h1

Who’s working on sharing also-liked data?

2007-04-16

Code4lib folks: I’ve got some questions that I’d like us to discuss. Please read on.

It sounds like a few sites are starting to collect user ratings (or just bare, anonymized circulation records) for their items so that they can be used in a “People who liked this item also liked…” suggestion.

This is a very good step, but there are advantages to the ability to use equivalent data from other libraries.

Some libraries will have use patterns that would be unhelpful. For example, just because the technical college’s library has a small popular-fiction section doesn’t mean that most people who liked the latest bestseller will also be fascinated by Chilton auto-repair manuals.

Other libraries’ data might be very helpful. For example, a small public library might be unable to generate a useful amount of also-liked information on its own, but would benefit from using similar information from a larger public library nearby, or from small public libraries across the country.

If we, as a community, haven’t already come up with answers to these questions, we ought to begin:

  • What kinds of information do we want to share? Are there any kinds of information that we know we don’t want to share, regardless of whether a user might actually want us to do so?
    I, for one, would consider attaching my demographic profile and perhaps my identity to a list of books I’ve checked out or enjoyed. Let’s preserve my right to share information as well as my right to keep it private.
  • How can we represent that information in a standard way that allows for optional components (like Dublin Core)?
  • Where should this data live?
    I have a suggestion here: each library’s data should live on one of its servers, and there should be a more centralized repository of information about what these libraries are sharing. Updates pulled through RSS could keep my library’s server up-to-date on what your users think.
  • How can we find only the information we want? (For example, you might want to retrieve a full data set with demographic information for each user. Or maybe you just want to retrieve the average overall rating for each item.)

Any thoughts, folks?

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: