This will be a controversial take here on HN: I'm not too excited about governments getting directly involved in the development of software, let alone open source. With possibly a few exceptions (internal software for national agencies, etc.), it is way outside their area of expertise. I think it would be better to pay a vendor like Redhat or SuSE or Cannonical to do it. And, the gov't can write the support contract such that for X EUR per year they get Y competent developers to work on LibreOffice, or whatever they like.
This is hilariously ignorant, especially when it comes to France.
There are a ton of open source government projects from various agencies and contributing universities - from the government SSO (https://github.com/france-connect) to the Covid contact tracing and health pass management app (https://gitlab.inria.fr/stopcovid19) to the tax code to the unemployment app to a million other things (https://code.gouv.fr/sources/#/awesome). And all of them are good, usable, and (almost always) with permissive licenses.
Why hire an external vendor that has to add a profit margin, and lose the competency when that vendor changes for the next contract, or become a hostage to them? You literally can only lose.
You realise that when governments write software, they just hire software developers, and designers, and project managers like any other organisation does, right?
They're not just asking around in parliament "so who has dabbled in python?" or what have you.
The governments fund projects which are of public interest. They don't actually have development teams in-house.
This is much like the EU funding open source projects of public interest through grant calls.
It's exactly what we want: to fund individuals who's interests align with the public. In fact, it would be great if there were a browser which matches this criteria: publicly funded and designed for the average citizen (as opposed to designed to maximise ad revenue).
> I think it would be better to pay a vendor like Redhat or SuSE or Cannonical to do it.
We don't want governments to fund for-profit corporation. These corporations typically have interests opposite to end-users. E.g.: less digital rights, less digital autonomy, more vendor lock-in, and solidifying their position of power.
In an ideal world, you'd have none of these type of organisations, but much smaller teams and individuals working on individual projects which can inter-operate.
I have worked on open-source software that was government/university funded. It's not uncommon in Europe. And yes, typical death-by-committee issues exist, but there is something to be said for a piece of software that holds people's data and is not outright owned by one government, corporation or random group of cats.
I don't know about the US, but I can see some crucial user data software suites moving to an open-source model where nobody has absolute power over or ownership of the data.
How to get such a multi-player project organised efficiently without burning through a load of money and time is another matter...
