On 12/16/03 8:57 AM, Eric Lease Morgan <[EMAIL PROTECTED]> wrote: > Upon further investigation, it seems that MARC::Batch is not necessarily > causing my problem with diacritics, instead, the problem may lie in the way I > am downloading my records using Net::Z3950....
Thank you to everybody who replied to my messages about MARC data and Net::Z3950. I must admit, I still don't understand all the issues. It seems there are at least a couple of character sets that can be used to encode MARC data. The characters in these sets are not always 1 byte long (specifically the characters with diacritics), and consequently the leader of my downloaded MARC records was not always accurate, I think. Again, I still don't understand all the issues, and the discrepancy is most likely entirely my fault. I consider my personal catalog about 80% complete. I have about another 200 books to "copy catalog", and I can see a few more enhancements to my application, but they will not significantly increase the system's functionality. I consider those enhancements to be "featuritis." Using my Web browser I can "catalog" about two books per minute. In any event, the number of book descriptions from my personal catalog containing diacritics is very small. Tiny. Consequently, my solution was to either hack my MARC records to remove the diacritic or skip the inclusion of the record all together. The process of creating my personal catalog was very enlightening. The MARC records in my catalog are very very similar to the records found in catalogs across the world. My catalog provides author, title, and subject searching. It provides Boolean logic, nested queries, and right-hand truncation. The entire record is free-text searchable. Everything is accessible. The results can be sorted by author, title, subject, and rank (statistical relevance). A cool search is a search for cookery: http://infomotions.com/books/?cmd=search&query=cookery Yet, I still find the catalog lacking, and what it is lacking is/are three things: 1) more descriptive summaries like abstracts, 2) qualitative judgments like reviews and/or the number of uses (popularity), and 3) access to the full text. These are problems I hope to address in my developing third iteration of my Alex Catalogue: http://infomotions.com/alex2/ My book catalog excels at inventorying my collection. It does a very poor job at recommending/suggesting what book(s) to use. The solution is not with more powerful search features, nor is it with bibliographic instruction. The solution is lies in better, more robust data, as well as access to the full text. This is not just a problem with my catalog. It is a problem with online public access catalogs everywhere, but I deviate. I'm off topic. All of this is fodder for my book catalog's About text. Again, thank you for the input. -- Eric Lease Morgan University Libraries of Notre Dame
