On Dec 10, 2018, at 11:38 PM, Eric Phetteplace <[log in to unmask]> wrote:
>> 4. Metadata is not necessary.
> Could you unpack this? I think I know what it means but I'm not sure. Thanks for the concise report.
Actually, the full quip was "Metadata is not necessary. Well, that was sort of a debate, and (more or less) deemed untrue."
Different flavors of artificial intelligence (AI) are being used to generated metadata, and there was some discussion regarding the need for human-generated metadata, or more specifically, metadata in the form of controlled vocabularies. Controlled vocabularies, specifically subject headings, are not objective (without bias), but they are/were all but necessary in a world sans full text. On the other hand, a person now-a-days can use any number of different techniques to generate subject terms or tags which are centered around the reader or the content as opposed to an outside entity such as a library. Examples include topic modeling or the calculation and ranking of term frequency / inverse document frequency (TFIDF) scores.
In the end, we all thought metadata is/was necessary, but not necessarily as necessary as before. :)
University of Notre Dame