Gredig and gmisub

Gredig was intended as a communal toy for sharing linguistic expertise.

But it didn't get a community, and activity dropped to zero a few days after it was first announced. This didn't suprise me, I know this is how these things tend to go. But it seems a shame.

So I just revisited it with a mind to fostering the network effects which are oxygen to such a system. I don't want to manipulate anyone into participating against their better judgement, so I rule out any artificial gamification. The question is how to make it convenient for those who might be vaguely interested in participating to do so, and to keep doing so.

Here's what I came up with:

• The per-language lists of unrevised sentences are now in gmisub format.
• The user page, where revisions to your sentences are shown, also has updates in gmisub format.
• The user page now includes a personalised version of the unrevised sentences list on the front page, showing only the languages you are fluent in.

Small tweaks, I know. The thinking is based on my mildly cynical understanding of human behaviour: people *will* selflessly help out others in small ways with no expectation of award, but they typically *won't* go out of their way to find opportunities to do so. So expecting people to manually check for unrevised sentences is overoptimistic. But if one pops up in their gmisub aggregator, they might well take a moment to deal with it. Meanwhile, those who don't use gmisub will manually check their user page when they're waiting for help from someone else -- and while they're there, they might be willing to pay forward the (anticipated) favour.

Do I expect this to work and lead to some real activity on gredig? No.

Now a couple of questions for you, dear reader:

• Has anyone written a gmisub -> atom converter? This would help those of us who are hooked on a gemini-ignorant aggregator like newsbeuter. I expected one to exist, but failed to find any. I started to hack one together in sed, then realised that it has to resolve relative uris, so needs to be a little more sophisticated. I could write one in haskell easily enough, but that seems like overkill, and anyway this isn't a wheel I particularly feel like reinventing.
• Do you have any other thoughts on what could make gredig work?

EDIT 2021-04-10: Well I ended up implementing it after all. But in C rather than Haskell, which is much more appropriate really.

/gmisub2atom/