Let’s try the "Wiki database" idea.

In the comments to my previous post, Nick Matzke* suggested the idea that one could create an open database on the Wikipedia model in which people could compile and contribute published data and access the complete dataset freely. This would not replace other databases that give a pile of information for each species (e.g., FishBase), but would deliver one parameter at a time for a number of species. It would be good if the datasets could be exported to a spreadsheet for analysis as well.

Having been through the extremely time-consuming task of building a database from the bottom up (and, thankfully, having a pro web designer build it, which took a massive effort on his part as well), I would be all for trying a community-assembled database for other kinds of information. Kehan has suggested that something like the scratchpad used by the European Distributed Institute of Taxonomy (EDIT) would work nicely for this. I think that looks very promising.

So, I am totally up for trying this experiment. I should tell you that WikiData.org, WikiBase.org, OpenData.org, and a bunch of the obvious domains are unavailable. Anyone got a good name? Maybe a biological term that reflects a bunch of individuals contributing to an emergent database? (Please don’t say “EmergentDatabase.org” — try to be more creative). I’ll register it, then hopefully we can recruit someone with web experience to help set it up. I have data on red blood cell sizes that we could start with that is a reasonably complete dataset, and of course chromosome numbers could also instigated and then contributed to by users.

No point letting a great idea pass by!


* Incidentally, Nick, when are you going to get your own blog?

8 thoughts on “Let’s try the "Wiki database" idea.

  1. This is a really cool idea, and something I was getting ready to blog about myself. David Lancashire has an open Chinese-English translation database (Adsotrans) that has been really successful over the years. By fetching Chinese news articles, and adding mouse-over annotation using the open database back-end, he has also created a very interesting way to allow users to interact with the database. It becomes very easy to see what words are lacking a good translation, or very easy for someone who knows the translation to contribute to the project.

    I think this could have a lot of potential for biological datasets as well. One idea that might be helpful to incorporate would be some kind of rating system for database entries. For example, if you want to have a species specific page which includes a brief “introduction,” rather than allowing only a single entry for the introduction at any one time, allow multiple entries. Users could then “vote” on the one they feel is the best, and it would get a higher weight accordingly.

    All in all it sounds like a great project. Keep us posted on how it goes.


  2. This is a great idea, Ryan. I think NSF actually has funding to support such initiatives, in the otherwise little known DBI section. Check it: http://tinyurl.com/ysvaw3

    P.S. wikibase.org, datawiki.org and database.org are taken.

  3. Yeah — I have been waiting for someone to have this idea but I haven’t seen it yet…glad people are intrigued.

    Scratchpad might work:

    …But in an ideal world, NCBI or someone would set something like this up for biologists. Really, all biological datasets and databases everywhere should just be hosted/mirrored on NCBI or something like that. Not just every new sequence, but every new specimen, fossil, species, etc., would get entered into the database. Journals would make entering this data into the databases a requirement for publication.

    All this will happen When I Become King. While I’m dreaming, BTW, I’d like a pony.

  4. Hi John,

    The NSF BD&I looks promising. One catch:

    “International collaboration is encouraged; however, financial support for any non-U.S. participant organization must be provided from within the participant’s country or other non-NSF sources.”

    I may be able to conjure up some funds to get started, but my hope is that this will be pretty inexpensive to get off the ground if we use/exapt existing interfaces.

  5. Talk to some astrophysicists. They do amazing things with their data, and it sounds like some of the same principles would apply here.

  6. You don’t need to have web experience to set up a wiki. Try http://www.wikispot.org – I have two wikis there and it’d be welcome in that community. You can register whatever web address you want for it and they’ll help you point it to the domain.

Comments are closed.