when open source is not actually open source, or what i learned at midwinter.

I typically recycle every last bit of vendor flotsam I receive before the ALA Midwinter and Annual conferences, not being patient or seizure-resistant enough to handle the exhibits floor in any meaningful way. This time around, however, I decided to take part in an extremely lucrative focus group on (anonymous) new software for “Making Database Searches Easier.” The promise of $200 simply for jawing about something I’m interested in for 1.5 hours was too much to resist, and it turned out to be an experience that gave an intense amount of insight into how vendors develop and market library-related products.

The focus group consisted of 5 librarians from ARL institutions (UCSD, Princeton, Penn, Chicago, and Ohio University), all of whom had considerable interest in public services and search/discovery tools. Four groups had come in before us, giving you an idea of the potential pricetag associated with this round of testing alone (5 times 5 times $200 for starters equals a healthy budget, in my mind).

The software in question actually sounds great, in the event that it works and is marginally affordable – the facilitator described a hosted metadata harvesting and indexing one-box search tool that will ostensibly layer over all of a library’s search and discovery resources and allow users to conveniently filter results based on a number of criteria, etc. The interface would be highly tailored to a library’s needs and aesthetic, apparently, and would support development by local tech staff. The product seems in the very early stages of development – no interface testing or images were given, only concept-related descriptions and vague marketing pitches.

The thing that stood out to me was the idea of a hosted harvesting one-box searcher, which I think could eliminate two of the biggest headaches of federated searching or metasearching – namely, the amount of time it takes to generate results and the limited array of resources that are queried simultaneously. I took away from the session that this tool would layer over the top of all of a library’s resources, including OPACs, databases, ejournals, digital collections, blogs, wikis, and outside sites, and would harvest updated metadata from all of these sources on a frequent basis. This current metadata would act as the source of the query, the actual database or catalog being engaged only once relevant information is located. In the absence of more information this is about as far as I can go in terms of a description, and if anyone knows of a similar tool currently available I’d love to hear about it. Summa Integrated Search comes to mind as an open source alternative.

Speaking of which, a troubling aspect of the experience was the product developer’s totally incorrect and unfortunate misuse of the term “open source.” The facilitator at one point made the unbelievable statement, “this product will be open sourced,” leading several of us to wonder where in the world they got the funds to pay untold numbers of librarians $200 to give feedback on their product. After going a few semantic rounds with the facilitator it turns out that the product will most definitely not be open source, but highly customizable and subscription based. More disturbing than the incorrect usage of this concept was the fact that none of the four previous groups had caught the mistake, and that whoever is developing/marketing this tool either doesn’t know what open source is or is using the term incorrectly to capitalize on its current cache.

I was also struck by the number of exceedingly good ideas that were thrown around by our group in the span of 90 minutes – if wishes were keystrokes we would have one hell of an integrated search interface operational at the moment. In addition to strengthening my resolve to improve my coding skills by leaps and bounds over the next few years, this session reinforced my opinion that librarians and systems folks should focus on pooling expertise and developing the types of open source software solutions that will make us less reliant on vendors (and increase competition among the vendors that remain, which may result in some much needed price cuts here and there).