Lightweight protocols for lightweight applications


I believe that a lot of the buzz in the library world to do with emergent technologies is related to the fact that we’re seeing more and more lightweight protocols and data formats delivered by library service providers. I say protocols and data formats because a lot of the existing protocols used by libraries are pretty simple, whereas the content has typically been provided in a heavyweight format (at least in terms of what a typical library employee is accustomed to).

Things like SRU and JSON — Search/Retrieve via URL and JavaScript Object Notation — are making it easier to write lightweight web applications without recourse to specialist programming libraries (remember the bad old days of Z39.50?). Lightweight-web-applications programming needs no server-side wizardary, only simple proxies to facilitate cross-domain requests (and even this is unnecessary in cases where data is provided in JSON format).

This kind of democratization means that libraries can — with the help of a few tech-savvy librarians — provide impressive interfaces for their content, create added value at little cost and deliver something that works well enough to be acceptable for end users. The only problem is that the core library systems are typically antiquated enough that they don’t provide any real support for such lightweight protocols. An example of this is a system that provides an HTTP API for data harvesting, but no corresponding way of ordering an item that has been harvested.

What is needed is an understanding from library systems developers and content providers that libraries with a strong online presence want to integrate services into their web, and present information from diverse sources within a common template as far as this is possible. The traditional focus on harvesting relates to a similar trend: we used to want data in our OPACs and reference tools, now we want it on our pages.

A really important point for me is that presenting data from diverse sources means that this data can then be re-purposed according to the display device that the user has chosen (laptop, tablet, text-to-speech, a DIV in a user’s social networking site, mobile device, retinal HUD). I think that libraries are going to have to focus more on providing re-purposed information in this way in the future, and these lightweight protocols are needed in order to gather the information to be re-purposed

Tags: , ,

%d bloggers like this: