5 ms·
> One, I don’t know that RECAP’s expenses need to be particularly high. Open source software development often does not entail paying developers anything at all
by morisy 9y ago
> One, I don’t know that RECAP’s expenses need to be particularly high. Open source software development often does not entail paying developers anything at all.
It's incredibly hard to run an open source project like this with zero paid developers. I've done it! And others have too, but it's nearly impossible to pull off in the long term and having paid staff with some kind of recurring revenue is almost essential to keeping these kinds of projects alive and growing.
Sure, you can launch that kind of project and run it for a few years, but over time you have customer support queries to deal with, obnoxious bugs that take hours to ferret out, uptime and DevOps challenges that will ultimately otherwise fall under one volunteer who, if unpaid, has a day job and outside commitments and gets sick or bored or run down.
> With the right motivation including public praise, some people may be inclined to donate their skills. Certainly RECAP needs new features and improvements from time to time, but most such improvements should last indefinitely once built, reducing RECAP’s ongoing expenses.
What's the model here for having an organization with zero expenses run this kind of broad-based public good?
It's very easy to underestimate what it takes projects like RECAP to operate, or to think that open source contributions can bring costs close to zero, but the reality is that these projects are very hard to keep running on the cheap, which is why we keep seeing small projects — even when that get a lot of praise and acclaim — shutter.
And relatively speaking, RECAP is an incredibly cheap project already: It's got one full-time staffer, and in 2016 its expenses were under $100,000:
https://free.law/pdf/taxes/2016-990-EZ.pdf https://free.law/pdf/taxes/2016-990-EZ.pdf
RECAP is still making its data broadly accessible for free, while also trying to find some kind of sustainability model beyond a hope and a prayer. Donations and grants come and go, and while open source contributions can be helpful they can almost never keep a centralized service like this running for the long-term.
- morisy 9y agoTo follow up, look at percentage of contributions to CourListener from just one individual vs. other volunteers: https://github.com/freelawproject/courtlistener/graphs/contributors https://github.com/freelawproject/courtlistener/graphs/contr... "The community will keep it running for free" sounds great until you've actually tried managing that community.
- thinkcomp 9y agoThe model is RECAP before Free Law Project took over. It was backed by Princeton (which didn't spend much), had far fewer expenses (certainly no one got paid $90K a year; it was a grad student project), and moved forward code-wise via the coding challenge my foundation sponsored which cost one-fifth as much. See https://free.law/2013/04/02/two-recap-grants-awarded-in-memory-of-aaron-swartz/ https://free.law/2013/04/02/two-recap-grants-awarded-in-memo.... The problem here is that RECAP may need money, but it is not being transparent about how it gets its money and is misleading users as a result. Mike has made deliberate choices about how he runs FLP. I run Think Computer Foundation and PlainSite, which does the same thing as CourtListener. PlainSite is financially self-sustaining. CourtListener is not. There are alternatives to FLP's new "model."