A Wiki for the Planet: Clay Shirky on Open Source Environmentalism

An interesting article. But I'm getting a bit tired with reporters that equate online collaboration with open source. To be fair, Clay Shirky doesn't say a word about open source at all, but the reporter does in the headline. I see it more and more these days: People who want to apply the open source methodology to something non-programming related, then pointing to Wikipedia as their shining example. It's wrong IMHO.

Open Source doesn't just mean that anyone can contribute in an open fashion. It also means that all contributions are peer reviewed before they are accepted into the project. That's where the quality in FOSS comes from: peer review. Linux has that. Wikipedia does not. That's just a free-for-all. Not that Wikipedia is bad but it could be better if it had implemented all of the open source methodology instead of just half of it.

Unfortunately, "Open Source" means much less than you state (and imply). It could be no more than you get to read the source if you so desire, with conditions. Unless it's a GPL like license it's near to a meaningless buzz word that was supposed to give comfort to the timorous corporate types. Solaris and even Open Office are free/open projects, however, do you really think you could get some of your code included? Certainly not easily with mostly paid Sun developers in control*.

* This comment is NOT a knock at Sun, it's just hard to share and in the case of the OS it diminished it chances (at least up to this point).

Quoting:Unfortunately, "Open Source" means much less than you state (and imply).

That depends on how you look at it. If you say "open source as a license" then you're right. But I'm not talking about the license. I'm talking about the open source development model. That's a different beast. A good example: SugarCRM. It's open source; The code is GPLv3. But it's development model is closed. There's no public code repository. No mailinglists. No decent way to submit patches. No peer review. Just code dumps (releases) and a half-assed bug tracker.

Here's a good read about the development model: http://www.producingoss.com/producingoss.html

Quoting:I'm talking about the open source development model.

That boils down to the same thing as 'scientific publishing'. That's probably not what the author was intending to say with open source. The article is about the need of a license for 'actions'; and because the article is about the world lacking such a license - there's no name for such a thing - the author must have been limited to using 'open source' to point out resemblances with other licenses that do exist

Genuine question (I'm not being sarcastic): does this "code dump" mean code is available for people to take away and use? Could you then fork it? I'm not looking to do that, just idly wondering about it.

DiBosco: Yes. Basically, SugarCRM develops it's product behind closed doors. No peer review. No publicly available cvs/subversion/git repository. Nothing. It then makes a release which is GPLv3 and which you can download from their website.

Sander,

Again, no if the license is wrong type. It needs to be a free version that explicitly allows taking the code for another use. Open Source can mean, "look but don't touch". Again the meaning, Open Source is nebulous and too open to misinterpretation. Sure it applies to Sugar, but from the start they believed in free use. Being closed does not imply the code is locked.

Look at some of the older (approved) Sun open source licenses or Apple with the webkit and you will see it is hard to impossible to use the code for your own purposes. With the former, you lose the copyright to anything you contribute, the latter has a mine field of licenses that can make it impossible to use necessary libraries to build a product you want to give away. You can build it, play with it and maybe talk about it. But that could be as far as it can be taken. [Get on the webkit-dev mail list and read about the knotty licensing issues.]

Txt.

TxtEdMacs: Re-read the posts above me.DiBosco was asking about the "code dumps" from SugarCRM, which are GPLv3.

Sander,

When I see Sugar, I think sweet things for kids, which to me originally meant for use on OLPC laptop. When you add the "CRM", that I effectively ignored, you are right when the conversation pertains to it alone. [See partially dyslexic.]

Nonetheless, I really think it dangerous to allow a term as imprecise as Open Source to be used as a descriptive adjective when applied to software. GPL versions 2 and 3 are much more than that term. I would rather not associate such a sloppy term with freedom. Open like Democratic are popular terms embraced by groups that have immiscible goal and practices. The last I heard, MS is very interested in its definition of being Open to All, until the gates clang shut.

Txt

@Sander:

You're wasting your time here, I fear. TxtEd has his mind made up on the subject, and nothing you can tell him about "freedom" is going to stick, outside of the strict definition provided by RMS and company. The idea that they may be actually promoting "it's free only if you use it as we approve" is incomprehensible to said group, as they are "right". That's ok, I wish the "freedom by RMS" camp the best of luck. I just don't see it as being any more practical or less extreme than the MS version.

@dumper4311:

Can we please leave RMS out of this for once? You are the one bringing him and his believes and your view of it in the discussion, while you don't have any idea how TxtEdMacs thinks about it (ok, you may have), and it's not even related to the current discussion (RMS is about Free Software, ESR is about Open Source). Please refrain from telling what others think if you're not sure what they think. I'm not going to tell what you think either.

Anyway; it's not about 'RMS Approved' or not. TxtEdMacs is referring to people like those at MS spreading their 'shared source' as being open source. Most shared source may not be forked. Therefore, not all open source can be forked. Actually, not much can be done with 'shared source'; though the source is open. If somebody once looked at it and writes a program thereafter, he or she could be sued for patent/copyright infringement quite easily. Therefore 'even' ESR points to shared source as a poison pill, though in fact it IS open source.

@hk:

>"while you don't have any idea how TxtEdMacs thinks about it (ok, you may have) . . ."

Stay calm, my friend. My post was in direct reference to TxtEd's words - the expression of his thoughts - thus he has told us what he thinks. Indeed I may have an idea of what he thinks, based on his words. If you read them again, you may as well.

>"I really think it dangerous to allow a term as imprecise as Open Source . . . I would rather not associate such a sloppy term with freedom"

It was TxtEdMacs comparing Open Source to "Freedom" - my response is related to TxtEd's discussion. Read more carefully. RMS is the champion of this particular movement, by definition it's difficult to leave him out of any such discussion of his views.

I agree that "shared source" is not worth much, so don't use it where you shouldn't (we're all pretty much on the same page on that subject). Sander understands that as well, and tried explaining the difference several times. (edit: he was more focused on open development models than open source code) But that wasn't TxtEd's focus. He was explaining how Open Source - in general - is bad wrt 'freedom'. His focus is what I was commenting on.

I'm sorry if that discussion (or this one) has offended you. I'd recommend relaxing a bit, but whatever gets you through the day. :)