It seems it wasn't moved; the intention wasn't to make the final release on New Year's, but to have the release occur around PyCon. Rather it seems what they meant was that they'd stop fixing issues (except critical ones/blockers) past new year's.
Accidently realised I had a lot of code that still worked under 2.7 when I provided the wrong interpreter path. The tests passed with only a few failing.
I doubt I'll ever update them to run under 2.7. Ironically I still support some scripts under 1.x for a bunch of industrial systems that aren't networked. It will be nice when July comes along and they all vanish and replaced with the 3.x goodness.
If there's legitimate need and enough support, it will be. But I haven't seen any company announcing their fork and services so far. It may not happen on a larger scale.
But Python is not GPL or similarly restrictively licensed, so RH is not obliged to publish the changed code they produce for their paying customers. Whether they will do e. g. via CentOS I have not investigated.
to be fair, all the CVEs I see in the last few years are related to urllib (https://www.cvedetails.com/vulnerability-list/vendor_id-1021...) - there are tons and tons of applications which aren't even using any kind of network communications, does it matter for them that Python gets updated ?
https://www.python.org/psf/press-release/pr20191220/