Immanence
De Landa tackles Open Source
06/02 edit: Yeah, it’s a hella confusing text, and the Hacker News discussion on it made me consider writing this note. The elevator version is (0) Open source caters to hackers’ needs, not end-user’s needs(1) Open source has only produced cheapie knock-offs of ideas developed in the proprietary market and (2) who the hell cares about source as long as there are open APIs?
The Open Source movement is a bit unfashionable as something to think comprehensively about, in the era of interoperable web applications which are not open source yet fit an “open society” model of the web that many former open source enthusiasts have moved on to. Still, it’s worth pondering this 2001 paper by philosopher Manuel de Landa, whom has later discussed a general ontology for social thought in his 2006 book A new Philosophy of Society: Social assemblages and complexity theory.
In hindsight, De Landa underestimated how strong the link between the hacker culture and open-source was (in a nutshell, open-source software tends to fulfill the needs of hackers, not those of end-users) and appears to overestimate the stability of open-source movement as a social assemblage. But it’s interesting to look at the guts of how open-source was ran over by the “Web 2.0″ project as the (possibly temporary) core of the more general program of an information open society.
De Landa is a “real” philosopher (as opposed to inside-out racconteurs of the social processes in open source development by like Eric “esr” Raymond), and is particularly interesting in that he’s attempting to bridge together the opposing strands of philosophy (”continental”/phenomenological and analytic). To quote the introduction of Intensive Science and Virtual Philosophy, “history and geography have conspired to divide this world [of Western philosophy] into two almost exclusive camps, the Anglo-American and Continental camps, each with its own style, research priorities and long traditions to defend”, and De Landa larger program is “to present the work of the philosopher Gilles Deleuze to an audience of analytic philosophers [...] and scientists interested in philosophical questions”.
De Landa’s account of the open source movement is significantly titled Open-Source: A Movement in Search of a Philosophy. At a first glance he’s merely providing an outlook on the open-source generation and its sociology, but there lies in between the lines the prehistory of a more general thought on social assemblages. Quoting from the open-source essay (emphasis mine),
The power of the hundreds of people that do not belong to this core group [of open-source programmers] lies precisely in the local information that they can bring to bear, information which can only be gathered by users of a program who know what is relevant to them. Like Simon’s markets, these users are a “parallel computer”, a vast geographically dispersed army of programmers working simultaneously (in parallel) finding bugs and, as Raymond puts it, collectively exploring the space of possible program designs.
It’s interesting to see how the demise of Linux as a consumer platform (now relegated to server-side computing and only because it’s the cost-effective solution, given the price of an actual Unix and the shortcomings of Microsoft’s server-space offerings, or computing dissemination efforts like the EEE PC and OLPC projects) contrasts with the starry-eyed hopes of the open-source project. Torvald’s law (”given enough eyeballs all bugs are shallow”) fails dramatically as poor quality assurance and the general lack of priority given to end-user needs. De Landa is aware of this, as he presents the central challenge of the open-source model as the one of expanding its coverage of the space of program designs (given that the vast majority of open-source projects, from Linux up, are knock-offs of proprietary software where the conceptual innovation has already been done):
The development model, on the other hand, has proved itself worthy of certain production tasks (such as rapidly evolving a piece of pre-existing software) but it has yet to show that it can fulfill all the different needs of software production (including the initiation of brand new types of software).
Open-source eventually became passé as a media and computer-savvy users darling with the rise the platform-free, loose-commitment (specially since industry leaders’ denouncement of “data silos”) high information mobility-oriented “subspace of of program designs” that pundits came to nickname “Web 2.0″. The focus moved on from software features to informational convenience.
Sure, Emacs can do everything but the girl and the open-source equivalent of popular proprietary software have more and more open features (Photoshop competitor The Gimp can be scripted in Python), but what I really need (and without the inconvenience of setting up a ssh tunnel) is to have my daily dose of information in convenient form (begetting RSS readers) and facilitating social interaction (begetting first “social networks” and most recently friend-feeds like Twitter and FriendFeed.
FriendFeed is particularly interesting because it illustrates the power of interconnecting protocols. No one cares that the software behind GoodReads is proprietary as long as they can get on their FriendFeed. This is a whole different “subspace of program designs” the open-source movement failed to catch on, being grounded on hacker culture as it is.
Open-source is still alive and kicking the dust, Richard “rms” Stallman still posing as a hippie prophet of a new age all while emphasizing petty issues like claiming his version, long-abandoned by people who have moved to greener pastures as the One True Emacs. Yet we, who are creating a civilization of Mind in cyberspace, no longer care. But have we inherited any (positive) lessons from the golden era of open-source? Maybe — if we believe the “Web 2.0″ movement (in its most abstract form) fulfills the more generally-minded epilogue of the De Landa essay:
But even if the movement failed when confronted with any of these challenges, it would have already proved its worth by showing the potential gains of creatively experimenting with alternative institutional environments and governance structures. Even non-programmers have a lesson to learn from this daring institutional experimentation.
Breadcrumb trails:
Analytic Philosophers, Anglo American, Assemblages, Complexity Theory, First Glance, Gilles Deleuze, Hacker Culture, Hacker News, Hindsight, Intensive Science And Virtual Philosophy, Manuel De Landa, New Philosophy, Ontology, Open Source Development, Open Source Movement, Open Source Software, Philosophical Questions, Proprietary Market, Racconteurs, Research Priorities, Social Processes, Social Thought, Source Generation, Western PhilosophyRelated posts:
Tags: Analytic Philosophers, Anglo American, Assemblages, Complexity Theory, First Glance, Gilles Deleuze, Hacker Culture, Hacker News, Hindsight, Intensive Science And Virtual Philosophy, Manuel De Landa, New Philosophy, Ontology, Open Source Development, Open Source Movement, Open Source Software, Philosophical Questions, Proprietary Market, Racconteurs, Research Priorities, Social Processes, Social Thought, Source Generation, Western Philosophy Discussion
11 comments for “De Landa tackles Open Source”
Post a comment
You must be logged in to post a comment.
Categories
TwitterCounter
I’m sure you have something important to say, but after trying to read and make sense of the very first ..cough .. sentence, I give up.
Try reading your writing out loud - exactly as you wrote it, not as you understand it should be read - and get back to us when you realize that even you can’t understand it.
It’s simple really: open source wins at infrastructure and fails at general consumer goods. Closed source is just the opposite.
I think you can say the same things about socialism and capitalism. Coincidence or conspiracy?
Where has open source innovated? It created the internet from the ground up. For some reason the global network is based on TCP and not on ATM. Hmm..
Your big mistake is that you limit your definition of success to consumer goods. But even there you don’t really get it. OS X for instance has an open source infrastructure and a proprietary consumer interface.
The thing is, the internet was built ground up in academia and the military. Whether the source was open is a moot point — the thing is that the open source movement as such, the one that promotes Linux and Inkscape, etc. — could never have built such a huge infrastructure. Not only the knowledge would never have been built because the incentive is always to fix, not to innovate. They wouldn’t even have the hardware.
And the internet was made usable by the advent of graphic browsers like Netscape. Had hackers “built” the internet, it’d be accessible only at university terminals by people who had bothered to acquire the knowledge to utilize arcane interfaces no one bothered to polish.
Poor communication. You give the impression of someone who is trying to be impressive by using big words and long sentences. The unfortunate result is that your message is hidden by poor grammar, disorganized logic, and lengthy sentences.
But even after taking the time to understand what you have written, I find it naive and uninformed, the product of an inexperienced mind lacking perspective and life experience.
If you want to write, you might do better by sticking to topics you know something about.
Trust me, I’ve done all the grand tour. I too thought it could never fail. I’ve just already grown to be cynical; it’ll take you a couple more years before you think like me. People generally jump off the open source bandwagon after a while — unless they’re rms or something.
Seriously - Who gave Peter Griffin internet access and the LSD-laced thesaurus?
Comments are moderated but I’m approving pretty much anything. There were some pingbacks that seemed bogus, so I spammed them, but don’t take it as “censorship” or anything.
I’m also going to sleep now, so thrash on, I will approve every single non-spammy comment tomorrow, no matter how below-the-belt.
Wait, what? The two hackers who invented the graphic web browser (called Mosaic, of which Netscape was a commercial re-implementation) were not actually hackers because they happened to be college students at the time and thus were part of academia?
If you limit your definition of open source to “software written by teenagers” then yes, everything you say about it is true.
Did Mosaic mean anything, as far as the popularization of the Web goes? Moreover, did the Mosaic team did any further work, like supporting early versions of JavaScript, which do very much matter?
I agree with Jamie jwz Zawinksky when he says in his resignation letter “We are the reason there are URLs in movie trailers. We changed the world”. They did. Emacs didn’t, as powerful as it is. Rsh didn’t.
I’d say you’re probably premature if you conclude that open source has failed. Open source is probably the competition that is keeping the oligopoly of Microsoft and Apple on their toes. Linux has gotten much easier to install and use; I would only use Windows on one machine to use specialized devices that don’t have drivers in Linux, or for games that don’t work in Linux. Since the platform is open, theoretically anything is extensible, not just the API. The biggest advantage might be that one doesn’t need to pay for every little program — one can easily try something from the browsable/searchable “apt” repository (maybe they should make this more easily browsable from outside Linux) for free and continue using it for free forever. The true standards seem to be those things that are free and open.
It is obviously better on the whole when information is shared. If we used a small fraction of our total creativity, I think we should be able to fund our non-military info-products through a “crowdsourced” proposal/reputation system that allots federal funding based on (a crowd of) acknowledged experts’ votes, and then the results can be shared for free with everyone, since it was federally funded and it doesn’t cost anything to copy it. So there are still ways of compensating people for their work, such as paying them for their reputation and proposals, without creating monopolies.
I think a person’s quantified reputation in an area should increase if their recorded opinions are migrated toward by masses of other people (so your reputation increases if you can convince people about your opinions). Besides funding federal info-product proposals (such as for open source software), there would be enormous application for reputation/trust systems, from helping with legislation, to solving the problem of oversight of classified programs.
http://www.jwz.org/card1.jpg