LK2008: The values of the Linux community [LWN.net]

By Jonathan Corbet
October 10, 2008

The opening keynote speaker for the 2008 Linux-Kongress was James Bottomley, who presented his views on the Linux community's values. What these values are, says James, is not entirely obvious. Related groups - the free software community, for example - have well-articulated value systems which define them. The Linux community's values are not so clearly expressed, but, he says, they are central to what we do.

James started with a bit of history, noting the the initial value placed on software was entirely commercial. Once the industry realized that software could be worth far more to its users than it costs to create, the proprietary mode became dominant - and that has affected the evolution of programming in general. The value placed on the code by its developers became irrelevant, leading to "paycheck coding." There is no value placed on creativity, and such a model leads to bad code.

Eventually Richard Stallman came along and challenged the commercial view of software. But, during this time, about the only alternative to commercial software was the BSD Unix distribution, and that got caught up in the lawsuit by ATT. So closed software took over; Windows won on commodity platforms, but proprietary software also became dominant in the Unix arena.

In 1991, Linux hit the scene; since then, it has become the most popular and vibrant free software operating system available. In a sense, this is interesting, in that Linux is licensed under the GPL, a license that many companies hate. Apple explicitly chose BSD as the base for MAC OS to avoid GPL-licensed code. But, despite this antipathy, lots of companies use Linux, and even contribute to its development. It is interesting, James says, to look at why that is.

The reason is the Linux community's values. In particular, the community prizes technical merit above all other considerations - including small things like what any company or user would like to have. Also prized is passion; code supported by a developer who clearly cares about it will generally fare better in the review process. If the code quality and the passion are there, the community does not care about much of anything else. Factors like the source of the code or who might benefit from its incorporation don't really matter.

In particular, contributors to the kernel are not required to sign on to any particular belief system or any specific view of freedom. A contributor may have an FSF-like belief in free software, or, instead, be a corporate developer who does not care about software freedom at all. Even the BSD community requires acquiescence with a specific view of freedom. A Linux contributor, instead, need only be willing to contribute the code under the share-alike rules of the GPL.

As a result, anybody can play with Linux, regardless of philosophy or corporate status. We have a community which is defined by contributions, not by a specific set of values regarding software freedom. That has allowed the formation of a very diverse community with a specific shared interest: creating the best kernel we can.

There are some significant benefits from this approach. It forces companies to recognize their engineers' values; that, in turn, makes for more motivated developers. Developers who are interested in improving Linux can get resources and support from corporations. Users get high-quality code from developers who care about what they are doing. Companies get the ability to focus on their little piece of the problem while taking advantage of the community-maintained kernel for the rest; they can also offload their older code to the community for long-term maintenance.

James compared the Linux way of doing things with the US constitution. That document only mentions freedom three times, yet it has become a blueprint which has supported freedom for over 200 years. It is a relatively short document. The proposed EU constitution, instead, is about 20 times the length, before taking into account other documents which are referenced. That document would appear to be somewhat bloated; the goals would be better served by a more concise formulation.

Similarly, the Linux community spends little time talking about freedom. Instead, the focus is on a set of brief principles involving code quality and passion. Freedom is not legislated; it arises as an emergent value inherent in the Linux way of doing things. Linux has managed to bring about software freedom without talking about it, and without imposing a view of software freedom on its contributors. In the process, Linux has succeeded in creating something which is as free - or more free - than the GNU system envisioned by the Free Software Foundation.

During the question period, James wished for a free software advocate who would argue the point with him, but no such person emerged. He will, it seems, have to repeat the talk in a different venue before he can have that debate.

Index entries for this article
Conference	Linux-Kongress/2008

to post comments

LK2008: The values of the Linux community

Posted Oct 10, 2008 21:18 UTC (Fri) by stijn (subscriber, #570) [Link]

Apple is an exception, precisely because it is entirely its own ecosystem. For companies operating
in a larger ecosystem there is much value to be had in open standards and a common platform that
is a commodity. This is achieved by the fact that the GPL removes the possibility of forking off
branches and then closing them off. One of its great strenghts is that branches can by and large
only contribute to the original software (by moving code between branches), and never take away.
The community matters a great deal, but I think the license choice is a huge factor as well in the
success of Linux in the market. It garantuees a level playing field and it helps in making the
platform a commodity. I do no think the license and the community are independent variables.

compare to openoffice

Posted Oct 10, 2008 21:23 UTC (Fri) by johill (subscriber, #25196) [Link]

There was an article by Michael Meeks about openoffice commit stats where he compared it to Linux as well: http://www.gnome.org/~michael/blog/ooo-commit-stats-2008....

Everybody can draw their own conclusions, but I think this (the things James says) is a large part of it.

LK2008: The values of the Linux community

Posted Oct 10, 2008 22:40 UTC (Fri) by JoeBuck (subscriber, #2330) [Link]

Linux developers accomplished great things, but despite philosophical differences, they aren't really that separate or distinct from the developers who do the GNU tools and libraries they rely on; in fact, there's a substantial overlap between contributors to the Linux kernel, and contributors to critical GNU/FSF packages like GCC, glibc, and the binutils.

In the past, the FSF projects have been impaired less by an emphasis on freedom-as-ideology than an overly centralized management style, but that's shifted a great deal, and the success of the Linux development style has been largely responsible. These days, there are plenty of contributors to FSF-owned projects who don't agree with the FSF's full agenda any more than Linus does; likewise, there are plenty of free software purists contributing to the Linux kernel. And even RMS is more of a pragmatist than he is often given credit for.

LK2008: The values of the Linux community

Posted Oct 10, 2008 23:06 UTC (Fri) by riddochc (guest, #43) [Link] (3 responses)

"Similarly, the Linux community spends little time talking about freedom."

I dispute this assertion.

I believe that one of the main things that distinguishes the Linux community from that of other communities is that we talk a lot about freedom. I also believe that this is a group of people that nobody can really speak on behalf of the whole group, and yet almost everyone tries to anyway.

Crowd: "Yes, we're all individuals!"

Random guy in the back: "I'm not!"

LK2008: The values of the Linux community

Posted Oct 12, 2008 15:45 UTC (Sun) by nlucas (subscriber, #33793) [Link] (2 responses)

Crowd: "Yes, we're all individuals!"
Random guy in the back: "I'm not!"

Thanks for the laugh!

It's been some time since I've seen such a good humoristic characterization of the reality.

LK2008: The values of the Linux community

Posted Oct 12, 2008 16:29 UTC (Sun) by nix (subscriber, #2304) [Link] (1 responses)

Everyone agrees: we need another Monty Python.

LK2008: The values of the Linux community

Posted Oct 13, 2008 11:58 UTC (Mon) by ewan (guest, #5533) [Link]

"I don't!"

LK2008: The values of the Linux community

Posted Oct 10, 2008 23:34 UTC (Fri) by lmb (subscriber, #39048) [Link] (22 responses)

Saying that one does not need to sign up to a specific view of freedom when using the GPL is a fairly bold statement, a point I also mentioned to James:

The GPL specifies a very aggressive form of freedom: freedom that must never stop. Whereas BSD allows you to take something into a "jail" - as embedding code into a proprietary project -, GPL will have your head for it, and forces you to open up everything it comes into too close contact with. And the proprietary to GPL step is irreversible.

It is exactly this viral form of freedom - the domain of GPL can only grow, never shrink - which I personally think is why the license is so successful.

(Double licensing, or re-licensing if you get every copyright holder to agree, does not change this argument, as it cannot be applied retroactively nor does it detract from the power of the GPL.)

When I heard Team America sing "Freedom is the only choice now", I definitely thought of the GPL ...

All other points about passion and meritocracy etcetera also apply, of course; but that also works for BSD. Maybe Linux had an edge there in the 90s due to the unclear *BSD situation. But the reason why Linux sucks everything in is that the GPL is a black hole. (Luckily.)

But an emergent value? Hardly. It is built into its very core.

LK2008: The values of the Linux community

Posted Oct 11, 2008 0:12 UTC (Sat) by JoeBuck (subscriber, #2330) [Link] (2 responses)

I was a UC Berkeley grad student in the early 1990s, so it would have been natural for me to wind up as a BSDer, but it didn't work that way. It wasn't just the license. The BSD crowd was brilliant and talented, but they were extremely arrogant and did not get along with either outsiders or themselves (I personally witnessed some of the bitterness between the BSD camps).

Linus was much more welcoming of contributions than the BSDers were, and much more interested in providing support for the kind of hardware the BSD crowd turned up their noses at. The result was that around 1993-4 it was just much easier and more fun to play in the Linux world, even for those of us old guys who grew up with BSD Unix.

LK2008: The values of the Linux community

Posted Oct 11, 2008 11:35 UTC (Sat) by ajb (subscriber, #9694) [Link] (1 responses)

Interesting, since these days, the kernel mailing list has something of a reputation as an intimidating place. Would you say its still more friendly than BSD was back then?

LK2008: The values of the Linux community

Posted Oct 11, 2008 15:33 UTC (Sat) by jlokier (guest, #52227) [Link]

I think the kernel mailing list was less intimidating back in 1994, because there were far fewer contributors so you didn't have to fight for attention or prove every blue-sky idea with a quality working patch before you'd be taken seriously.

LK2008: The values of the Linux community

Posted Oct 11, 2008 16:16 UTC (Sat) by danielpf (guest, #4723) [Link] (17 responses)

You rightly pointed out what was missing in the original article.
Let me expand a bit.

In my opinion the success of GPL over BSD, or else, has almost nothing to do directly with ideology, but with the property of being able to spread faster than others in a given environment, the famous viral property.

Freedom in the GPL sense is a factor encouraging the viral propagation, but this is not necessarily the only and last one. For example, Microsoft's predatory methods are until now more successful in the biological sense.

The software world can be seen as a complex dynamical ecosystem, far less complex than life, but still including similar features like birth, death, reproduction, competition and selection acting on programs.
Life offers many examples of spreading organisms ready to use any tricks to spread better than competitors: fast reproduction, direct or indirect destruction of competitors, parasiting others, etc. No need of ideology there, just methods and acts within a dynamical systems.

Individual freedom is just a notion that has proved superior in human societies over the last centuries, but there is no ground to assume that it will always be the necessary condition to be the best fit for survival. Until Microsoft's empire collapses I may keep doubts that their aggressive methods may be effectively more viral than the GPL.

LK2008: The values of the Linux community

Posted Oct 13, 2008 3:03 UTC (Mon) by k8to (guest, #15413) [Link] (16 responses)

Oh let that dead horse rot in peace.

The GPL is not viral. Viruses reproduce whether the host likes it or not. It's a slur, and an inaccurate one.

LK2008: The values of the Linux community

Posted Oct 13, 2008 6:34 UTC (Mon) by danielpf (guest, #4723) [Link] (15 responses)

So what is your proposition? Infectious, self-propagating, contaminating, ...? I don't mind finding better terms, but for the moment "viral" is the meme that is popular. Typically your unhapiness with it should stimulate your creative side to find a better, superior wording that propagates and makes "viral" really a dead horse.

LK2008: The values of the Linux community

Posted Oct 13, 2008 11:50 UTC (Mon) by csamuel (✭ supporter ✭, #2624) [Link]

Why not "meme" ?

Viral properties of the GPL

Posted Oct 13, 2008 11:58 UTC (Mon) by Felix.Braun (guest, #3032) [Link] (4 responses)

The qualification of the GPL as "viral" is utterly wrong and it's sad to see people who should know better perpetuate this meme.

The GPL only requires you to put any derived works of GPL'd software under the GPL. This means that you are not allowed to take GPL'd software, modify it and then distribute the result under a different license. However, you are free to publish your own changes seperately under whichever license you please. Even if you have once published a derived work under the GPL, you may later change your mind and publish new versions of your work under a propietary license or any other license that you can think of.

So the GPL only does one thing: it prevents appropriation of work by other people. You can not take my code, slap a patch on it and then take the whole thing for yourself. So the GPL's only effect is, that the work other's have placed under it is removed from your control. You are completely free to do whatever you please with your own creations. You just can't climb on the shoulder of giants and claim the whole view for yourself.

Characterising this as "viral" can only be called misleading propaganda, IMNSHO.

Properties of the GPL leading to wider use

Posted Oct 13, 2008 15:06 UTC (Mon) by danielpf (guest, #4723) [Link] (3 responses)

So what is your proposition for a better wording? If you leave the space empty the current memes will just fill it. Actually your new thread title helps the propagation of what you dislike!

The idea to convey is that once GPL is adopted for a piece of software, it cannot be changed back, only expand the GPL realm as the software grows.

Here are a few adjective conveying roughly this idea:
1) disease propagation: infectious, contagious, catching
2) drugs consumption: addictive
3) biological reproduction: fertile, prolific, spawning
4) warfare: invasive

Since English is not my original language, I am convinced better propositions with more positive connotation can be made.

Properties of the GPL leading to wider use

Posted Oct 13, 2008 16:06 UTC (Mon) by martinfick (subscriber, #4455) [Link]

The better word is "copyrighted". None of the restrictions of the GPL are any greater than any other standard non-licensed copyrighted work. The GPL simply adds to what you can do with a copyrighted work, it does not take anything away. So debated the poor use of the term viral if you will, but it neglects the primary point, this "viral nature" is not a GPL specific feature, but a copyright specific feature.

So it's silly to say that the GPL is viral unless you are saying it as an extension of the fact that any copyrighted work is viral. If you insist on using the term viral, simply apply it to copyright and the way it applies to derived works. Point the finger in the right place, at copyright. The GPL takes advantage of this feature of copyright law just like almost any other copyrighted work, especially proprietary software, does!

Properties of the GPL leading to wider use

Posted Oct 13, 2008 18:53 UTC (Mon) by Felix.Braun (guest, #3032) [Link]

Good point, that I was propagating the meme myself :-) I should have been more careful :-)

Anyway, my main point is, that what you discribe is not a property of the GPL or even GPL'd software. It's the lazyness of people who shy away of re-inventing the wheel. People who moan that the GPL is viral are just too lazy to do all of the work themselves. They want to base their work on work done by others and claim the end result as their's to do as they whish.

Nothing in the GPL prevents them from doing all the ground work, coming up with their own solutions to age old problems long solved by some other piece of code. Nothing except their own lazyness. That is not bad at all. As we all know the correct mix of lazyness and hubris lies at the heart of all innovation ;-)

However, it strikes me as pretty disingenuous to start whining that all the people how have come before you, whose work you are building on, somehow force you to choose the GPL. The the GPL "infected" your work without your consent. That the GPL is doing something.

In fact, all the GPL does is to pander to a very powerful human trait: lazyness. The GPL allows you to stand on the shoulders of giants, or, more specifically, those giants allow you to use their previous work by means of placing the fruits of that labour under the GPL. You are by no means forced to do that. So if we'd have to label the GPL by some adjective, I'd say it is "seductive" because it attracts lazy people who like to innovate and allows for that lazyness through the exchange of ideas.

Not "viral". Try "Protective" (vs. "permissive").

Posted Oct 16, 2008 18:36 UTC (Thu) by dwheeler (guest, #1216) [Link]

A common term for GPL-like licenses is "protective"; in contrast, BSD-like licenses are often termed "permissive".

For example, see: http://www.dwheeler.com/essays/floss-license-slide.html.

LK2008: The values of the Linux community

Posted Oct 13, 2008 14:27 UTC (Mon) by rgmoore (✭ supporter ✭, #75) [Link] (7 responses)

So what is your proposition? Infectious, self-propagating, contaminating, ...?

The problem with terms like viral, infectious, contaminating, etc. is that they imply that other software can be brought under the GPL just by associating with something currently under the GPL. That just isn't true, which makes those terms wildly inappropriate. Code only needs to be GPLed if it is a derivative work of another program that's GPL licensed. A better description of that would be "inherited".

LK2008: The values of the Linux community

Posted Oct 13, 2008 15:17 UTC (Mon) by lmb (subscriber, #39048) [Link] (6 responses)

It is a bit more than that. It trumps most other licenses when GPL'ed code becomes embedded into a project. I personally think that "contagious" actually describes the behaviour quite well. Or, if you prefer a better spin, "pervasive" ;-)

Genetic metaphor

Posted Oct 13, 2008 16:18 UTC (Mon) by rgmoore (✭ supporter ✭, #75) [Link] (5 responses)

I would describe the GPL as being dominant, while the BSD license (and similar licenses) are recessive. The analogy isn't 100% perfect- real organisms inherit half of each parent's genes, so they have a chance of receiving the recessive gene from a heterozygous parent, while software projects inherit all of their parents' licenses- but it's reasonably close.

I think that it also does a good job of highlighting the philosophical differences between permissive and copyleft licenses. The whole point of permissive, BSD-style licenses is that the code is supposed to be mixable with anything out there. That only works with a recessive license. Copyleft licenses are the exact opposite. Their goal is to be as dominant as possible so that no other license can replace theirs.

Genetic metaphor

Posted Oct 13, 2008 22:31 UTC (Mon) by njs (subscriber, #40338) [Link] (3 responses)

Oh, I like that one.

Another space of positive terms to explore: tit-for-tat, fair...

Genetic metaphor

Posted Oct 13, 2008 22:41 UTC (Mon) by rgmoore (✭ supporter ✭, #75) [Link]

If you want to get into that sphere, I'd call the GPL (and other copyleft licenses) a "share and share alike" license, while the BSD (and other permissive licenses) is a "gift" license.

Genetic metaphor

Posted Oct 14, 2008 9:55 UTC (Tue) by danielpf (guest, #4723) [Link] (1 responses)

What about "equitable" ? It conveys an idea similar to "share and share alike" in one word, without being negative.

Genetic metaphor

Posted Oct 16, 2008 5:04 UTC (Thu) by lysse (guest, #3190) [Link]

"Mutual" might be another suitable term.

Genetic metaphor

Posted Oct 23, 2008 13:30 UTC (Thu) by anton (subscriber, #25547) [Link]

I would describe the GPL as being dominant, while the BSD license (and similar licenses) are recessive. [...] real organisms inherit half of each parent's genes [...] while software projects inherit all of their parents' licenses.

If you don't take all the code, you only inherit the licenses for the code that you take (which actually might, in some cases, be half of the licenses of each parent).

Moreover, if you, e.g., take only BSD parts from each parent, your resulting software can be distributed under BSD, even if the parents have GPL parts and therefore have to be distributed (as a whole) under the terms of the GPL, just like a recessive trait can show up in a child of two parents expressing a different, dominant trait.

So your analogy is pretty perfect.

viral comparisons considered harmful

Posted Oct 23, 2008 12:50 UTC (Thu) by pdundas (guest, #15203) [Link]

Sure there are aspects of how the GPL works that resemble the way a virus propagates its DNA. But there are two objections to the term that have very little to do with how technically accurate the parallels are.

The term has negative connotations. Just making the comparison sounds like a criticism - one that could be a barrier to use of both the code and the licence.

The metaphor is ambiguous. If you say the GPL is viral, people will assume a whole load of baggage that comes with viral, and start to imagine that it will "infect" far more than it does. This is a gift for opponents of open source and free software, and another barrier for less-aware potential users of the licence and the software.

Basically, the metaphor confuses more than it explains. SO whatever its technical merits, it's confusing and unhelpful in the real world.

LK2008: The values of the Linux community

Posted Oct 15, 2008 19:42 UTC (Wed) by jejb (subscriber, #6654) [Link]

I felt I partially addressed your comments.

The point about GPLv2 is that it's not entirely an agent for the FSF four freedoms which is one of the big reasons the FSF went to GPLv3. The other point I made is that really all we need to make Linux successful is the quid pro quo give back. We don't need the GPL to do this, but it was certainly the first to try.

That the give back requirement represents a significant requirement for freedom I won't dispute, it certainly restricts the rights of the original author (but then so does placing stuff in the public domain). However, my essential argument is that this is a pragmatic rule to live by rather than a statement of "freedom" principles.

I'd also dispute the fact that the GPL is "aggressive". You can put aggressive derivation constructions on it, but they do depend on a definition of derivation that has never been addressed in case law, so it's not a given position. I also argue that the give back requirement will work whether the ultimate construction of derivation turns out to be aggressive or conservative.

If you want to see the whole talk by the way, LinuxMagazin.de now has it on line:

http://streaming.linux-magazin.de/events/lk08/archive/jbo...

this is a java stream with the slides, if you just want the ogg, it's at

http://streaming.linux-magazin.de:8080/lk08/jbottomley.ogg

The sound isn't very good, I'm afraid.

Apple & BSD

Posted Oct 11, 2008 11:30 UTC (Sat) by Felix_the_Mac (guest, #32242) [Link]

"Apple explicitly chose BSD as the base for MAC OS to avoid GPL-licensed code. But, despite this antipathy, lots of companies use Linux, and even contribute to its development."

Apple was shipping an OS over which they wished to retain strict control, GPL software was clearly inappropriate.
Most other companies are shipping something on top of the OS, e.g Oracle's database, Tivo's PVR so the OS is an enabler not the main event.

LK2008: The values of the Linux community

Posted Oct 12, 2008 16:00 UTC (Sun) by nlucas (subscriber, #33793) [Link] (3 responses)

The proposed EU constitution, instead, is about 20 times the length, before taking into account other documents which are referenced. That document would appear to be somewhat bloated; the goals would be better served by a more concise formulation.

Except that the only people in Europe that call it a "constitution" are the the same ones against it (to make the document scarier for the common people).

It's not a constitution, is a treaty made by constitutional experts from dozens of individual countries made in a way that would minimize any change to the countries individual constitutions.

It tries to consolidate all the major ideas from every one of this individual constitutions into a single paper and that also means stripping some ideas that are not general (including some values that although accepted by the majority can not be signed without major changes in the constitution of some countries).

Any way, I'm no international law expert, so call it whatever you want...

LK2008: The values of the Linux community

Posted Oct 12, 2008 21:47 UTC (Sun) by dark (guest, #8483) [Link] (1 responses)

I'll call it a Constitution just like the European commission does :)

http://europa.eu/scadplus/constitution/introduction_en.htm

http://europa.eu/institutional_reform/index_en.htm

LK2008: The values of the Linux community

Posted Oct 13, 2008 11:55 UTC (Mon) by nlucas (subscriber, #33793) [Link]

Have you read the last paragraph of the link you supplied?

Following the ratification problems encountered in certain Member States, the Heads of State and Government decided at the European Council of 16 and 17 June 2005 to launch a "period of reflection" on the future of Europe. The idea was to initiate a broad debate with European citizens. At the European Council meeting on 21 and 22 June 2007, European leaders reached a compromise and agreed to convene an IGC to finalise and adopt, not a Constitution, but a reform treaty for the European Union. The final text of the treaty, drawn up by the IGC, was approved at the informal European Council in Lisbon on 18 and 19 October. The Treaty of Lisbon was signed by the Member States on 13 December 2007.

LK2008: brevity and constitutions

Posted Oct 23, 2008 12:39 UTC (Thu) by pdundas (guest, #15203) [Link]

There may be reasons why the "constitution" is as long as it it - but the original point (on the concise expression of powerful ideas) stands.

It's interesting to note that the GPL seems to be getting more complex as it evolves. Like laws and constitutions, almost.

Regarding the side-issue of whether the EU doc is a constitution or not - the first, rejected attempt was called a "constitutional treaty" in the UK ("It's not a constitution, honestly!" advocates said).

The second one is also a treaty, but its apologists say it's completely different because "it's not a constitution", some of the jobs like EU foreign minister have been renamed, a few more veto rights remain for the member states, a lot of text is now referenced rather than included, and it's worded as a set of amendments to previous treaties rather than a free-standing replacement treaty.

Sceptics point out that the provisions of the new version are almost the same as the old one.

You get to decide -- if you can read the thousands of pages, and you're either an Irish citizen, or a member of parliament in one of the other EU states.

LK2008: The values of the Linux community

Posted Oct 12, 2008 20:03 UTC (Sun) by dvdeug (subscriber, #10998) [Link] (3 responses)

I found "The value placed on the code by its developers became irrelevant, leading to 'paycheck coding.' There is no value placed on creativity; it is a model which leads to bad code." very questionable. Does paycheck building lead to bad buildings? Paycheck coding has a habit of producing code that actually usable and useful to someone other than the developers...and note that many major Open Source programs are largely maintained and extended by paycheck coders.

LK2008: The values of the Linux community

Posted Oct 12, 2008 23:12 UTC (Sun) by ewan (guest, #5533) [Link]

Does paycheck building lead to bad buildings?

Maybe not, but I'd suggest that's because building bad buildings leads to getting sued, and for smaller products making faulty ones leads at least to giving refunds. Building bad software doesn't tend to have those sorts of consequences, so there's less motivation to do a good job.

LK2008: The values of the Linux community

Posted Oct 13, 2008 23:58 UTC (Mon) by flewellyn (subscriber, #5047) [Link] (1 responses)

Does paycheck building lead to bad buildings?

Modern "McMasion" style houses, built within the last few years during America's housing bubble, are uniformly shoddy, hideous, inefficient, overlarge, and badly designed, in my opinion. Given that the impetus here was to get them built fast and sold fast, I would say the answer to your question is "yes".

LK2008: The values of the Linux community

Posted Oct 14, 2008 10:20 UTC (Tue) by dvdeug (subscriber, #10998) [Link]

And you've never met the code that was quickly scraped together for personal use and continued to be used despite the fact that the output always needed to be hand-edited, when it worked well enough for that? Even working off your example, I'd say there is a lot of personal code out there that's "hideous, inefficient" and "badly designed". Furthermore, anything must be judged against the requirements; when "over"large is in demand, and efficiency not considered important, holding those against it is absurd. The tradeoff of "built fast" at the cost of being "shoddy" is not one unheard of in personal code, either.

Even accepting that, one example doesn't prove the whole case. Since virtually all buildings are paycheck buildings, you need to argue that virtually all buildings are bad buildings.

I thought it was a rhetoric question ...

Posted Oct 12, 2008 22:44 UTC (Sun) by robilad (guest, #27163) [Link] (2 responses)

... rather a particularly interesting venue to discuss. The idea that Linux kernel development can be used to argue emergent behaviour one way or another strikes me as odd, given the sample base of one project used in the keynote. My freedom is freer than yours is not a discussion for auditoriums, it's one for pubs. ;)

Apple actually doesn't hate the GPL, they ship quite a bit of the code in Mac OS X under it, and provide the source code for it, just not in the kernel. So the remainder of talk from there seems to be based on a mistaken premise.

Factors like who might benefit from inclusion of some code certainly matter, otherwise some 'stable kernel driver API' would have emerged by through economical interest among proprietary kernel module suppliers in that particular area of the kernel, etc.

Apple does not hate GPL... Apple hates freedom...

Posted Oct 13, 2008 8:15 UTC (Mon) by khim (subscriber, #9252) [Link] (1 responses)

If you think about it then you'll understand principal difference between kernel and userspace tools. GCC, CUPS, even WebKit are not really vital for the lock-down schemes. They all can be limited by kernel. Kernel, on the other hand, controls everythig in the computer: without support from the kernel side you can not really limit freedom of user. All such tries can be effectively bypassed by appropriate kernel module. Because kernel controls any access to hardware it can emulate said hardware. As long as kernel is free the whole system is effectively free, if kernel is non-free the whole system is effectively closed! Creator of kernel (even free kernel) can cripple said kernel (like Apple does), but it's not very convenient to do with GPL kernel.

Of course choice of kernel was not really up to Apple: they bought Mac OS X kernel and when the said kernel was in design stage linux was not an option: that's was done in 1985! Linus was a teenager back then and still had few years to learn before he'll try to create it's own "clone" of minix...

But today this problem is acute (linux usage is widespread) - and companies invented few "creative" ways to solve it:
1. TiVoization - freeze the free software and make it non-free this way.
2. Phone industry way - separate "don't tamper with" parts and put them in separate hardware box.
3a. Microsoft way and
3b. IBM's way - put freedom-restricting libraries in a box and connect this box directly to the user program.

So far only Phone industry way and IBM's approach work (first one - in huge number of smartphones around the world, the second one - in PS3). Tivoization (implemented by iPhone and/or PSP) and Microsoft's way of control do not work very well: if you put big and complex system in the "forezen state" sooner or later some bugs in it can be used to unfreeze the system, but as iPhone shows that the approach Apple favors is still classic "freeze everything" one...

Apple does not hate GPL... Apple hates freedom...

Posted Oct 13, 2008 10:23 UTC (Mon) by robilad (guest, #27163) [Link]

Well, not really - the Darwin kernel is Open Source, etc. and is heavily used by the Hackintosh community, for example. Apple keeps publishing the xnu sources along with the rest (including all the other APSL licensed components in the system) for each OS update.

The non-free parts of Mac OS X are way up the software stack, where Apple's got a unique selling point. The kernel isn't, really - it would be entirely possible for Apple to do the same thing on top of Linux (like Google does right now for Android, where pretty much everything on top of the GPLd Linux kernel is kept proprietary & locked down), if they cared about hacking on the Linux kernel.

Apple doesn't, though, as it's much easier for them to hack on a kernel they have control over - less communication & negotiation overhead.

Freedom is legislated: GPL

Posted Oct 13, 2008 17:42 UTC (Mon) by martinfick (subscriber, #4455) [Link] (1 responses)

"Freedom is not legislated; it arises as an emergent value inherent in the Linux way of doing things."

Hmm, me thinks the GPL does exactly that, it legislates. There is no need to debate what freedom means in the community (even though it still is debated.) When it comes down to it, the GPL legislates exactly what it means as far as the linux kernel is concerned. You don't have to agree with what freedom means, you just have to abide by what the GPL says it means if you want to participate (distribute code). It certainly seems legislated to me.

Freedom is legislated: GPL

Posted Oct 23, 2008 17:50 UTC (Thu) by Duncan (guest, #6647) [Link]

That sentence brought me up short as I read it as well. And after a
minute's contemplation, I came to the same conclusions you did, that the
GPL itself is the legislation. It's only within the context of the GPL
that the whole Linux (kernel) community exists, and it is precisely
because it is preconditionally agreed to for the purposes of Linux
development, that the whole situation as described, people who emphasize
freedom /as/ freedom, people who couldn't care less about that but are
just interested in Linux for the practical reasons, and the majority
somewhere in between, exists. As you said, people don't have to agree
with it, only to abide by it, if they want to participate in the
community.

Duncan

LK2008: The values of the Linux community

Posted Oct 14, 2008 12:48 UTC (Tue) by NAR (subscriber, #1313) [Link]

In particular, the community prizes technical merit above all other considerations - including small things like what any company or user would like to have.

The distributors can (and in a lots of cases they do) add stuff to their kernel that a company or a user would like to have. Since the Linus (and -stable) kernels are not made for the users, but for the distributors, I think some (or most?) users end up with a kernel where not the technical merit is the most important...

LK2008: The values of the Linux community

Posted Oct 17, 2008 2:59 UTC (Fri) by gnu_andrew (guest, #49515) [Link] (3 responses)

It's interesting how this article manages to completely rewrite history to make it sound like the FSF and the Linux community are completely distinct from one another. Richard Stallman didn't 'eventually' come along and challenge the 'commercial view of software'. He started the GNU Project with the aim of providing a Free clone of the existing UNIX operating systems. This wasn't non-commercial, as they initially made money by charging for the duplication and distribution of the programs.

Linux didn't come out of nowhere in 1991. Its development was based on Andrew Tanenbaum's MINIX, with the aim of extending it from an educational to a full kernel, and only when linked to the rest of the system already produced by the GNU Project was it actually usable.

The Linux community clearly has concerns about freedom e.g. the tainting of the kernel by proprietary modules. While you may want to distinguish their thoughts from the more stringent views of the FSF (or more specifically, RMS), you can't just rewrite history in order to do so. Linus was not a hermit writing Linux in a cave from which he suddenly emerged in 1991; at least as far as I know...

LK2008: The values of the Linux community

Posted Oct 17, 2008 23:24 UTC (Fri) by nix (subscriber, #2304) [Link]

Linux didn't come out of nowhere in 1991. Its development was based on Andrew Tanenbaum's MINIX, with the aim of extending it from an educational to a full kernel

I don't think so. Minix just provided an OS to run the development environment on before Linux was self-hosting.

LK2008: The values of the Linux community

Posted Oct 18, 2008 16:51 UTC (Sat) by jejb (subscriber, #6654) [Link] (1 responses)

> It's interesting how this article manages to completely rewrite history
> to make it sound like the FSF and the Linux community are completely
> distinct from one another. Richard Stallman didn't 'eventually' come
> along and challenge the 'commercial view of software'. He started the GNU
> Project with the aim of providing a Free clone of the existing UNIX
> operating systems. This wasn't non-commercial, as they initially made
> money by charging for the duplication and distribution of the programs.

You should probably look at the slides and talk before jumping to conclusions like this. If you look at slide 6, it says everything you did above plus a bit more. Realistically, when you have one page to summarise an hours talk, I don't think "Eventually Richard Stallman came along and challenged the commercial view of software" to be an unfair summary, especially when the history was just an introduction.

> Linux didn't come out of nowhere in 1991. Its development was based on
> Anndrew Tanenbaum's MINIX, with the aim of extending it from an
> educational to a full kernel, and only when linked to the rest of the
> system already produced by the GNU Project was it actually usable.

Neither of those is actually a true statement. Firstly, as has been stated many times since the Alexis de Toqueville propaganda (and confirmed by Andrew Tanenbaum), Linux isn't derived from Minix. You could say it's "based on" MINIX in the same way that GNU is "based on" Unix (see ast's nice history), but basically that means having similar interfaces.

You don't have to take it from me, here's the horse's mouth:

http://www.cs.vu.nl/~ast/brown/

Linux also had a Minix emulator, and was basically usable with a full Minix userspace.

> The Linux community clearly has concerns about freedom e.g. the tainting
> of the kernel by proprietary modules. While you may want to distinguish
> their thoughts from the more stringent views of the FSF (or more
> specifically, RMS), you can't just rewrite history in order to do so.
> Linus was not a hermit writing Linux in a cave from which he suddenly
> emerged in 1991; at least as far as I know...

I still can't see your point. There was no connection between GNU and Linux other than you could run Linux with both a GNU and a Minix userspace, the former ultimately proving to be more popular.

The history was really just an intro. The main point was that the Open Source principles of Linux (excellence in code and give back) directly ignited the open source revolution to a much greater extent than either the FSF or the BSDs.

LK2008: The values of the Linux community

Posted Oct 18, 2008 23:38 UTC (Sat) by gnu_andrew (guest, #49515) [Link]

>> It's interesting how this article manages to completely rewrite history
>> to make it sound like the FSF and the Linux community are completely
>> distinct from one another. Richard Stallman didn't 'eventually' come
>> along and challenge the 'commercial view of software'. He started the GNU
>> Project with the aim of providing a Free clone of the existing UNIX
>> operating systems. This wasn't non-commercial, as they initially made
>> money by charging for the duplication and distribution of the programs.

>You should probably look at the slides and talk before jumping to >conclusions like this. If you look at slide 6, it says everything you did >above plus a bit more. Realistically, when you have one page to summarise >an hours talk, I don't think "Eventually Richard Stallman came along and >challenged the commercial view of software" to be an unfair summary, >especially when the history was just an introduction.

I was referring to the article, which contained the offending text, and not the slides themselves. I think there are many more ways of phrasing that brief history in the space allowed that would more accurately reflect RMS and the FSF's contribution. This makes it sound like RMS came along late to the party and just muttered some stuff, rather than founding the Free Software movement about 7 years before Linux came along.

>> Linux didn't come out of nowhere in 1991. Its development was based on
>> Anndrew Tanenbaum's MINIX, with the aim of extending it from an
>> educational to a full kernel, and only when linked to the rest of the
>> system already produced by the GNU Project was it actually usable.

>Neither of those is actually a true statement. Firstly, as has been >stated many times since the Alexis de Toqueville propaganda (and >confirmed by Andrew Tanenbaum), Linux isn't derived from Minix. You could >say it's "based on" MINIX in the same way that GNU is "based on" Unix >(see ast's nice history), but basically that means having similar >interfaces.

>You don't have to take it from me, here's the horse's mouth:

>http://www.cs.vu.nl/~ast/brown/

>Linux also had a Minix emulator, and was basically usable with a full
> Minix userspace.

I'm well aware of the whole Alexis de Toqueville debate and I think you just misconstrued what was a bad choice of words on my part. When I said 'its development was based on MINIX', I didn't mean as in the code itself, but as in that's the platform on which the development work took place. Tanenbaum's states this in the article you referenced:

'He [Linus] had my book, was running MINIX, and undoubtedly knew the history (since it is in my book)'

My understanding is that he used MINIX and it motivated him to develop his own kernel which would go beyond its limited educational purposes. One part of that change is moving away from a microkernel structure. This is the bane of the HURD and, being still mainly a research idea now, it definitely was then.

>> The Linux community clearly has concerns about freedom e.g. the tainting
>> of the kernel by proprietary modules. While you may want to distinguish
>> their thoughts from the more stringent views of the FSF (or more
>> specifically, RMS), you can't just rewrite history in order to do so.
>> Linus was not a hermit writing Linux in a cave from which he suddenly
>> emerged in 1991; at least as far as I know...

>I still can't see your point. There was no connection between GNU and >Linux other than you could run Linux with both a GNU and a Minix >userspace, the former ultimately proving to be more popular.

>The history was really just an intro. The main point was that the Open >Source principles of Linux (excellence in code and give back) directly >ignited the open source revolution to a much greater extent than either >the FSF or the BSDs.

Things ignited because all three contributions came together to make a working system; Linux as a kernel, the toolchain, C library, etc. from GNU and the networking stack from BSD. That was my point; that each party deserves credit, and we wouldn't be where we are today without all three.

Without getting into a whole debate about Free Software vs Open Source, I'm sure the FSF would be glad not to have ignited the open source revolution...