[WikiEN-l] a few new wrinkles in the external link issue

[Thomas Dalton]

> 1) If we're going to blindly attach "nofollow" to all the external
> links, why are we allowing Wikia links to be propped up artificially?
> Are we in the business of conflict of interest now?

Looks like Wikia is treated as an interwiki link, which assumes it's
part of the same project. I'd report it as a bug, if I were you.

> 2) Myspace blogs were recently added to the spam blacklist by Raul per
> request of Jimbo, although no one else seems to know why, how, or per
> what rationale.  I won't pretend to know what Jimbo's been up to past
> not having edited Wikipedia since the is-it-or-is-it-not-a decree, but
> perhaps some more explanation on this would be worthwhile?  Seems like
> we're blocking a shitload of otherwise worthwhile primary source
> material for many of our articles for the sake of...well...nothing.
> Meanwhile, a blog ''not'' hosted on MySpace is still a-okay, which is
> patently absurd on its face.  I'm wondering what the thought process was
> on this, since no one else seems to want to chime in.

MySpace blogs are never a reliable source for anything other than a
MySpace blog, and they are virtually never notable, so I don't see a
problem.

> 3) Did you folks know we have a bot that reverts links that are
> arbitrarily considered spam?  I didn't until today.
> [[User:Shadowbot1]].  I convinced him to post the blacklist where we
> could see it, and while some (most?) are useful, others are pretty
> screwy, and I'm not sure this is helpful in the long run.

I knew, but I've never looked at the blacklist. I've just trusted
it... was I wrong to?

> I'm starting to think that our focus on spam is becoming a problem
> rather than a benefit to the project.  How much collateral damage are we
> willing to accept in the project to take care of this "problem" that
> people think is massive?  One out of every 10?  5% poor hits?  Do we
> have some sort of measurement we're using here?

I don't think there is much if any collateral damage from our
anti-spam efforts. I can remember one problem with ShadowBot removing
a link that it shouldn't have, and I think it was resolved by fixing
the blacklist (I don't remember exactly, I wasn't actually involved to
a great extent).