2007-04-16
I originally replied to Chris Messina’s post on The relative value of open source to open services via a comment on his posting. My follow-up was getting long enough that I figured I’m better off posting it here…
How does this situation though (effective data lock in through lack of effectual data portability) affect the developers trying to make a living?
Oh, don’t get me wrong. I have no problems whatsoever with a developer electing to charge for hosting or anything like that. I’m not keen on the start-open-then-close-it approach, but it’s a free country (WARNING: “free country” offer not valid in all jurisdictions). I was merely commenting on the value of the source vis a vis an open-enough service.
instead of Ilija being able to make this into a business and leverage the value of his work, people should take his code, fork the project, and leave him in the lurch, having already invested a great deal of time and effort and having no income to show for it.
Legally, Ilija has all the rights to do what was done. I’m a believer that open source projects have a social contract beyond the legalese of a license. It’s the same sort of social contract that exists when firms define standards — witness the recent outcry when Netscape wanted to stop hosting the RSS 0.91 DTD at the URL they told everyone to use back in the day. It’s the same social contract that exists when an open-enough free service starts charging for the formerly-free stuff. There’s nothing legally to prevent breaking social contracts, but you can’t exactly expect those on the other side of the contract to necessarily follow along without complaint.
How is he supposed to sustain himself if we can’t figure out a model that works for both open source advocates and the developers writing the code?
There’s lots of potential revenue sources available to open source developers, whether operating as sole proprietors/hobbyists, non-profits, or full-fledged businesses:
- Fundable.org-style fundraising for particular development efforts (or, organize as a co-op, with paid-in co-op members getting to vote on directed development) (or, try one of those “completion bond” systems, though I can’t think of anyone who’s tried that particular approach).
- Paid hosting service, as suggested by JH in your original posting (or sponsored hosting service if you want it free for users, following the PBS model)
- Sales of ancillary related materials, such as a book
- Sales of ancillary unrelated materials, such as CafePress stuff
- Donations (if organized as a non-profit)
- Sales of add-ons (e.g., aC Web interface is free, but a desktop client with hooks into Outlook costs $X/year, and I’m not an aC expert so forgive me if this particular example is off the mark)
- Sales of compiled binaries (not relevant here, but more so for desktop apps: give away the source, but sell pre-compiled binaries)
- Sales of telephone support or other forms of priority support
- Sales of consulting services not strictly related to the project (i.e., open source as resume fodder for freelancers)
- Sales of consulting services tied to the project (e.g., custom unusual authentication systems, custom modules for use in an intranet environment)
And these are just the ones off the top of my head. None require closing up of the original open source project.
Now, it’s entirely possible that pursuing these lines of revenue will take as much time as Ilija spends on a regular “day job”. In that case, “monetizing” aC is more a vehicle for independence and focus than it is a way to spend more time on the open source project itself.
2007-04-17 at 9:03
Thanks for your follow up.
In the examples you cites, I’d be interested to see whether or not you could make a living doing these things exclusively. On the contrary, I’ve seen many developers take those routes only to end up consulting and/or getting a day job, which kind of defeats the purpose.
Perhaps it’s just super-difficult to pursue the dream of building a successful *and* lucrative open source project… but I think do think it’s worth it, even if the current economics are against it.
Put it this way: if a typical developer makes $60K-$80K on the low end, do you think, given the options you suggested, he’d be able to make that much in a year?
2007-04-17 at 10:30
“I’d be interested to see whether or not you could make a living doing these things exclusively”
As an example, WordPress seems to do OK with the hosting model. The Apache Foundation presumably runs the foundation off donations. Wikimedia does too, though in their case the emphasis is far more their variation on “open service” than on the MediaWiki open source itself. On the smaller end, you have people like Karsten Lentzsch, who has open sourced parts of his JGoodies Java Swing toolkits, and sells the others.
“On the contrary, I’ve seen many developers take those routes only to end up consulting and/or getting a day job, which kind of defeats the purpose.”
No doubt. But that’s not significantly different than being an actress, or a musician, or any number of interests where there’s a fairly steep hurdle for making it a full-time career.
“Put it this way: if a typical developer makes $60K-$80K on the low end, do you think, given the options you suggested, he’d be able to make that much in a year?”
A typical open source developer with a typical open source project? Probably not. That’s because the typical open source project doesn’t attract that much of an audience, and audience drives much of the potential revenue off my list.
Would Ilija of activeCollab be able to make a go of it using these techniques? No reason why not. He’s well-suited to the WordPress-style hosting model, either charging for the base service, or leaving the base service free and charging for fancier features (like WordPress does). Also sell $100/year subscriptions which gives the buyer access to updates to an ebook on aC and an hour of telephone tech support. Add to it some form of paid directed development, and it’s eminently possible aC could cover a solo developer.
I suspect what you’re really concerned with is scalability of the model. There’s no question that most of what I list there doesn’t scale, certainly not with the hockey-stick type growth curves that venture capitalists like. Scalable models aren’t readily genericized, other than a few that have their own execution challenges (”run a hosting service”, “ads, ads, ads”, etc.). For example, while Second Life is presumably bringing in revenue, it’s hard to genericize “make a virtual world with a virtual economy and make money off selling virtual real estate”…