This post is entirely off-topic, though it ends with some stuff meta
to this group.
On Wednesday, June 5, 2013 1:50:28 PM UTC-7, Peter J Ross wrote:
> In humanities.classics on 03 Jun 2013 22:04:26 -0400, Rich Alderson
> wrote:
[but this is Peter Ross first]
> You're using Gnus, which is perfectly capable of understanding
> MIME-encoded Greek characters. So are slrn, tin and nn. So are all
> other popular newsreaders that I know of - even Outlook Express and
> Google Groups - with the exceptions of XNews and trn, both of which
> were abandoned by their developers many years ago.
Apparently; I knew trn4 had been neglected for some time, but was
taken aback not even to find it listed in Wikipedia's lists of
newsreaders.
> The principal reason I put Greek verses into my .signature is to show
> users of obsolete software that they need to upgrade.
Problem is, for me trn4 *is* an upgrade. The only software I've used
that comes close in functionality for me is MacSOUP, which is even
more outdated, and only comes close. (Can't begin to describe how
much I'm looking forward to *header control*, for example, which
MacSOUP doesn't offer.)
I assume Gnus can do anything trn4 can do, but I doubt I have enough
time left in my life to master anything involving Emacs, so that's
no help; and I've tried Gnus once, so I know I'd have to master it,
and not just use it, to get trn4's abilities from it.
> > Newsgroup creation was thankless then, and I would imagine more so
> > now, if it's even possible.
> You can get a Big-8 newsgroup "created", in the sense of "added to an
> official list" more easily than in the past.
Hmmm.
Is interest in this form of "creation", then, as low as would appear?
I was involved in the creation of two groups (soc.history.ancient and,
as principal proponent, soc.history.early-modern), and worked on a
bunch more, starting in 1995. Up to 1998, when she-m came in, it
wasn't that hard to create a group if you had the necessary clues and
didn't happen to pick a topic too close to any of the main permanent
floating flamewars. The necessary clues could be hard to get (I won
fame on news.groups by writing an FAQ about them), and tough luck if
you wanted to create a group about, say, Pakistan or Armenia, but for
the most part, it wasn't that hard. (This was a change that kicked
in during maybe 1992 or 1993 for comp.* groups, and 1994 for the rest.
Prior to that, newsgroup creation had been harder, varying in
difficulty, since about 1983.)
From 1998 until the system definitively broke down in 2005, it got
harder and harder because votes became harder to find, until
finally *in* 2005, not a single group passed its vote. I was one
of the committee charged with fixing this, and wound up in what I
called the "loyal opposition" when the committee decided to become
the new operators of the Big 8; when the "term" for me that this
resulted in ended, I left newsgroup creation.
So let me see what happened, both before then when I was ignoring
that kind of discussion in my inbox, and after, when it wasn't
there. All this comes from the archives of news.announce.newgroups
at <
http://www.big-8.org/> (the hyphen matters):
2006 - 10 creations. Huh. I had no idea.
2007 - 21 creations (16 of them, however, essentially without
proponents, as a sort of board experiment, after my "term")
2008 - 3 creations
2009 - 0 creations
2010 - 2 creations
2011 - 1 creation
2012 - 1 creation
2013 so far - 3 creations
Creations have been swamped until this year by a *long* series of
group removals, totalling easily ten times as many as the above
and possibly more. I don't know how this played out in
news.groups.proposals, a group I've never read, but it can't
have been *that* helpful. (I noticed one of the 16 groups
created without a proponent being removed; didn't look for the
others, but my guess would be that the experiment more or less
worked. I wonder if they'll repeat it.)
> But server admins pay
> less attention to the "official lists" than they used to. Big-8 is a
> lot more like alt.* than it used to be.
Official lists mattered at first because the "backbone" admins
agreed to adhere to them, back when Usenet traveled mainly by
UUnet feeds. Then they mattered because there were a whole lot
of conscious admins who cherished their right to run their own
server, but to varying degrees exercised that right carefully
around the Big 8. (Abby Franquemont-Guillory was widely respected
on news.* as news-admin of Tezcat, a Chicago ISP of those days and
my first ISP; but after she left, I documented a ton of cases where
Tezcat differed from the official lists, many dating to her tenure,
and often reflecting a clear intentional policy.) Finally they
mattered because they fed the list at
isc.org, and the transit
servers used those.
Propagation to actual *user* sites, which the transit servers by
definition were not, was probably never all that automatic. We
had a lot of trouble with it when sci.classics was renamed to
humanities.classics (not assisted by the then-czar of the Big 8
screwing up the renaming [1]). It didn't come fast to she-m, and
was, I think, one factor (though not the main one) in the group's
failure to thrive. This is in 1998 now.
I didn't get much by way of clues later than that, although
propagation should've been *helped* when it became possible soon
after 2000 to issue proper Big 8 control message checkgroups
again. So I don't have any way to judge whether things have
really deteriorated since 1998.
Joe Bernstein
[1] David Lawrence, aka tale, ran news.announce.newgroups from I
think 1991 to, oh, 2003? If anyone cares I'll pin these dates
down, but although I actually have researched this topic's
history extensively, I don't have my notes here and have limited
multi-tasking where I'm writing from. Anyway, tale cared a lot
about automating tasks, which required carefully formatted
documents. We found out the hard way that he had never sorted
out with the Usenet Volunteer Votetakers a format for a one-off
renaming's Result posting, and it turned out that he used
carefully formatted Result postings to trigger scheduled events
in renamings. So those of us who'd worked in support of the
renaming spent *months* cajoling people and bullying ISPs to get
them to carry the renamed group because sci.classics was going
to go away. And then it didn't go away, and didn't go away,
and didn't go away ... And *then*, when tale found out his
mistake, he removed it instantly, with no chance to let people
know "OK, now it's for real". Basically, *.classics was treated
about as badly as I ever saw that system treat any actually
created group.