MARC – the dabate goes on – what is it?
I just read a thread on the Code4Lib news group regarding whether there was a MIME type for MARC records. Inevitably – as in any conversation touching however briefly on MARC – the topic of ‘what is a MARC record?’ comes up. Now I know that there are not only different flavours, but different implementations of those flavours, and it doesn’t stop there because there are even different representations of the flavours. All of that is why people want to use MIME types to explicitly declare what the content of a http response is.
Then a thought flashed through my mind – “This is a bit like the semantic web”, followed by – “But isn’t the semantic web supposed to fix this?”
Let me explain what I mean. As I see it there are several flavours of ‘semantic web’ out there. There is the flavour that is microformats: an attempt to bring meaning to web pages and other xhtml based documents. There is the flavour that is ‘pure’: rdf, a means to describe documents to within an inch of their lives. Then there are things like RDFa, embedded RDF, and, somewhere in between, a whole raft of ontologies, schemas and vocabularies to describe the things that turn up in RDF.
Sounds (and feels – when trying to use semantic web techniques) a bit like the current state of MARC to me….
So, who is going to be the first to write an application that is solely designed to help me describe things?
Feature set:
- Look up suggested vocabulary as I type
- Show me how that vocabulary fits in with other vocabularies (could I be using better words?)
- Show me the same vocabulary in my preferred language
- Show me pictures and diagrams to allow me to see how my objects interact.
- Expand the above interaction to include objects that have already been described by other people.
- Don’t bother me with having to learn specific markup. (but let me see it and explain it to me if I am interested)
- Let me then save that data in as many formats as I could possibly wish, including: Human readable thesaurus, taxonomy etc, as well as various flavours of XML, turtle, RDF, OWL etc.
Anyone want to bite?