Recessions and Free Software

I’m still not convinced I’ve got much of an idea what’s going on with the economy and global finance and all that, but one thing that seems apparent is that to at least some extent we should be expecting a widespread extended recession.

Naturally, that’s inspired various folks to explain how that’s a good thing for open source in one way or another for open source — much like you could probably expect news that the economy was absolutely fine all along to be equally heralded as good news for open source. I certainly don’t think it’s all good news — it’s definitely been tougher for the lca organisers this year for instance, with more speakers than usual having to pull out, somewhat fewer attendees than hoped for, and not quite as readily available corporate sponsorship as there’s been in the past. So I wonder how that’s all going to add up?

Recessions measure a sustained decrease in GDP, so at its simplest, you get a recession when collectively the average person and business is spending less money over the course of a year (and there aren’t sufficient new people or businesses to make up for the difference). Taking that as a given might mean good things for open source adoption, in so far as the purchase price is cheaper, so people and companies that can switch to free software are able to cut their costs without losing out on features (or businesses that have already chosen free software can keep their costs low, profits reasonable or high, and watch their competitors either catch up or disappear).

Ultimately, if everyone were to successfully switch to free software overnight — and the new software took no time to install, did everything they were used to, and was so intuitive and easy to learn that it didn’t require retraining or any lost productivity, and every single computer user was better off — that would result in a measurable recession simply because less money was changing hands. In that sense, there’s nothing wrong with a recession: it can just be part of the economy restructuring to take account of more efficient and effective ways of doing things. And in so far as open source software is effective and efficient, the fact of a recession likely will result in companies looking seriously at open source as a cost-cutting measure.

On the other hand, it doesn’t necessarily mean good things for open source support or training businesses, unless they’re either already, uh, high-value propositions (ie, cheap), or the costs there are more than made up for from savings from licensing costs. That’s probably a mixed bag for open source companies, but hey, web documentation, source code, conferences and IRC work for me, don’t they work for everyone? And I suspect there’re plenty of savings to be made in many IT budgets before the RHELs of the world start filling up CFO’s pie-charts.

The Fedoras, Debians, Ubuntus, CentOSs, OpenSUSEs and OpenSolarises of the world, otoh, probably get something of a boost though, as far as I can glean. In so far as they’re no-cost, non-budget items, they may get additional users, and hence additional contributions from those users (such as bug reports, extra helpers on forums, useful patches, more advocates or simply more relevance).

Of course, companies cutting budgets is going to hit salaries too, and if it hasn’t already hit significant numbers of people who are paid to work on free software, it will soon enough. If there’re any open source hackers getting paid big bucks to do visionary work that isn’t obviously going to make any obvious practical difference to anyone in the forseeable future, that’s probably going to not last so well, much like Eazel didn’t, though of course its code lives on. I can’t actually think of any projects falling into this category these days though — maybe they’re just all webapps this time round.

But in so far as free software development is done in people’s spare time, having more people with more spare time is likely to result in a net win for free software development anyway; whether that arises because people are just working less overtime, are laid off entirely and are filling in time (and their resume) until they’re working again, or they’re staying in university a little longer because a PhD suddenly doesn’t compare so badly to private sector salaries and stock options.

Donations and sponsorships and other freely provided resources are a bit more difficult though. Getting tens or hundreds of thousands of dollars to help run conferences like lca and debconf, or to fund development programs like GSoC, or to keep things like the Linux Foundation or the SFLC running will be more challenging. Challenging doesn’t mean impossible, or even necessarily hard, but it’ll probably mean things will look more local and unconferencey than like international trade shows, and non-profits will have to make more effective user of smaller budgets, but those aren’t necessarily bad things. I don’t think there’ll be much of an issue with in-kind donations — bandwidth and hardware is insanely cheap for all reasonable purposes these days. At worst I guess it’ll mean some mirror sites scale back or shut down, that distros will need to use P2P systems like bittorrent more often, and more people might have to actually deal with monthly bandwidth/usage limits… But none of that seems likely to have much of an impact on open source development or adoption as far as I can see.

I wouldn’t call any of that a huge win; more a lot of small benefits, with a few difficulties. Mostly it’s due to the free software being free as in beer, fun to do as a hobby, and not very expensive to work on; so in so far as all that applies to proprietary software development, it’ll be better off too. So free/cheap proprietary webapps that can be used as a cost cutting measure will probably also cope fairly well with economic recession, while expensive propeitary software had better be able to justify its price tag in triplicate if they want to stick around: these are the sorts of times people get fired even when they buy IBM

I wish I knew what it all meant for startups versus established businesses, or Australia versus the US or Europe, but oh well. Interesting times, either way.

One Comment

  1. Tim says:

    I think you got this mostly correct. The nominal GDP of the Netherlands and Germany is boosted compared to Australia because the higher heating bills of the Northern European economies show up as “output”, even though this use of resources is not increasing future output. So GDP can be a strange beast (and that’s why purchasing power parity GDP makes more sense). Open source software is hypothetically an innovation increasing productivity. Less is not necessarily spent when a hypothetical open source solution saves money; it will just be spend elsewhere and presumably spent “better”. IT in general increases GDP because it makes people more productive (it’s just machinery), and if you can get the same output for fewer resources outlayed, that’s a productivity boost. Like you hint at, one has to wonder if corporate investment in open source IT will fall victim to a “protectionist” mind-set. So there are pluses and minuses for the open-source model in a downturn. Some open source projects will benefit, and some will suffer. Those with critical mass will do well, I think.

Leave a Reply