But before I go any further, I need to disclose several tidbits about this whole endeavor.

First, this was a "black-ops" project. That is to say, I consulted no one before launching into this. I did not go to the public and ask their opinion. I did not go to my colleagues and solicit their input. I did not float this in committee. I didn't get any official authorization to do it. In fact, the whole thing flew under the radar from its conception through to fruition, which took exactly four days. In many ways, it was a spur-of-the-moment project that I did to keep busy during a few relatively quiet days. Sometimes, you need to throw caution to the wind and just do it.

Bear in mind that the only place this service was advertised was on this blog (where, admittedly, it was subsequently picked up by several others). I did not announce it to AADL staff--I let them discover it on their own. Many probably still don't know it's there. When you look at the numbers, know that they reflect those who either read about the cards on a blog or stumbled across it during their course of regular business.

It was a very fun little project, but the real value was that I could slip it into production very quietly and let it act as a proof-of-concept for some very real, very large changes I want to do to our OPAC (which I definitely need authorization for). Essentially, I wanted to answer a very simple question:

Is the public ready for a social OPAC?

What I found indicates something very special, indeed. In fact, as time progressed, I began double-checking my methods which consisted of a simple laundry list of basic queries, which I'll describe. As you can see, I only have 7 collection dates and they are not at regular intervals so the growth rates represented are a little deceiving, but as I've said before, this is very unscientific--I'm definitely no statistician! The results, however, are real.

The first graph here (above) is the simplest of all. It's the total number of marginalia comments in the system over time. The query was simply:

SELECT COUNT(*) FROM cc_marginalia

The count for each date, respectively is: 35, 107, 160, 413, 593, 634, 790. You can clearly see that the number or cards being marked up is experiencing a steady climb. I actually am not terribly surprised by this, but for an unadvertised service, it's still much higher than I had expected or hoped for.

Now, marking the cards up is one thing--it's a fire-and-forget process, it doesn't even require a user to be logged in. Adding a card to a collection is another. Adding a card requires that a user be logged in to a valid account and represents, what I consider, a higher participation factor. This means that users are not only adding marginalia, but they are taking advantage of the ability to build card catalog collections. To me, this is really exciting stuff because it tells me that a) the service is useful and b) there is a market for this kind of stuff in a library setting. FYI, the query for this data is:

SELECT COUNT(*) FROM cc_savedcards

The data for the preceding graph dates is 8, 43, 66, 82, 216, 320, 388.

If you're interested, you can take a look at my saved cards. In addition to knowing how many cards were being added to collections, I needed to know how many people were collecting cards. This is where I really started to feel my pulse quicken. These numbers told me that, despite a lack of advertisement, a substantial number of people were building card catalog collections. I have to be honest and say that I was expecting only 20 to 30 users participating after three weeks. As of 2/11, however, there were 75 users building collections! Check out the growth over time: 2, 6, 22, 42, 58, 62, 75 for each date respectively. Of course, to put that in perspective, we have 20,955 registered users on our Drupal site as of 2/14--our regular users are an extremely small fraction of that, however. To get this data, I ran the following query:

SELECT COUNT(DISTINCT(uid)) FROM cc_savedcards

So... How many of those users not only started to build card catalog collections, but opted to share their collections with the world? I was blown away by these results. Simply amazing: virtually every person to create a card catalog collection also opted-in to sharing it. To me, this speaks volumes about the types of services our users want and it tells me that a good number of our library users want to interact with each other--a vital ingredient if you want a social OPAC. It also sheds a little light on the sophistication of our users--a lot of them are already Web 2.0-aware and will be sensitive to new services like this. Our commercial counterparts are doing a great job of softening the market for us--now all we have to do is provide the tools on our systems.

There is a very good chance I'm overlooking something here--if there is a metric you'd like to see me add, please ask for it and if it's feasible, I'll do my best to provide it.

Our users are smart, clever, interesting, positive, intuitive, and social. They may not know it yet, but they're waiting for their public libraries to be a catalyst for the community. There is something wonderfully special and intimate about shared experience--that is why Web 2.0 is so successful. When those experiences are centered around books, movies, and music and they're aggregated at the local level, the product becomes highly personal--how inspiring would it be to have patrons who are proud of the job they've done for their library?

NOTE: The above graphs were created using the open source graph library, jpgraph.

[Update 4/16/2006 2:42PM]
Somehow, this post got flagged as "private" .. Odd. Very odd, but fixed now.
[/Update]

[tags]Web 2.0, Virtual Card Catalog, Library 2.0, Programming, Library, AADL[/tags]

First, this code is written for use with III's XMLOPAC and requires my PHP XMLOPAC class. That said, an astute PHP coder could easily port this app and make it usable on another system (I heartily encourage that). Someday, I plan to make it completely generic so that all you need to do is pass it the data itself. I'm a little busy at the moment, however.

I am not releasing the card stock images that AADL uses. There was no license info associated with the original images. Until I can contact the folks at UPenn and find out what the deal is, I'm not going to make them available. It'd be very easy to create your own stock, however.

Same deal with the fonts. I was able to freely download them--so should you.

At present, this code is really meant for developers, it is not a "package" that will work out of the box. You must know what you're doing in order to get this going.

I'm releasing it under GPL for all the world to use freely, forever and ever.

Included in the tarball is a README file with some basic instructions. Be sure to read that before emailing me your questions.

Enjoy. I'm looking forward to some possible creative uses. Please email me if you get it working as I'd love to see what you're doing with it. If you come up with a particularly interesting use, I'll be sure to showcase it here.

[tags]PHP, Programming, Virtual Card Catalog, AADL, Web 2.0, Library 2.0, Open Source[/tags]