The problem with open source, part I

Since my first interview, I’ve learned something very important about blogs: you can outsource content-generation… I mean, provide a platform for co-thinkers with innovative ideas. And that’s what I’m doing today, with an anonymous contributor. Ever since my horrible experience with Linux while travelling last year, I’ve grown suspicious of claims that an open-source operating system could subvert corporate monopolies and make information free. Okay, I had those suspicions beforehand, as I don’t believe there’s any technical fix to capitalist ownership. The drive of capital to expand means that even the most utopian DIY project will be excluded if it’s a threat to monopoly, or appropriated if it’s got some marketable potential.

This discussion does get technical, but I dont know anywhere else where you can get this kind of analysis, and I think it’s an important corrective to some of the stronger claims of DIY software enthusiasts. I’m including photos to break up the text, not to minimize the subject matter. My contributor knows far more about this than I do and, best of all, he backs up my irrational prejudices with hard facts. So I’ll turn the floor over to him, suitably edited to make me sound more erudite than I actually am.sophisticated homer

We began innocently enough. He had sent me revisions for an article I wrote, and I couldn’t see them. I was using Preview on a Mac. He explained that:

The only freeware, apart from Adobe’s own, that can consistently ‘see’ the mark-ups inserted using Adobe Reader is Okular (a Linux application). But unfortunately it can’t edit the mark-ups or insert its own. My guess would be that the mark-up feature is a proprietary connivance, built on top of the Portable Document Format’s non-proprietary open standard, which, for reasons unknown, software developers aren’t especially keen to explore.

Q: It makes sense that a corporation would attempt monopoly control over a technology, to the extent of suing others for attempting to develop/distribute alternatives. But for something so minor as being able to hover over mark-ups and read inserted text? It shouldn’t surprise me that capitalism is deep as well as broad.

We next discussed the Sklyarov case, in which Adobe backed the prosecution of a Russian for developing the means to modify e-books. This was legal in Russia, but not in the US – they nabbed him when he was at an American conference.

A: If Adobe are willing to prosecute a non-US national for activities otherwise quite legal under their own country’s jurisdiction, what chance, say, a US national who distributes a program that inadvertently implements an Adobe patent?

Q: Yeah, they’ll err on the side of caution rather than get sued. And this kind of chill effect seems to spread. Nothing gets in the way of patented data!

A: While the actual abolition of software patents in the US isn’t on the cards, reform of the system probably is, because when royalty-bearing patents were more or less the exclusive preserve of industry oligopolists, a corporation’s willingness to sue a competitor would be held in check by considerations of their own potential exposure to patent suits. Today, however, there are companies (dubbed ‘non-practising entities’) whose function isn’t to sell products but to buy up patents and sue the likes of Apple, Microsoft and Motorola for whatever they can get.

A free software application might find itself in violation of a software patent either way, of course, and if you’ve ever used a commercial variant of Linux (or a version that officially serves the development of a commercial variant) you’ll immediately notice how various elements are hobbled: mp3 support (Fraunhofer Institute), plus the other proprietary codecs; font rendering (Apple, Microsoft); reverse-engineered versions of Silverlight (Microsoft’s answer to Adobe Flash)… It’s been suggested that distributing programs by source code alone would circumvent patents, since distributing what are, in effect, the instructions for creating software would be no more problematic than distributing the patent document itself — which is fine, until you realise that compiling the source code for your PC’s operating system and software applications would take the better part of a working week!

Homer asleep over book

Q: That’s a very good point you make: individuals are supposed to make up for the (externally imposed?) limitations of open-source. I struggled with Linux for a few months last summer and I grew to appreciate the clunky but fully-functioning evil that is Windows… because my laptop was designed for it. Oligopoly indeed. Unfortunately the response of advocates has been voluntaristic: you dont want to hang around for a week for the programs that you use on a daily basis to finish compiling? What kind of libertarian are you? 🙂

A: What never quite fails to amuse with – most – free software advocates (and, in the interests of full disclosure, I’ve been running Linux and FreeBSD to the exclusion of MS Windows since 2006) is their total failure to countenance specifically political obligations, notwithstanding, say, the General Public License’s commitment, via the source code, to freely redistributable and freely modifiable software as a political desideratum.

In fact, in 2007 there was an attempt by key Linux kernel developers, ‘The Linux Driver Project’, to make non-disclosure agreements (NDAs) with manufacturers their default arrangement. NDAs  effectively make the notion of ‘freely modifiable software’ a nonsense, since, in the absence of so-called ‘register level programming’ information, only those party to the NDA can meaningfully modify (which is to say, improve) the very software that permits your laptop to work correctly with its video card, sound card, and power-management system. But free software advocates by and large thought it was a terrific idea, because it would allow more PC hardware to be supported under a (nominally) free software license!

Q: I don’t understand all the technical language, but restricting the hardware for which developers can write code starts a monopoly by any other name. I personally think it’s a weakness of the libertarian streak among open source programmers: so focused on individual freedom from monopoly, they’re willing to countenance freedom to make a profit without considering how those actions will affect others. But not to pick on open source, there’s a lack of understanding of political obligations in many communities, free or not 🙂 It’s really important you can critique that from the inside.

Homer in court

A: Among advocates of ‘copyleft’ software licences – that’s to say, licenses which legally oblige one to feed any alterations to the code back into the public realm under the same contractual terms – there’s very little open recognition of these matters, and for various reasons.

Copylefted code (unlike the support services provided around it) is well-nigh impossible to directly monetise, which has made it popular among the majority of large IT corporations with the important exception of Microsoft. Or, more to the point, it’s become popular among failed monopolists, who from the outset have been happy to support the development of the Linux kernel project, even where they market their own proprietary versions of UNIX – Linux having originally been developed by Linus Torvalds as a reverse-engineered version of UNIX.

That these corporations are willing to invest so much in something the appeal of which is the fact that their principal competitor is effectively excluded from profiting from it (at least according to its business model as a vendor of actual software) is taken by the more excitable copyleft-inclined free software activists as a sign of impending world domination. Given the wantonly simplistic nature of their advocacy (I was once told not to ask awkward questions of political principle, as it wasn’t advancing ‘Linux evangelism’) they’re happy to restrict their demands to more code released under their preferred licence.

Money and power

Needless to say, not everyone in free software is quite that obtuse, but the more perceptive critics come from some unexpected quarters. The principal alternative to copyleft among free software licences is the Berkeley Software Distribution licence, which in effect is a public domain licence qualified by an obligation to respect the moral rights of the author. Unlike copylefted code, BSD-licensed code can be proprietised and incorporated into proprietary code, not least because, unlike as with copyleft, there’s no obligation to distribute any changes one might make to it in the form of source code.

But paradoxically, it’s the very fact that one can directly monetise BSD-licenced code by withholding its source code which has meant its (relative) neglect among software developers working for IT oligopolies. Mindful perhaps that ‘world domination’ isn’t within their gift, rival BSD-licensed free operating system projects such as OpenBSD have been very vocal critics of the moral and political chicanery involved in the corporate patronage of Linux.

Please return for part II…


One thought on “The problem with open source, part I

  1. Pingback: The problem with open source, Part II | The Rootless Metropolitan

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s