[Taxacom] The strain between Wikipedia and Science

dipteryx at freeler.nl dipteryx at freeler.nl
Sun Mar 6 02:37:15 CST 2011


Van: Jim Croft [mailto:jim.croft at gmail.com]
Verzonden: zo 6-3-2011 1:12

> The main issue for modern information management is not that 
> Wikipedia compilation can be mindless (and I think you are 
> being a little uncharitable - there is often quite a bit of 
> editorial judgement that goes on), as in this example but
> that it is manual.

***
I don't see that there is any editorial judgement in such cases
except in the selection of a picture. Very many Wikipedia entries
are not manual at all, but generated automatically and they are 
very similar to this one: some of these are rather silly 
(it depends to a great extent on the database that is harvested).

Wikipedia's reason for being is that it is manual, and for some 
topics this works very well indeed.  
* * *

> There appears to be no way to build in dynamic content that 
> is generated and maintained in other databases and generally 
> play the linked data thing and see an information element 
> in context. It is essentially a bunch of static hard coded 
> text hard linked to another bunch of static hard coded text. 
> Its apparent responsiveness is little more than a large and 
> active bunch of humans responding.

***
????????????

There is plenty in Wikipedia that is generated and updated 
automatically from other databases. Perhaps ever more so. There 
was once a Wikipedia in some language or other (perhaps still is)
that was generated automatically in its entirety (not manual at
all). And of course there is a project that uses the Wikipedias
as raw material and generates content automatically from that. 
The weakness of Wikipedia (at least for many of its pages) is 
the lack of context and the absense of a bunch of humans that 
respond to changes in the world out there.
* * *

> Wikipedia would be much more exciting (useful?) if it had, say, 
> a taxonomy block (from the Catalogue of Life, for example), 

***
That is being experimented with (boring), but fortunately not
based on the Catalogue of Life (horror!).
* * *

> or an embedded map (from GBIF, for example) that would be 
> updated automatically as the databases behind them are changed. 
> Don't get me wrong - I love the Wikipedia and the whole notion 
> of free and distributed knowledge.  I just wish it had more 
> dynamic and real time content, as is, warts and all, independent
> of Wikipedia editorship. Under this model Wikipedia editors 
> argue over the suitability of the link, and once agreed, what 
> is delivered is up to the link - no [citation needed] needed; 
> the link is the citation. But that is probably unlikely to happen.

***
As pointed out above, much of Wikipedia is like that (and there
are other such projects), but it pretty much stops being Wikipedia
in such cases. One of the big problems of Wikipedia is those
'contributors' who treat themselves as part of such an automatic
machine, copy-and-pasting 'data', combining the weaknesses of both 
approaches.

Paul



More information about the Taxacom mailing list