Updated 'table of biological wikis'!
Dan Bolser
http://www.bioinformatics.org/wiki/BioWiki
I carried over the information from the old table (now depreciated)
and filed in a lot of new data. However, the information for many
BioWikis is still quite sparse. Please feel free to fill in the blanks
or correct the data for your wikis!
The biggest improvement over the old table is the inclusion of results
generated by an automatic 'wiki activity monitor script'. The script
queries the (MediaWiki) API for each site to collect all the changes
in the previous month. It then summarises the changes and uploads the
results. You can see a prototype script (without the upload component)
here: http://www.bioinformatics.org/wiki/Talk:BioWiki#Wiki_activity_script
To all wiki maintainers... can you try to make sure that your
MediaWiki API is on, and supports the rcprop=loginfo parameter?
I'll put the full version of the script on GitHub and start running a
regular cron job to keep the activity statistics up to date. If you
add a wiki to the table, be sure to provide the full URL to the
MediaWiki API to be included in the update!
One final request... I've tried to incorporate suggestions and
recommendations that came up during the NETTAB conference, but any
feedback on (or direct improvement of) the site would be very welcome.
In particular, the BioWiki entry pages themselves are still crude and
ugly. If anyone wants to play with improving the presentation of the
data on the entry pages or on the hub page that would be great ;-)
Cheers,
Dan.
Benjamin Good
I've added the Gene Wiki to the list. Since it lives inside Wikipedia, I
listed wikipedia's API url for it.. I worry this could cause your script
some problems if you simply suck in all the listed API urls and let it
rip. (To access just the gene wiki data you need to filter things down by
limiting to pages that include the ProteinBoxBot template.)
-Ben
>--
>BioWiki mailing list:
>biow...@googlegroups.com
>
>Subscription & archives:
>http://groups.google.com/group/biowiki-l
>
>Unsubscribe:
>biowiki-l+...@googlegroups.com
Dan Bolser
> Dan,
>
> I've added the Gene Wiki to the list. Since it lives inside Wikipedia, I
> listed wikipedia's API url for it.. I worry this could cause your script
> some problems if you simply suck in all the listed API urls and let it
> rip. (To access just the gene wiki data you need to filter things down by
> limiting to pages that include the ProteinBoxBot template.)
Hi Ben, just getting back up to speed with this. Thanks for your contribution!
Indeed, that setting would have caused the script to try to extract
all edits in Wikipedia in the last month. Fortunately(??), the cron
job hasn't been running (I've just been reviewing the code, and it
should be back up soon). However, this case needs special
consideration, because the API doesn't allow this kind of filter [1]
to be combined with an rc list query [2]. The script could be modified
to first query a page list (if such a filter is set in the form), then
query the rc list for each resulting page...
I'll work on fixing the basics first ;-)
Dan.
[1] http://www.mediawiki.org/wiki/API:Embeddedin
[2] http://www.mediawiki.org/wiki/API:Recentchanges