At that time, Mark Shuttleworth was advocating that all FOSS projects adopt
a timed release process like Ubuntu's 6-month cycle. I believe Fedora does
this. One of his arguments was that regular releases makes planning easier
for distributors, but from your report I think that GIMP developers also
see the value in regular incremental releases.
In that thread from 2012 I proposed a 3-month release cycle. This is, and
was, a slightly slower pace than that used by the Linux kernel. The number
is arbitrary but I still think the results will be worthwhile. With 3
releases in 2019, a 4-month process may be a good place to start.
There are issues and challenges to doing this, and I think there are
answers to both, but I suggest this only to see if the interest is real.
PS - I use GIMP infrequently these days, but +1 on suggestion for a better
text tool. Given recent font tech (ligatures, font feature settings), it
seems something similarly sophisticated may be warranted here as well.
Perhaps direct import of HTML+CSS makes the most sense, even if it means
bundling an entire browser engine ... then maybe even a text edit tool
written in Electron? That can use browser-based editors? Short
justification: the web is defining modern typography, and raster
applications need to compete. Embrace and subsume...
On Sun, Jan 5, 2020 at 4:19 AM Alexandre Prokoudine via gimp-developer-list
<[hidden email]> wrote: