Hi List :-)

TL;DR  We should move to another mailing list implementation and host. The 
current list will remain as a read-only archive. I propose to move to 
Savannah.  I expect this move to be painless.

---

I have already talked to the people at Savannah. They are willing to create 
a list for Magit even though Magit development happens on Github.  They are 
also very responsive, so unless we get unlucky and hit a lazy day we might 
be able to complete the move in a single day.

I think Mailman on http://savannah.nongnu.org would be a good match for 
this project.  But feel free to suggest alternatives.  Gmane and other 
third-party archives would then replace the current web interface.

Are there other problems that could arise?

---

But why?

In the past I haven't paid much attention to the mailing list. The web 
interface is a pain to use and so is reading html mail in Emacs. Also I was 
under the impression that not much interesting was going on here. Maybe 
that impression was wrong. Regardless it is getting interesting now. It 
certainly is a problem if the primary maintainer avoids the mailing list 
for whatever reasons.   oO

It has also become apparent that some people thought following this list is 
enough to stay up-to-date with current development. Despite some false 
statements (which I have now removed/corrected) hinting into that direction 
that is not the case, development does happens on Github.

Simply switching to another mailing list won't change that.  But moving to 
something that does not actively encourage users to post html mails is a 
perquisite to read the list in Emacs (at least for me) and one cannot apply 
patches from the web interface.

I am writing this message in the *scratch* buffer and will later post 
kill-yank it into Google's web interface, which will then turn it into 
html.  Yes, I could probably configure mu4e to better deal with html mail 
but *everyone* who wants to use the list properly (including patch review) 
would have to do the same.  This is not acceptable if an alternative, which 
makes html mail much less likely, is available.  Doing most of the reading 
using the web interface (which sucks even for just this) and doing the 
patch work inside Emacs is also out of question.

---

If/since the above is unclear/to opinionated there might be objects.  If 
you agree with me, please help me out by providing a more coherent argument.

---

To avoid disappointment later, I have to emphasize the following:  After 
moving to a better list we will not instantly accept patches posted on the 
new mailing list, and to be up-to-date you will still have to follow 
development on Github. Supporting both pull requests and patches on the 
list comes with an overhead which I am currently not able and willing to 
handle. Discussing on both Github and the mailing list also leads to 
duplication but I am willing to make that sacrifice.

The goal is to support both.  But we should first extend Magit to better 
support a git/list based workflow - a first patch into that direction has 
already arrived, thanks.  We should also extend Magit to support Github 
pull requests better - there are several 90% solutions out there attempting 
this.

Or in other words and in good old Emacs spirit: while working on Magit we 
should focus on those features that make working on Magit more pleasant. 
 :-)

  Jonas

-- 

--- 
You received this message because you are subscribed to the Google Groups 
"magit" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
For more options, visit https://groups.google.com/groups/opt_out.

Reply via email to