Recently a couple things happened that make me despair of ever having prior work not be repeated.
The first incident was at a large library conference at the beginning of the year, with a panel about aggregating metadata from multiple contributors. The room overflowed with attendees, as the topic was much more popular than the conference had allowed for. I first sat on the floor, then stood, but I was pained not by my lack of a chair, but by the proceedings.
We listened to individuals who seemed to imagine that this was the first time anyone had attempted to aggregate metadata from diverse contributors. I was in awe of Diane Hillmann, who was on the panel. She sat there mute and expressionless as the other panelists completely failed to acknowledge the work of the National Science Digital Library that had preceded their attempt by some years and that Diane had been deeply involved with. How she remained silent I will never know, and I am embarrassed that I failed to speak up on behalf of her and her colleagues whose prior work in the field was unacknowledged.
But it wasn’t just the NSDL work that was being ignored — no one was cited as having contributed anything to their thinking about the issues or their strategies for dealing with them. Europeana is an obvious example. UKOLN perhaps. JISC is another. As is CIC. And CDL. You can’t take a step in a Google search of “aggregating metadata” without falling over prior art.
Then a few months later I was at another event where something similar happened. It was an informal discussion about data aggregation issues. In this case someone was describing writing code from scratch to cleanup dates, which are well known to be a big problem in metadata aggregations — particularly ones that include MARC data. It turns out they had never heard about the utility that CDL developed many years ago to do the same basic thing. Because they hadn’t even looked. A simple Google search on “date normalization” turns it up as the first hit.
When faced with a challenge, the very first step should be to investigate what has come before. The very first. Only by doing so can you avoid relearning lessons that were learned years ago, and that could help you avoid the bumps and bruises that learning as you go will entail. This will not always be the case — you may need to do things completely differently and create things from scratch. But if you do, you will know with a certainty that you must.
Believe it or not, it is possible to learn things from the previous experiences of others. Sometimes, just sometimes, it didn’t start with you.