Information Desk Del.icio.us Adventures

Challenging questions often lead to the discovery of interesting web resources. Over the years our reference folk have assembled a collection of useful sites for answering the FAQs (and the not-so-F AQs) that we get at the desk. These are currently saved as browser bookmarks on the various information desk computers. There are several disadvantages to this system: consistency on all info desk computers is difficult to achieve (especially for those in other locations), and it is time consuming to find and weed out broken links. So, a few months ago we decided to try converting to del.icio.us, which has proven to be more challenging than we expected. Here's a chronicle of our (mis)adventures to date:

Step One: - Setting up the hplinfodesk del.icio.us account
We uploaded the bookmarks from our reference computers to a del.icio.us account, tagged them and bundled the tags (http://del.icio.us/hplinfodesk). Aside from a minor uploading mishap (it appears there's a limit on how many can be uploaded at once) this process went fairly smoothly. We also discovered "Fresh Delicious" which enabled us to identify outdated URLs with one swift mouse click. So far so good.

Step Two: - Initial Feedback
We asked a pilot group for feedback and discovered that our tag list/cloud was so enormous that they were overwhelmed. Two potential solutions were suggested: (i) to weed both posted sites and tags to make it more manageable, and (ii) to provide a keyword search function. Since we couldn't part with enough sites/tags to accomplish (i) we started working on (ii).

Step Three: - The Quest for a Keyword Search
1. The Del.icio.us Search Function:
Unfortunately in order to search our del.icio.us page you would have to be logged in to the hplinfodesk account. This was not an option. Controling tag proliferation is difficult enough with only one person adding sites. Giving access to more than a dozen people would be a sure way to court chaos. We experimented with del.icio.us "networks" and "subscriptions", but neither of these methods made it possible for someone with a different del.icio.us account (like mine) to search just the hplinfodesk bookmarks.
2. Mashups ... DeliGoo:
Our search for a suitable mashup led us to DeliGoo. While this is a great little application there were two reasons we decided not to adopt this solution: (i) it requires the installation of a firefox extension which would pose a significant hurdle for adoption system-wide, (ii) DeliGoo performs a Google search on all of the sites posted to our delicious account (i.e. it does not search just our hplinfodesk page). So, it's an overkill and in typical google fashion produces a large number of search results.
3. Google Custom Search Engine:
A Google Custom Search Engine would be a fabulous solution ... if we could get google to index our page! We submitted the URL to Google for indexing at the beginning of January, but I guess the Google spiders are too busy crawling a different part of the web. Google is more likely to index a site that is linked to from other indexed sites. So, that's why I've put so many links to http://del.icio.us/hplinfodesk in this blog post. I'm still hopeful!
4. APIs and a DIY Solution:
Given that our lower tech "serve it up to me on a platter" approach has not yet yielded a solution, we called Ryan today to find out if the del.icio.us API would accommodate developing a customized search solution. It's looking promising, and the race is on - Ryan v.s. the Googlebot! :)