Managing Google Groups headaches

[Rich Kulawiec]

(comments from a lurker on python-list)

- Google "groups" is a disaster.  It's extremely poorly-run, and is in
fact a disservice to Usenet -- which is alive and well, tyvm, and still used
by many of the most senior and experienced people on the Internet.  (While
some newsgroups are languishing and some have almost no traffic, others
are thriving.  As it should be.)  I could catalog the litany of egregious
mistakes that Google has made, but what's the point?  They're clearly
uninterested in fixing them.  Their only interest is in slapping the
"Google" label on Usenet -- which is far more important in the evolution
of the Internet than Google will ever be -- so that they can use it
as a marketing vehicle.  Worse, Google has completely failed to control
outbound abuse from Google groups, which is why many consider it a
best practice to simply drop all Usenet traffic originating there.

- That said, there is value in bidirectionally gatewaying mailing lists
with corresponding Usenet newsgroups.  Usenet's propagation properties often
make it the medium of choice for many people, particularly those in areas
with slow, expensive, erratic, etc. connectivity.  Conversely, delivery
of Usenet traffic via email is a better solution for others.  Software
like Mailman facilitates this fairly well, even given the impedance
mismatch between SMTP and NNTP.

- Mailing lists/Usenet newsgroups remain, as they've been for a very
long time, the solutions of choice for online discussions.  Yes, I'm
aware of web forums: I've used hundreds of them.  They suck.  They ALL
suck, they just all suck differently.  I could spend the next several
thousand lines explaining why, but instead I'll just abbreviate: they
don't handle threading, they don't let me use my editor of choice,
they don't let me build my own archive that I can search MY way including
when I'm offline, they are brittle and highly vulnerable to abuse
and security breaches, they encourage worst practices in writing
style (including top-posting and full-quoting), they translate poorly
to other formats, they are difficult to archive, they're even more
difficult to migrate (whereas Unix mbox format files from 30 years ago
are still perfectly usable today), they aren't standardized, they
aren't easily scalable, they're overly complex, they don't support
proper quoting, they don't support proper attribution, they can't
be easily forwarded, they...oh, it just goes on.   My point being that
there's a reason that the IETF and the W3C and NANOG and lots of other
groups that could use anything they want use mailing lists: they work.

- That said, they work *if configured properly*, which unfortunately
these days includes a hefty dose of anti-abuse controls.  This list
(for the most part) isn't particularly targeted, but it is occasionally
and in the spirit of trying to help out, I can assist with that. (I think
it's fair to say I have a little bit of email expertise.)  If any of
the list's owners are reading this and want help, please let me know.

- They also work well *if used properly*, which means that participants
should use proper email/news etiquette: line wrap, sane quoting style,
reasonable editing of followups, preservation of threads, all that stuff.
The more people do more of that, the smoother things work.  On the other
hand, if nobody does that, the result is impaired communication and
quite often, a chorus of "mailing lists suck" even though the problem
is not the mailing lists: it's the bad habits of the users on them.
(And of course changing mediums won't fix that.)

- To bring this back around to one of the starting points for this
discussion: I think the current setup is functioning well, even given
the sporadic stresses placed on it.  I think it would be best to invest
effort in maintaining/improving it as it stands (which is why I volunteered
to do so, see above) rather than migrating to something else.

---rsk