Prior art at T244075

As always, the first, and hardest, step is

WMF stewardship: Find at least one WMF team (or staff member on behalf of their team) to agree to offer basic support for the extension for when it's deployed to Wikimedia Production.[1] Do this step before any other steps. Due to technical debt, WMF capacity to steward extensions is limited.

Community-Tech was interested in filling that role 5 years ago. Hopefully they still are.

(Sidenote: I wish the people in charge would stop stifling volunteers' innovations. But there's nothing I can do about it)

Deploying this to all Wikipedias would be risky. Chess is very popular, and without a dedicated team for this extension, bug reports and feature requests could pile up quickly.

A more cautious approach could be to reach out to a specific community to see if they're interested in partnering on a pilot. For instance, since the extension originated (AIUI) as a gadget on Hebrew Wikipedia, perhaps that community would be receptive? (cc @Kipod)

It could be worth suggesting this idea there to see how they respond. If you do, it would be good to clarify that this is still an early idea, since the path to deploying the extension isn't clear yet. It would also be helpful to think about how it might be rolled out (maybe to just a few pages at first?) and what the plan would be if it needed to be removed (like who would revert its use on articles?). In general, setting reasonable expectations is important.

Although there are still deployment blockers (no security review, no WMF sponsor), my hunch is that these would be easier to address if there was a clear plan for a focused pilot on an initial wiki.

I think it's inevitable that deployment would spark feature requests (fairy chess pieces anyone?), but there are bugs here that have been open for years and years, but nobody's fixed them yet. There is no deadline. https://en.wikipedia.org/wiki/Wikipedia:There_is_no_deadline

There might be a case for deploying it to a testwiki (https://testwiki.wiki/) first.

Then might Commons be a suitable place? I think if it were deployed to a small Wikipedia first, it might get ignored.

Also, there is the example of Extension:Graph for what needs to be done to undeploy an extension if it all goes wrong. It's not an ideal scenario, and we can hope it never occurs, but it wouldn't be too large a task if it needed to be done.