Richard writes:
> Thanks for the thoughtful reply, Steve...
Thinking is what I do ("I was born this way"). But if others find it
useful, that's very happy-making. :-)
> Honestly it's been a while, so I'm not 100% sure memory serves but
> I think on my distribution they were pushing the dockerized version
> and I was pressed for time and didn't realize that was just my
> distribution's default package and not the preferred way, which is
> what you seem to be saying.
If it was a distro version, then I think they're making a mistake. We
are not fans of distros' Mailman packages in any case, though. That's
a different rant.
> I'm glad you responded, Steve because I appreciate the potential
> benefits. It looks like I have to give this another look-see,
> probably over this coming holiday season...
Feel free to consult. For best results use
[email protected]. There are a lot of people besides Mark
and myself with expertise there, some of which goes beyond ours.
> In my view Python and a good handful of other systems in (too-)
> wide-spread use were foisted
I could go into a long interleaved rant about this, but my opinion is
that nothing was foisted, because very few users are competent at
writing requirements. They get what they explicitly ask for. And
those who do know what they want are not willing to pay the full cost
of implementing those requirements.
> I KNOW it's like trying to hold back the ocean with a broom but I
> beg ALL developers everywhere to learn about and consider the
> down-stream effects,
The people I work with know about them. That's why Mailman advocates
the virtualenv install, and why my employer is a pretty big fan of VMs
and containers. I know it grinds your gears, but these are good-
enough solutions to those problems as experienced by our users, and
they are cheap vs. the alternative of the complete rewrite of whole
ecosystems.
Employer is also trying to move our bigger clients to a "business as
usual" model (aka managed services) for a (mostly) fixed price -- and
that puts most "apres install" costs in *our* lap, providing much
better incentives. (It also gives us a hold on existing clients, a
hold which is only partially mitigated by concentrating on open source
solutions where requirements allow.) Those contracts are devilishly
hard to write, I'm told.
> there ARE other ways to write your code that aren't unfriendly to
> those who would like to host it!
I don't know of any, myself, because none of them solve the "nobody
demands it enough to pay for it" problem.
> But I know I can be just some opinionated asshat, who just happens
> to have seen a lot and not been abjectly stupid.
Sure, but those often make good reads!
> > Have you read Jamie Zawinski's screed about the advantage of
> > Netscape 2's flat-file approach over formal databases for mail
> > reader message summaries?
> DO send me a link if you have one!
https://www.jwz.org/doc/mailsum.html
And there are followups!
https://www.jwz.org/blog/2004/03/when-the-database-worms-eat-into-your-brain/
https://www.jwz.org/blog/2011/07/mork-keeps-on-giving-when-the-database-worms-eat-into-your-murder-trial/
https://www.jwz.org/blog/2011/07/kids-today/
Aaand I have story about the "Kids today" episode. Richard Mlynarik
wrote me personally (not the -devel list where bugs were supposed to
be reported, which is presumably where he got my address) to complain
(in very similar terms except more four-letter words) about a really
nasty network stream corruption bug in XEmacs.
Or it would have been nasty, except that he supplied a small client-
server implementation that very quickly reproduced the bug. Now,
Mly's description suggested it was intermittent and state-dependent.
Which is why it was terrifying to me. At the time I hadn't discovered
Richard Stevens' book on Unix networking. Turned out to be an off-by-
one error every 64k bytes, and wouldn't you know it, that was the
buffer size. ;-) I had a tested fix in CVS in 90 minutes. Would have
been faster but I was terrified of the networking code. :-)
Never got another bug report from that source again. :-(
> Thanks again for the reply, Steve,
You're welcome!
Steve
--
GNU Mailman consultant (installation, migration, customization)
Sirius Open Source https://www.siriusopensource.com/
Software systems consulting in Europe, North America, and Japan
------------------------------------------------------
Mailman-Users mailing list -- [email protected]
To unsubscribe send an email to [email protected]
https://mail.python.org/mailman3/lists/mailman-users.python.org/
Mailman FAQ: http://wiki.list.org/x/AgA3
Security Policy: http://wiki.list.org/x/QIA9
Searchable Archives: https://www.mail-archive.com/[email protected]/
https://mail.python.org/archives/list/[email protected]/
Member address: [email protected]
