Re: Nature: A call for a public gene Wiki

Phillip Lord wrote:
> 
> Matthew Cockerill wrote:
>> I agree that the details of gene function probably don't belong on
>> Wikipedia.
>>
>> <snip>
>> 
>> It's at least conceivable that Wikipedia may play an important role
>> in providing widely accepted identifiers for such high level  classes
>> and instances, since the high level of usage of wikipedia would tend
>> to keep those high level concepts  far better maintained and curated
>> than they would be in the backwaters of a specifically biomedical
>> ontology.
>> 
> 
> The problem with wikipedia in this context is that it does not
> have any specific policy on data preservation. Links can break, or
> refer to other things than they started out.
> 
> The general wiki community notion is a good thing, in general,
> but there are limitations. Wikipedia has succeeded well in getting
> to the size that it has, been it has it's limitiations. The here
> today, gone tomorrow, nature of the web is certainly one of them.

To some extent, these problems are self-correcting as more scientists
participate in the construction of the resource. Certainly there will always
be a lack of control for links to resources outside of the wiki, but links
within the wiki should be more reliable.

To me, the main point of the resource is getting more scientists involved as
contributors of knowledge, making this a community effort rather than
something constructed by one group. The amount of knowledge to be captured
is truly daunting and will require worldwide participation, as suggested by
the title of the original letter:

Gene-function wiki would let biologists pool worldwide resources
http://www.nature.com/nature/journal/v439/n7076/full/439534a.html

The trick is to figure out how to channel the energies of these worldwide
contributors so that the information they contribute is semantic web
friendly (e.g., ontologically controlled) without stifling participation.

Perhaps there could be different sections of the wiki, one that can be added
to and modified in a minimally controlled way, and another more
ontologically controlled section, which could be derived from the raw
section by designated curators. This permits different types of
participants: scientists who just want to add knowledge without worrying
about ontologies, and scientists that are willing and able to act as
curators.

At the end of the day, we'd like to have more than just a set of informative
and browsable web pages. We'd like to be able to build tools that can
intelligently mine the collected information (preaching to the choir here,
of course).

Steve

Received on Friday, 24 February 2006 00:47:33 UTC