[WikiEN-l] Interesting debate over reliable sources

[Andrew Gray]

On 10/07/06, Anthony <wikilegal at inbox.org> wrote:
> On 7/9/06, maru dubshinki <marudubshinki at gmail.com> wrote:
> > On 7/9/06, Anthony <wikilegal at inbox.org> wrote:
> > > On 7/9/06, Erik Moeller <eloquence at gmail.com> wrote:
> > > > And if you want to say "blog X said Y", then of course "blog X" is an
> > > > excellent source for that. The question in both cases is more one of
> > > > notability and relevance than one of reliability. What needs to stop
> > > > is the blind worshipping of printed paper.
> > > >
> > > One problem with citing "blog X" when saying that "blog X said Y" is
> > > that the blog might very well not exist in a few years. A copy of a
> > > New York Times story, on the other hand, will certainly exist for many
> > > years.
> > >
> > And it is *exactly*  transience problems like this that show that we
> > need to be working more closely with the Internet Archive people (more
> > closely that is than simply letting ourselves be spidered by 'em and
> > letting'em d/l database dumps); I've suggested several times that when
> > a valid ext. link is submitted, it be discreetly tagged with an IA
> > link (assuming it exists), somewhat like how we're *supposed* to be
> > validating ISBN links
> > (http://bugzilla.wikimedia.org/show_bug.cgi?id=2391; I really wonder
> > about this bug- it should be pretty simple and useful, but you know
> > the devs... got other fish to fry). Or at least a robot to go around
> > adding links (http://sourceforge.net/tracker/index.php?func=detail&aid=1500288&group_id=93107&atid=603141)
> > if the original is gone.
> >
>
> I agree that this would be useful, to the extent it's possible ("there
> is a 6-12 month lag between the date a site is crawled and the date it
> appears in the Wayback Machine").  Also, since IA can and will remove
> pages from the archive under certain circumstances (one of which is
> when a robots.txt file appears later), you still have to worry about
> an archive disappearing later.
>
> And of course this doesn't at all address the issue of original
> research - rather than proving something by referencing original
> sources Wikipedia generally (if not exclusively) should be referencing
> facts which have already been analysed by experts in that field.
>
> Referencing in Wikipedia is currently a big mess.  The incremental
> tweaks seem to be making things worse in many ways.
>
> Anthony
> _______________________________________________
> WikiEN-l mailing list
> WikiEN-l at Wikipedia.org
> To unsubscribe from this mailing list, visit:
> http://mail.wikipedia.org/mailman/listinfo/wikien-l
>


-- 
- Andrew Gray
  andrew.gray at dunelm.org.uk