On Mon, Nov 20, 2017 at 06:59:16PM +0000, Jonathan Dowland wrote: > > and given NeoMutt the privilege of using the mutt package to establish > > itself in Debian and all its derivatives, how about you let mutt > > package users actually *use* my work for once before you move on. > > For most of the package's life in Debian, mutt has indeed been mutt. We > can't speak for derivatives, and we can't be responsible for them > either.
I haven't been involved in the packaging team for a while (largely because of this drama), and I'm comfortable with whatever plan of action Antonio decides here, but I'd like to state for the record: Debian has not been packaging pristine mutt… since forever. Debian has been carrying a number of invasive, new feature patches (both in the stock "mutt" package, and even more so in the "mutt-patched" package) for a number of years. The version right before I got involved in, 1.6.0-1, had 35 patches in debian/patches (6 of which only in mutt-patched), amounting to 14323 lines total of unified diff (plus context). Rebasing and maintaining these patches was a PITA, as was fixing bugs in them, which was my original goal when I started getting involved. I wasn't involved during the introduction of those patches, but my understanding is that these were introduced,because Mutt upstream was unresponsive and effectively dead. Kevin took over later, and has since moved the needle considerably, and has merged some of those patches or added the same features using new and different code. Other distributions and operating systems did the same, and have been for many years. NeoMutt was an effort to collect all of those patches that people were running for years (and their variants), rebase them on top of each other and on top of latest Mutt, clean them up, track and fix their bugs, and provide quilt patch series for everyone to consume again. Its original stated goal, was not to be a fork, but a collection of patches that would keep tracking mutt upstream, with an ultimate goal of merging them back upstream (very similar to our debian/patches workflow). Their homepage still lists that "not a fork" policy[1]. Ultimately, this did not happen, due largely to social differences between the two projects, and in the midst of significant pushback by Kevin, they decided to do more invasive changes, rename their binaries, documentation etc. It's all a very sad unfortunate turn of events. It'd be for the benefit of everyone to have those efforts be merged again and work under the same roof, but sadly, I don't see this happening anytime soon. Renaming the package in Debian seems to be inevitable now, so I concur that this would be a good idea at this point. That said, Debian users have been relying on some of the features only present in NeoMutt for more than a decade now and by the time buster gets released, will be accustomed in even more fancy things shipped by NeoMutt. I don't see what Debian users would have to gain by Debian shipping vanilla Mutt again. The NeoMutt project is a healthier one in every possible aspect anyway: > 1 people working on it, a more vibrant community, more downstreams, easier to work with, technically and socially. Regards, Faidon 1: https://www.neomutt.org/about.html
