Why bother with information architecture? — RenaissanceCMS. I was happy to be asked to write something on information architecture generally for Rob’s blog. It’s easy to forget that not everyone takes for granted the usefulness of IA, so I have tried to inspire people who aren’t sure what it is or what it can do.
Rob creates charming ethereal designs as well as working on marketing, branding, and visual identity and being generally ethical and sustainable. I particularly liked his latest post on tagging. I tend to approach folksonomy from a management and retrieval point of view, and so find myself arguing that just because it is cheap, doesn’t mean it can replace all other KO systems. However, I have been thinking about image retrieval, and one area where social tagging is useful is in labelling vague and abstract ideas like “mood”. If most people tag a photo as “sad” or “mysterious”, that is probably going to be useful for creative people who don’t need a specific image but are just after something that evokes a “feeling”.
Social Tagging and the Enterprise: Does Tagging Work at Work? | Earley & Associates confirms my suspicion that people aren’t as enthusiastic about content tagging at work as they are at home, tagging for fun or when it is their own content. As usual, it’s all about context. There are lots of ways tagging can work at work, but it isn’t reliable for core or high value asset management and information retrieval.
Taxonomies are sometimes criticised for not being scalable (or at least not scalable cheaply), but folksonomies only work really well when the scale is massive (e.g. web-wide). Most enterprises are right in the middle, which is presumably why synergistic solutions remain popular.
Here’s a similar take: Explicit and implicit metadata @ CommonPlace.Net.
The ever excellent Green Chameleon covered another aspect: Green Chameleon » The War Between Awareness and Memory.
Public knowledge, private ignorance by Patrick Wilson is one of those fascinating books that reads as if it had been written yesterday, but in fact was written in 1977. In what struck me as such a contemporary theme, he discusses the importance of personal contacts and trusted authorities as sources of knowledge - a theme that has returned with a vengeance in the form of “social search” and leveraging social networks for recommendations etc. A wonderful example of this was given to me by a friend last night who told me about how their archives division suddenly gained recognition when the new media lot realised they needed metadata for their rapidly growing digital repositories. The new media folks were in a panic until the talked to the librarians and archivists and realised they had already worked out - and been assiduously cataloguing all the digital assets. Without that personal contact, the new media folk might have ended up building their own catalogue, and duplicating all that work! It’s a sadly familiar story even in these days of information abundance when you’d think such communication would be easy, but even the problem of how to make sure “public” knowledge is actually used is nothing new. Wilson quotes Lord Rayleigh at a meeting of the British Association of the Advancement of Science in1884 (yes - eighteen eighty four - it’s not a typo) as noting how much scientific knowledge was published but unread, saying “It is often forgotten that the rediscovery in the library may be a more difficult and uncertain process than the first discovery in the laboratory”. Wilson points out that “knowledge existing only ‘in the literature’ is no different from knowledge possessed by undiscoverable or inaccessible individuals”.
Wilson also claims that “where there is knowledge, there must be a knower”, which struck me as a challenge to notions that publishing alone - whether it be tweets or academic research papers - is only half the battle. This reinforces to me the absolute fundamentalness of findability and serving the needs of the user both within individual publications and across the whole of our “public” digital repositories. Again, this is nothing new. Way back in the 19th century Charles Cutter was anxious to serve library users better, Grace Kelley (disambiguated from the other one by the extra “e”!) in 1937 was what we would now refer to as a “usability evangelist” and both Ranganathan (1959) and Bliss (1935) had a passion for getting the right information out to the right people.
As Computer Science from the 1970s took over much of information retrieval and then with commercial products being heavily marketed, I worry that this sort of passion has been lost in a blur of what you can get an algorithm to do, rather than what people actually need. As Wilson says “we do not make knowledge available simply by making available documents in which knowledge is represented”. Google is wonderful, but that is, essentially, all it does. Of course the more sophisticated programmatic tools we have the better, but as information providers we should never be afraid to say “this gets us so far, but not far enough”. We need to keep reminding everyone that it is the minds of the knowers and potential knowers we need to be serving and so we should not be afraid to keep demanding ever more sophisticated systems that are mixed, variable, and downright difficult to automate.
I’ve just read Steve Bailey’s book Managing the Crowd: Records Management 2.0. It is a thought-provoking and timely read and very enjoyable as well. There’s an RM 2.0 Ning site too. There’s a good summary on the TFPL website. Bailey is clear that he is trying to provoke debate, so I will raise a question. There is a widely held belief that people like tagging, but I’m not sure that this applies once you get into the office. People love to tag their own photos on Flickr, but is that because they like tagging or because they like their own photos? Similarly, people like to tag their own blog posts, but is this not a rather self-selecting sample? If you have the time, motivation, and energy to blog, the additional burden of adding a few tags to try to get yourself a few more readers is hardly great. So, is there any evidence out there that people tag work documents just as enthusiastically as they tag stuff about themselves? Are they really as enthusiastic about thinking of appropriate tags for financial reports and product information sheets as they are about tagging their favourite songs or You Tube clips?