On 2/27/20 2:44 PM, Mark Sapiro wrote:
I have said before that a much better use of time and resources would be the implementation of a light weight, non-Django web UI for Mailman 3, but I don't see anyone raising a hand to do either.
Let me be more clearer on this.
I have hired a professional PHP developer to begin work on a new admin/forum interface for Mailman 3. The work begins next week. We are only focusing on the admin (Postorius) interface initially. I also am in the process of hiring a front end developer for the new interfaces. I believe in the Mailman project, and as someone who has benefited from offering Mailman hosting services for years, I don't want to see it go away. The problems however are this:
- Mailman 2 is great. The interface for it is very outdated. This turns
people off. UI design has come a long way and people are use to using modern UIs.
- Mailman 2 does need to be ported to python 3 eventually but Mailman 3
is already there so why spend extra time and resources on doing that? That is a good question and the answer may be to look at the way Mailman 3 development is currently being handled. Also Mark Sapiro clearly wants us to put our support behind Mailman 3 which is good and he has my support with that.
- Mailman 2 doesn't integrate well with other applications due to no
REST api which I think is what modern users want these days. Mailman 3 has a REST api which is great but again I am having issues with the way Mailman 3 is being developed.
- So lovers/users of Mailman are stuck between a rock and hard place:
Mailman 2 or Mailman 3? Which way to go?
For me, Mailman 3 is the way to go but I can no longer wait on the two interfaces of Mailman 3 to be brought to modern standards. Postorius/Hyperkitty came out flawed right from the beginning:
All of Mailman Core features/functions not being revealed in
Postorius. This is something I intend to fix quickly with the new U.I. as soon as I figure out what those features/functions are. Anyone want to provide a list of that to me off-list so I can pass it on to my programmer?
- The decision to use Django. Maybe great for Python users but not for
me and perhaps for others. This is also brings additional confusion. Mailman 3 has THREE interfaces: Postorius, Hyperkitty, and the Django admin interface.
- Very poor documentation for Mailman 3 and way too many methods of
installing it which means all kinds of versions of Mailman 3 are in production today because Mailman versions are dependent upon what method of installation a person chooses: Distro Package, Docker, Source, others?
- MM3 DMARC handling seems to have improved from reviews I have seen
but NO BOUNCE PROCESSING. My goodness. How do list managers keep their mailing lists clean? I know how much a hit to an IP address reputation can be done when a server is sending messages out to invalid email addresses. However I don't think I can fix that right? It is a Mailman core issue? So let's say it does get added to MM Core. How long will it show up in Postorius? Especially since not every feature in MM Core is revealed in Postorius already.
Social Media integration via Django is awful.
Hyperkitty just does not cut it in appearance and usability when it
comes to a modern list forum. I am simply unable to compete with the growing number of applications that are being offered that has a better browser UI for communicating with list members.
I think the highest priority is to get Mailman 3 core up to speed in offering everything that Mailman 2 offers such as bounce processing. Then perhaps a whole new approach to interfacing with Mailman 3 core is in order. That part I have decided to work on because no matter how great MM3 core is, if the interfaces are poor then modern users will move on to something else.
I hope I did not offend anyone here or show disrespect to the hours and hours that have been spent on the Mailman 3 project. That was never my intention and I love Mailman and its community of users.
I would be interested to know if the developers of Mailman 3 are interested in the initiative I have taken to develop new interfaces for Mailman 3 that are more modernized and user-friendly.