Last Tuesday I gave a talk in Philly at the 50th anniversary NFAIS (National Federation of Abstracting and Information Services) conference. The audience consisted of some librarians and some people from the database industry. I took the opportunity to say a bit about the read/write web, what libraries have done with it, and how databases would be better if they took cues from it.

One thing I presented was a mockup of what a social or participatory database might look like. My slides might not be too helpful without commentary so I’ll just write a bit here. Here’s what I’d like databases to have:

→ a usable interface. Simple enough it seems, yet we direct patrons to interfaces that have 6 or more places to input data and all sorts of options to refine their searches. This is not acceptable. The example that I gave of a great interface, maybe my favorite in all the library world, is VuFind. It is an open source OPAC overlay being developed by Villanova University. I can even paste the entire link(!) to a book’s record right here because it isn’t long and ugly! http://www.vufind.org/demo/Record/397263. Here’s a bit more on VuFind from Roy Tennant in Library Journal.

→ usable urls. See above. To put a finer point on it,

http://web.ebscohost.com/ehost/detail?vid=1&hid=113&sid=85d806a6-91d6-40a8-b53b
-dac0e9002bf8%40sessionmgr102

is plain unnecessary. Why not something like http://database.com/authorID/article#?

→ history of articles retrieved. Not just saved articles in folders. The ability to turn on a feature to log all articles accessed could be useful. If the system could then generate statistics about a user’s viewing history, that would be great. Think: most accessed journals, authors, etc…

→ sharable, persistent spaces. Here I’m envisioning something like David Lanke’s Scapes concept which I find compelling. People can assemble content from databases, discussion, notes, links, a todo list, and other relevant documents like audio, video or pictures. These spaces have granular privacy and collaboration controls so that they can be shared with the world, certain networks, specific groups, all research friends, or only certain friends. Just like Yelp features their power users, so too could outstanding examples of research be highlighted by the database company. This would engage people and show off the full potential of their product.

→ embraces the open web. Library research does not exist in a library only vacuum. Research tools should reflect this and not segment themselves from the rest of the web. I don’t expect results from closed databases to appear in the open web, but it would be great to include stuff from the normal web, the web that people are accustomed to using, into databases.

→ full RSS support. Some databases (ebsco and proquest only?) are starting to get this. Customizable feeds for queries, authors, periodical titles, keywords, subjects, most saved/emailed/printed would be good. Providing javascript for RSS –> HTML would be great.

→ easy to use citation tools. In any relevant format.

→ user profiles. If something is going to have participation, there must be participants. If participant’s are meant to have meaningful identities over time, there must be user profiles. So here is what someone’s profile on a participatory database could look like, with details below.

Click through to flickr for a better view.

Things to note:

• Remember, this is someone’s homepage for a database or maybe all library databases. There’s one simple search box in the upper left.
• People can be favorited or friended. In this case the friending mechanism is a red heart by the username. Maybe a bit strong of a symbol for actual use.
• Networks link up people of common interests.
• On the left again there are links to things within FindBook and links to the open web. Customizable.
• Saved articles and shared articles. Easy access for the user, or others interested. There’s granular control over which networks, research friends, groups receive a shared article.
• Under “Networks” appears a class listing - Phil 367: Epistemology and Being. This is an example of a “persistent and sharable space.” Expanding the tab lists research friends associated with it, and clicking on the title opens it up.
• Research friends function just like friends in FaceBook. Here we have a link for easy access to articles research friends have shared.
• Groups could consist of project teammates or entire classes.
• Library information appears on the left. Here there’s information about books on reserve, eBooks and fines. Some library news and/or upcoming events would be nice too.
• Notes section is self explanatory. Maybe these should be sharable.
• And finally, there is a customizable section for specific search boxes, including open web sources.

Even though I didn’t intend for it to, this has some courseware functionality in it. I don’t know how much I’ll pursue this line of thought, but it does make sense in that a user-centered, practical research tool would be course based. A participatory database certainly wouldn’t replace anyone’s main social networking site, but it would look more like what people are accustomed to using. Because of the ability to make useful social connections people would be motivated to do research.

I hope you find my thought experiment worthwhile. A quick reminder that while light-years away from what databases currently look like, this stuff is library science, not rocket science. Isn’t like sending people to Neptune to establish a colony. Ideally I’d like to say that libraries could produce such a product but I don’t think database vendors provide the requisite APIs. So if we want something like this, we’ll have to collectively pressure them.