If you were to go into my Outlook mail box(es), you would be lost with little hope of finding your way back. I don’t know what I use for organizational categories and I rarely plan it out. Generally, I use what comes to mind. The reasons for all this chaos is that I rarely think about a repository for my own reuse – I have search engines for that. I tend to think in allegories and metaphors anyways; it is easier for me to recall a phrase than a tag. Given that situation, a full text engine works better for me.
That’s just a personal preference, however. If I am designing a system for other people to use, then I always consider my grandmother. (The link takes you to my growing digital collection of photos from before the Depression; I built this for LIS-665). My grandmother would have had NO IDEA what a browser was, let alone a search tag. She knew the Dewey Decimal System pretty well though. It was all they really had back then. So designing a set of tags for recall is something to consider carefully.
As Markus would point out, the design of a repository is driven by the intended use at the time of design. Often the use case changes over time and so the repository may not be suited for the questions put to it. It is not always necessary to consider the repository reuse question at design time but – and this is a large ‘but’ – you should at least consider if the repository is for you, or for others.
For this course, the requirement was to use Bibsonomy.org. You can see my posts here and the tag cloud on the right shows that I’m not terribly organized about it. I didn’t have an intent or plan to reuse it for anything, so a tagging system was just one more thing to do. I tend to use EndNote for bibliographic management and that connects to Clarivate OnLine. Ultimately, these are just tools.
I took LIS-665 at the same time as this course. The final project(s) are to study a large library’s digital collection and to build one of your own. I studied the New York Public Library’s Digital Collection which is just huge; some 30 Million items. You could really get lost in there.
But they do a fantastic job and it’s great to just wander around. Until I built my own collection, I did not realize the value of the tagging scheme. Much like Carolyne Millsap, I just had no idea of why I would need this. It was later on, when I started to amend the tagging system to fit my own recall and reuse needs. For example, I started to tag all the articles I read for the class with “LIS658”. I still need to go back and update many of them. Articles used for the paper were tagged, “Thesis”. I briefly considered using ‘Paper’ for the tag but that’s too common a word. An example of that would be my experience helping Philip Morris to use our taxonomic classifier – an engine that would read the documents and suggest keywords for you. They chose “Nicotine” as a major keyword but it didn’t really help much; 80% of their documents dealt with nicotine. They later abandoned that idea. Another was the BBC, who wanted to use the engine to classify their news stories. The problem with the BBC’s implementation was that the taxonomy would change 50x a day – literally – because news is about a constantly changing world. It’s really hard to have an organized approach to tagging when the fact base is constantly changing.
The point of the tagging system on Bibsonomy was simply to fulfill the requirement and to help me a little bit with the scoring. I still use EndNote, for example, it puts the citation directly into my word documents. Like Wil Silberman said, “whatever you do, you should use something”. Wil’s post reminds me somewhat of the scene between Shepherd Book and Mal in the movie Serenity:
and I include that here because it is a siren call to do well whatever it is that you do, to believe hard in yourself. Pick a scheme and stick to it.
Be like a postage stamp. Stick to one thing until you get there. -Josh Billings