On Sun, 07 Mar 2010 17:55:20 +0000, palladium wrote:
> Jim Henderson wrote:
>> say “you know, a minority of the community uses NNTP, why don’t we just
>> get rid of it and force those users to use the web interface like
>> everyone else?”
> that too would solve the problem brought up in this thread…
Sure, but at the same time, it would likely alienate those who do use
NNTP - some of whom (such as yourself) contribute quite a lot. So it’s a
> as far as i can tell, the only real reason nntp was kept was in the hope
> that the devs would embrace the entire community (rather than stick to
> themselves)…an idea that for most part has not taken hold…
Well, some devs perhaps lurk - there’s no real way of knowing. But it’s
also here because the Novell forums (from which one of the communities in
the merge) have it, and it was functionality that was deemed that was
wanted. Again, offering choices because of an understanding that people
work in different ways.
> you know, i didn’t even know there was an http forum until one day i had
> an OS problem, went (back) to the newsgroup and saw a note there that
> the group was moving to nntp//:opensuse.org.* so i did also…and
> suddenly there were about a million broken links and weirdness and . . .
> i was told there was nothing that could be done because of the http
> sides requirements…
> yet, the broken links are almost gone now…
> maybe this too will be fixed…eventually, even if it is impossible
> to revisit the send-then-edit decision now…
Perhaps. I’m not saying that it would be impossible to fix the
dissimilarity, but that it would require custom coding, and that
introduces additional constraints on the system - such as maintaining the
custom code, updating it, and additional QA testing when a version update
comes along for vBulletin. All part of the balancing act.
As is ensuring that functionality that’s in place today doesn’t
inadvertently disappear when an upgrade takes place - which is one of the
bigger points in favour of not heavily customising the code but saying
that where we are today is “good enough”.
It took me years to figure out that aiming for “perfect” is a losing
person’s game - because you never get to “perfect”, but if you wait for
it, then you never make your code available. There comes a point where
you have to say “this is good enough” and put your system in place.
That doesn’t mean it can’t be improved, but it does mean that when you
make improvements, you have to be much more meticulous and aware not just
of how the changes affect the system, but also how the changes affect the
users of the system.
openSUSE Forums Administrator