discuss: wiki spam
Subject:
Re: [discuss] wiki spam
From:
Rick Moen ####@####.####
Date:
31 May 2008 07:46:36 +0100
Message-Id: <20080531065340.GM1870@linuxmafia.com>
Quoting jdd ####@####.####
> So banning this IP and making the home page write accessable by the
> admin could fix the problem for now...
I'll give you one guess about how useful and worthwhile I predict that
fix will prove, based on experience elsewhere. (You're welcome to more
guesses, but really need only one. ;-> )
> Notice: most discussion about spam in the wiki world (enough to search
> for "spam" in the moinmoin doc) is mostly pointless, because paranoid.
For the more-obscure wikis / blogs: good point. But I imagine that a
thriving LDP wiki (much more than the current contents!) would, by
contrast, become more of a target.
> LDP wiki *may* be very much exposed, but if this happen it will be a
> good sign :-). Is most cases simple system like a publicly spread
> password have very good results (I use it an I'm never spammed).
> Robots *can* be very smart, but most spammers don't even worry doing so.
The problem has never been keeping spam load low; the problem is keeping
spam load low while not impairing legitimate use overmuch. (The very
idea of wikis being wide participation, you have an inherent conflict of
goals.) Several types of solution suffice to keep generic spambots out,
e.g., a site-specific question ("What's Einstein's surname?"), a widely
known but site-specific password, and so on. That leaves customised and
fully human-driven defacements, which seem, by and large, to go after
only prominent sites. Which gets back to my point (above) about
obscurity and targets.
