I think what stuck with me most from this discussion that, in case of doubt, prioritise user community over developer community. It sounds very obvious - be used, be relevant, evolve as needs evolve - but for a long time there was (or still is?) a narrative that part of the inherent sustainability of open source projects is that contributors can join, and that somewhat magically volunteers will sustain it, if only everything was well-documented.
This is not to say contributions - as code, through documentation, promotion - are not welcome. On the contrary!
But I think first and foremost we ought to be thinking about what makes a tool relevant and practically useful for those documenting. In fact, they need to be part of that discussion, and we need to have it on an ongoing basis.
Then, I think the question is one of financial sustainability. This is easier the more relevant a tool is, as it helps the argument that funding would go towards impact, rather than a somewhat abstract set of features. But if it is focused on developing what is new, it gets harder to support essential maintenance, and this is where I see most projects struggle most.
One way of addressing is through earned income or fees for service that have good recovery for these “hidden cost”. This, however, is also far from easy and brings with it the risk of mission drift, as it advances the roadmap of those that are willing and able to pay over where there is the most potential for impact.
It also opens the discussion on where to draw the line in terms of accepting work. If we think of the potential market for investigation/documentation tools, it is not completely crazy to see one with law enforcement or private investigators. While potentially great to have income that supports work for underresourced civil society organisations, at least for us that is not where I see our mission and, on the contrary, see big risk to undermine what we are hoping to achieve.