There’s been a lot of discussion recently about the state of the catalog, both here and on other blogs. It’s such a hot topic that I get double or triple the number of messages from ngc4lib than from other lists. (If it weren’t for the incredibly interesting and insightful posts on that list, this kind of volume would force me to unsubscribe instantly, but luckily it’s worth every minute.)
Over on One Big Library, dchud has been arguing that there’s a problem with this bill of rights because we signed the contracts so we’ve only ourselves to blame. And surprise, surprise, this has sparked off more discussion.
Now, don’t lynch me before you hear me through to the end, but I’ve got to add another bit to this whole discussion. At our last users group meeting Dinah Sanders was talking to people at the reception, and the topic of conversation was pretty predictable given the general state of things. These users all wanted to be able to do stuff with the catalog… web 2.0 stuff, fun stuff, necessary stuff, stuff that should be do-able. And Dinah’s response was always the same. She’s been wanting to do all that, too, but there’s no time, and they aren’t supposed to fix stuff unless they’re actively supposed to be working on that module or code. Her refrain was (and I quote), “As we touch it, we can fix it.” YIKES!!!!!! That’s the definition of over-extended! It must be horribly frustrating to work on a project where you can see necessary fixes but can’t do anything about it.
Wait! Don’t tar and feather me yet. Hold off for a few more seconds. Remember, I said I can empathize. But this undesirable situation is exactly why we need to rethink the ILS. If it were modular, each part could be engineered by people who had time to keep up with technology. If it were modular, the discovery system wouldn’t have to be designed by companies that excel in collection or budget management, for example.
There would be new problems with this modular approach, of course, and standards would have to be clear so that the modules could be seamlessly interoperable. But having a system that’s so big that obsolete code can’t be fixed unless the entire section is being reworked anyway is a long way from a perfect solution.