Please do not use obnoxious spam filtering services

Tom Plunket gamedev at fancy.org
Sun Feb 17 01:05:36 CST 2008


> I attempted to correspond with a recent subscriber over a question
> he had, and was gifted with a bounce from one of those "answer this
> email and I'll grace you with the pleasure of reading your original
> message" services.
>
> That sort of setup is inconsiderate and rude, doubly so if you're
> going to subscribe to a public mailing list and ask people to spend
> time answering your questions.

Asked on mailing list, reply on mailing list?  That's kinda the rule
I always used unless I was going considerably non-topical.

Since it was me, I apologize to all for the list-noise.  I get many
hundreds of spam messages a day and whitelisting is the most
effective means of handling that.  I have whitelisted the mailing
list's address, so messages to the list will never receive the
autoresponse, and in the years I've been using it this is the first
time I've ever seen a complaint about it.  Plus, I have no idea what
the list of people who might possibly respond to my email is, so I
can't just add everyone.  ...of course, this is the first list I've
subscribed to in a long time where the default reply-to was not the
list itself.  If you get the autoresponder message from me have no
fear, you'll never get it again and you'll probably get whitelisted
anyway when I actually do sort through the junk (once a week or so).

Bryan, thanks for the hgbook.  It's a good read.

...and thanks for the email on binary assets; you've identified the
biggest potential problem (they don't diff, you get database bloat),
but that's a problem in centralized VCS as well (although
workstations don't tend to have terabytes of disk).  I would think a
DVCS could handle this "problem" in a number of interesting ways,
but I have yet to browse the hg sources.  One semi-fix would be not
to pull the entire set of incremental changes but just the one end
change, but I'm not up on the hg vernacular (or usage, heh) to know
if the mechanism to do this exists (although I assume hg update
without --pull is just this functionality).


-tom!


More information about the Mercurial mailing list