use of non-free software, and then proceed to
    >> criticize another project for doing so.

    > Well no observer is impartial.

    I fucking give up.

  23. jeff moe
    July 21st, 2008 @ 7:12 am

    The linux-libre kernel runs on most modern hardware just like the one from upstream kernel.org. There’s a high end SCSI card, it breaks 3D for ATI, some old ISA sound card went out the door, some palm pilot stuff, uh something for the s390 if you happen to have one of those sitting around, etc. The kernel supports an *enormous* amount hardware and something like 30 drivers went out the door.

    The fact is that for the vast majority of users linux-libre supports their hardware just as much as the upstream kernel.

    Alexandre Oliva is thankfully maintaining this kernel now at FSFLA:
    http://www.fsfla.org/download/linux-libre/

    There are generic tarballs there for use with any distro and packages that are fedora/blag compatible.

    -Jeff

    P.S. OpenBSD rules.

Comments are closed