[sf-lug] Version Control :-) (Re: Who done it? ...)

[Rick Moen]

Quoting Michael Paoli (Michael.Paoli at cal.berkeley.edu):

> Well, the "who done it" was mostly just an illustrative exercise;
> e.g. if I/we wanted to find out who made some change to some file ...

Yeah, I figured.

> As to version control ... nothin' prevents SF-LUG folk(s) from implementing
> that on the web page - and even other web content there
> (at least all those that have edit access to it):

You betcha.

A really brief reminder about how I came to host a mailing list for
SF-LUG in the first place:  Back in 2004-5, SF-LUG had been running a
thriving mailing list for a while.  It was quite a happening place.
Then, suddenly, it was down.  And the sense of discouragement was
palpable.  

(It's been so long that I'll give myself licence for some looseness in
reconstructing this.)  I chanced to be talking to Jim, and there had, I
think(?) been some catastrophic hardware or software problem.  I said,
well, but the good news is that this was Mailman, right?  Yes, it was.
But, I said, that means, worst-case, you use your backups of the
cumulative mbox.  Like, y'know, it'd doubtless be
/var/lib/mailman/archives/private/sf-lug.mbox/sf-lug-mbox .  That's the
crown jewels, you know that, right?  (No answer.)  [Optimistically;}
'Armed with the mbox, or even a week-old backup copy, and _some_
reconstruction of the membership roster, which you can get or fake in
various ways, or best case you occasionally ran
'/usr/lib/mailman/bin/list_members -f sf-lug > somefilename' in order to
dump the membership roster to a simple ASCII file.  Did you do either or
both of those things?  The mbox?  The roster?'  (Pained silence.)

'Jim, but, if not that, what about system backups of some other sort?
Any other sort?'  (More pained silence.)  'Um, Jim, don't tell me a
production system had no backups of any kind?'  (big long awkward
moment)

So, wow, I didn't want to put more salt in that wound.  Dunno exactly
what nuked the system, but somehow nobody ever secured usable backup
data, so it was the worst possible loss scenario, where everything is
just completely gone and there's no way to recover anything.  I guess.
Because the awkard silence was a very powerful statement, and it
appeared that I couldn't help so I dropped the subject... and switched
tacks:  'Jim, I'm really sorry about whatever happened.  Would it help
if I offered SF-LUG a home on _my_ Mailman server?'  Yes, as it turned
out, it did -- and I was glad to alleviat the gloom and offer a hand-up
in time of trouble.

Now, you, me, if either of _us_ suffered a catastrophic system loss like
that, especially one that was up in lights with the public because a
community depended on it, we'd make double-damned-sure it never happened
again.  Ever, forever and a day, amen.  Mechuleh.

SF-LUG, though?

Fourteen years after an ignominious total-loss server failure, did
SF-LUG learn from being cluebatted by reality?  I'd love to be able to
say I think the answer is 'yes', but I'm an honest man, so I won't say
that.

(I'm talking about other than what you _yourself_ brought to the
picture.)

> I do use quite a bit of version control, ... e.g. lots of RCS (relatively
> dead simple (mostly) & local); also do and/or have used many others
> too (SCCS, CVS, Subversion, Git, some commercial stuff (e.g. ClearCase),
> and probably some others too that aren't jumping to mind).

And any, no matter halfassed, is better than none.  Just like backup.