Wednesday, 6 October 2010

Standards

I find myself talking about "standards" all the time, but meaning it in at least two different ways. The first is to mean recognised international standards in cataloguing and indexing - which, at the moment, in the UK at least, means AACR2, M21 and Dewey. It means following generally accepted rules, best practice, because there is safety in numbers, when we do what everyone else does, and because we believe ourselves to be abiding by principles which the best minds in the business have devised over time, so that we not only gain the undeniable and valuable benefits of interoperability, sharing and cooperation, but that somehow we are also doing "the right thing" and as far as possible future-proofing our data. But that simply isn't the case any more. We may input the data in AACR and M21 and Dewey, but our LMS's hold that data in all sorts of ways, they break it down and fragment it and build it all up again into structures that we have no control over and wouldn't even recognise if we saw it. The people who design our Opacs (not that they even call them Opacs now - the Opac is dead, apparently, and what we have got is a resource discovery tool) are spending their days devising new ways of "mining" and "parsing" the data which we thought we had got into the best possible format in the first place. So much for standards, if they have to be broken and re-made before they can be used.

But I also talk about "standards" when what I really mean is "quality" - keeping high levels of accuracy and consistency so that our records are fit for purpose and can be used as the foundation of an excellent service. These are the standards that we know we mustn't drop, even when we have to cope with reduced resources - instead we must work harder and smarter. But hang on a minute - how are we going to define quality if we don't have anything to measure it against? If we don't any longer have any agreement about what constitutes best practice, if we don't have agreement about shared formats and codes, how can we tell whether our records are any good?

If a language doesn't have a grammar, how do we know whether that language is being used properly? How can we communicate if we don't share meaning and structures? Are we reduced individually just making it up as we go along? Can anyone understand us? Is anyone still listening?

No comments:

Post a Comment