[Web4lib] MARC to XML: the agony and the ecstasy

[Jonathan Gorman]

>I think one of the challenges you are going to face is that XML isn't the
>answer to your question. The answer is SQL or some other decent and open
>database. XML is the transmission and display technology, but the real power
>is that it is so much easier to rip into a database to be called by dymanic
>websites. 

I disagree here slightly.  Databases aren't always the answer.  Indexing engines can frequently be useful when talking about data and a good number of indexing systems out there happily process XML that would likely have issues with MARC.  

Add to that the fact that there are efforts behind XML databases that use query systems like XQuery and XPath that are promising.  It's been a while since I looked at any so I'm not sure how far they come along.

But this is getting rather far away from what I see as the two underlying issues in Karen's post:
1) Data formatting and how the underlying AACR2 and practices cause issues regardless of transmission or how the data is stored.  A bit of transcribed or free text can be awfully difficult for a machine to parse, no matter if it's in XML, Mork, JSON, or MARC.

2) Ease of manipulation and tools of the actual data formats themselves.

Anyhoo, back to work,

Jon Gorman