Re: getmail and Fedora 32

On Sun, Aug 23, 2020 at 03:59:36PM +0200, Stefan wrote in 
<[email protected]>:
>On Sun, Aug 23, 2020 at 06:26:21AM -0400, Remco Rijnders wrote:
>> On Sun, Aug 23, 2020 at 10:23:02AM +0200, Stefan wrote in
>> <[email protected]>:
>> > The problem is the way that the fork is handled, that it registers
>> > getmail6.org and seems to push itself aggressively on distributions.
>>
>> As the person who registered getmail6.org (though not the person doing
>> work to make getmail work with Python 3), I'd like to respond to this.
>> I disagree with your assertion that getmail6 is aggressively pushing
>> itself onto distributions. Do you have any evidence for this?
>
>I don't know if the person who asserts Copyright (c) 2020 is doing this,
>but two of the people involved in pitching getmail6 to distributions are
>also listed as contributors on GitHub:

I have updated/massaged the documentation a bit and added the 2020
copyright to it to acknowledge the work Roland has done on getting the
getmail code to work with Python 3. As far as I know neither he or I
have aggressively pushed anything anywhere, but as users of getmail
and various distro's, we do see that getmail is being dropped left and
right for not working with Python 3. I understand that people can
still use getmail 5.14 quite easily if they are willing to install it
and Python 2.7 themselves, but there is a reason people like to use
distributions to get a coherent environment that can be kept up to
date using that distro's tools. I think this thread itself is enough
evidence for there being at least a demand/wish for a way to use
getmail with Python 3.

>And then there's this one, which contradicts statements made on this
>mailing list:
>
>https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=248633
>
>"getmail-5.14 depends on python-2.7, and there is no upstream effort to
> port to python3."

That is a statement by another getmail user like any of us are; While
it might not be factually correct, I think the sentiment is
understandable and can be forgiven, we just want to continue using
getmail with as little hassle as possible in our distributions of
choice. People have been credited in the THANKS file for "their
patches, bug reports, suggestions, packaging efforts, and
documentation updates" which seems only just to me.

>> I learned about this fork on this
>> mailing list first, and Charles has been invited to join the project,
>> so it can hardly come as a surprise that people are working on
>> this. Our fork/port is not a hostile one and I am sorry if anyone
>> perceives it as such.
>
>Inviting the original author to a fork is rather tactless.

I understood it at the time as a friendly reach out to keep Charles
(and others interested in the efforts) informed as to the current
status of the porting effort. I do not know if and what communication
there was prior to that.

>> getmail6.org was chosen and registered as our domain name as we'd like
>> our project to have its own website, I hope you take no exception to
>> that?
>
>If a project has been distributed for decades under http://pyropus.ca/
>and a fork goes straight for the jackpot URL, I think one is entitled
>to speculate about the motivations.

You are entitled to speculate as much as you want of course, but I
feel you are a bit quick to assume bad intentions or motivations over
the perhaps more benign explanation that we have an interest in
keeping getmail usable "out of the box" for as many of its users as
possible. getmail6.org is a logical name to pick for the website as we
can't use pyropus.ca for obvious reasons. If you find that to be in
poor taste, then I am sorry to hear that, but I think it might not be
easy to find a name that you wouldn't object to as it seems more that
you object to the existance of the fork at all than anything else.

>Perhaps the motivations are good, but I don't like forks of projects
>that are clearly maintained.  And I certainly don't approve of copying
>the documentation verbatim, even if the license allows for that.

Noted. But the license does allow for it, and I've gone to great
efforts to properly credit all the work done on the code and the
documentation as well as to convert the full history of all released
versions into git so that each line of them can be properly traced and
credited. I am sorry you find that disagreeable.

And with that, I think I am done on this topic :-)

Kind regards,

Remco