It’s funny, when lcsh.info went live it was using the Fred2.0 data but 2 days later Barbara Tillett asked me to use the canonical MARCXML data from LC…which it was using up until last week.
The conversion code that I wrote should still be able to work just fine with the crawled fred2.0 data. The only thing that might need tweaking is where to locate the LCCN, since I believe this was moved from the 001 in your data (if memory serves).
Thanks for Fred2.0 btw, it and Barbara & Corey’s paper about authority data and the semantic web were the inspiration for lcsh.info.
The problem is that V’GER didn’t (doesn’t?) validate the LCCNs, so if there’s a typo entering the record, updates don’t match correctly. There are several duplicate LCCNs, and proposed headings that co-exist with un-proposed).
Simon
p.s.
Apologies for reference to odd numbered Star Trek movie, but I think that software only deserves odd version numbers
It’s funny, when lcsh.info went live it was using the Fred2.0 data but 2 days later Barbara Tillett asked me to use the canonical MARCXML data from LC…which it was using up until last week.
The conversion code that I wrote should still be able to work just fine with the crawled fred2.0 data. The only thing that might need tweaking is where to locate the LCCN, since I believe this was moved from the 001 in your data (if memory serves).
Thanks for Fred2.0 btw, it and Barbara & Corey’s paper about authority data and the semantic web were the inspiration for lcsh.info.
The LCCN is in 010.
The 001 is the local id from V’GER.
The problem is that V’GER didn’t (doesn’t?) validate the LCCNs, so if there’s a typo entering the record, updates don’t match correctly. There are several duplicate LCCNs, and proposed headings that co-exist with un-proposed).
Simon
p.s.
Apologies for reference to odd numbered Star Trek movie, but I think that software only deserves odd version numbers