Donating to open-source seems like such a good use of charity money. I never give to charity because it always seems so abstract, or there might be better ways to solve the problem; with open source people are usually laboring over it with no recognition, and even a little seems like it has such high marginal benefit.
I'm in-between jobs right now(occupied with a side project), but at some point I'd really like to fund feature development on some open source projects:
* GNUCash has a solid heart, but has some usability issues that make it a pain to use in practice.
* Freenet last I checked had only 1 fulltime developer. And he's probably taking a serious cut to market salary to work on it.
* GHC could stand to have some performance optimization done on the compiler.
* Inkscape or GIMP are handy to have around. Inkscape even has a page describing how you can host a fundraiser for targeted feature development, which is very rare for open-source.
* I don't know that TOR needs much software help, but I wouldn't mind funding some exit nodes. It'd be nice if you could buy a locked-down black-box exit node that you could plug into your wall or something, that was guaranteed not to incriminate you. Maybe outside the scope of this, though.
* Everyone has a little app or site they use where a few people are working without much benefit to maintain something you use all the time.
Is there a good list of needy open-source software?
> I never give to charity because it always seems so abstract, or there might be better ways to solve the problem
It can be difficult to know whether or not you are making an impact with a donation. Fortunately there are sites like Charity Navigator [0] that audit aggregate and audit charities to make things easier. Some sites (like Giving What We Can [1] and GiveWell [2]) make it even simpler by suggesting a few charities that they deem most effective.
For example, one of these top charities is the Against Malaria Foundation [3]. They're a fairly straightforward charity: they receive and review requests for mosquito nets and fulfill orders with donations. When you donate they tell you exactly how many nets were bought with your donation and where they were sent. In this way your dollars (approximately) directly help prevent the spread of malaria. Now, could this problem be better solved with donations to a malaria vaccine research group? Perhaps. For the time being, mitigating the issue with nets seems like a decent interim solution.
My point is that there are many, many charities out there doing great work. It's worth looking into.
(Another note: I don't mean to imply that you shouldn't give to open source projects. I am just trying to say that you could also consider giving to traditional charities.)
My problem with that is that they mitigate the issue, they don't end it. As great as against malaria is, there will always be another family that needs nets, no matter how much you donate it will never be enough. One a feature has been added to an open source software, it is available to all those who need it.
I get your point but that comes across as very callous to me, valuing a feature in software that benefits a small fraction of the most prosperous people in already-prosperous countries over the health of a family.
Consider reframing it: With a small donation you solve the problem for a family (where the problem is the risk that their children will die!), and you can solve it over and over just by repeating the action - it's the definition of scalability. You won't have to keep adding different colors of net to make that family safe.
But there's always another software feature to be added, the job is never done - and when you add a feature it's only for the one package! If you want to do it again you have to start from scratch.
Also, when you consider improvements in your projects, you no doubt consider both cost and impact. Compare the impact of the feature-add your donation might make in an open-source project, to the impact saving one or more family's lives.
The bigger issue for me is that they have such limited resources they can only investigate a tiny fraction of charities. If you have a personal interest in a certain area or want to find global optima the net is not cast wide enough. Very glad they exist though, could help raise standards everywhere.
Every piece of open-source software needs donations of some form (time/money/hardware/etc). Donate to the ones you use. If they accept your money, great. Many only accept time though (in the form of patches etc).
Tor needs help with software too.
Tor relies on exit node diversity for security. Jurisdiction diversity, organisation diversity, sysadmin diversity, hoster diversity, hardware diversity, OS diversity etc. Funding exit nodes reduces a lot of those. It would be best to setup a node yourself and convince people you know to setup nodes. If you still want to fund nodes, the Torservers project accepts money to do that. Tor itself had/has an experiment on funding exit nodes. Exit nodes tend to attract law enforcement in some cases, there are no guarantees. Relay nodes are much safer.
I strongly agree with the Inkscape/GIMP bullet point. In fact I'd say Inkscape/GIMP/Scribus/Blender and I'm probably missing some other tools. It's really a shame how focused design classes are on proprietary software. I've heard the sentence "don't use Blender, it will hurt your chances of employment" a lot and "clashed" with some of the professors here.
I think there needs to be a strong "not Adobe" toolkit and probably something similar for music and film (Blender is good for film actually). Those are creative and content oriented fields that often get overlooked by us programmers but there's so many young folks that need/want good and free tools to express their creativity.
[it doesn't help that digital drawing oards and the like often don't work with Linux so maybe some more hardware as well]
Similar to the TOR point you made, maybe it's not the projects that need the money (eventhough I'd say if in doubt give it to them). More tutorials that focus on the free tools would help a lot in that field. You can actually drive adoption by having better tutorials. If my first 10 hits when I search for "draw comic book character" all lead to Inkscape and not Illustrator tutorials that's a nice win.
I wonder, why they have chosen Freenet? Not I2P [1] as much more modern alternative and active community, bigger and faster network, bigger dev crew.
Also I'd like to mention C++ implementation [2] of I2P client, since it's more important for bringing it on the microcontrollers, IoT, mobile devices, like Tor. It will produce smaller binary and less memory footprint.
> I never give to charity because it always seems so abstract, or there might be better ways to solve the problem;
I decided a long time ago that there are too many good causes, and too many charities. While I don't donate much every single donation I've made for the past fifteen years has gone to the same three charities.
I can cheerfully ignore starving children, tsunamis, and all other good causes precisely because I know that I'm making a (small) difference in some other areas that are both meaningful to myself, and which I've "audited".
Too many charities seem very vague, or produce little measurable impact. I think the key is to pick something local, and something with which you want to be engaged. (Or sidestep the problem entirely and donate time, space, resources rather than actual cash.)
* RNIB, The Royal National Institue for the Blind.
* RNLA, Royal National Lifeboat Association, - their members get my respect for literally risking their lives to rescue others. (Alongside other people such as firemen, mountain rescue teams, etc.)
I'm in-between jobs right now(occupied with a side project), but at some point I'd really like to fund feature development on some open source projects:
* GNUCash has a solid heart, but has some usability issues that make it a pain to use in practice.
* Freenet last I checked had only 1 fulltime developer. And he's probably taking a serious cut to market salary to work on it.
* GHC could stand to have some performance optimization done on the compiler.
* Inkscape or GIMP are handy to have around. Inkscape even has a page describing how you can host a fundraiser for targeted feature development, which is very rare for open-source.
* I don't know that TOR needs much software help, but I wouldn't mind funding some exit nodes. It'd be nice if you could buy a locked-down black-box exit node that you could plug into your wall or something, that was guaranteed not to incriminate you. Maybe outside the scope of this, though.
* Everyone has a little app or site they use where a few people are working without much benefit to maintain something you use all the time.
Is there a good list of needy open-source software?