From: Brenda Chawner <Brenda.Chawner@vu...> - 2005-04-15 18:47:53
I think a wiki would be a great idea for this, as long as enough people =
are willing to help maintain it.=20
WikiEngines are starting to develop a range of techniques for dealing =
with WikiSpam. I'm most familiar with PmWiki, which supports blocking =
specific IP addresses or text, and the most recent version will also =
block posts that contain more than a specified number of new links (3 I =
I put a password on the LIANZA ITSIG wiki for several months because I =
was tired of removing spam, but since I took it off 6 weeks ago the spam =
has been manageable.
As a foreign librarian trying to keep updated on the OSS4LIB front, I second Ed's opinion. The more people gets involved with this the better it will be.
But I also think that one has to be catious with the solutions adopted, so as not to kill a good idea with precipitated eagerness. I think the best solution is one with a variable degree of control, matching the local spam intensity.
----- Original Message -----
From: "Ed Summers" <ehs@...>
Sent: Friday, April 15, 2005 3:57 PM
Subject: Re: [oss4lib-discuss] one long list of project updates, one big apology
> On Fri, Apr 15, 2005 at 08:14:43AM -0400, Daniel Chudnov wrote:
>> What do others think?
> wiki +1
> I've been impressed at the level of participation on this list for the
> past few years and I think that getting a wiki up and going for
> opensource software in libraries would be really useful. I mean it's
> not like we're trying to create a general encyclopedia or anything, this
> is a pretty small niche, and alot of the people who are in the niche are
> on this list it seems.
> The wikispam thing is definitely a problem, but wikipedia and the like seem to
> manage ok. I guess we'd just have to investigate what the best solution
> is. If it eventually means locking edits down to known editors that would be
> one way. Enabling some sort of plugin would be another. I would recommend
> erring on the side of being too lenient until something goes wrong.
> And at the very least DelayedIndexing  would be key:
>  http://c2.com/cgi/wiki?DelayedIndexing