It really depends on the requirements. For data with a long expected
life-time, XML may actually be a better choice than comma/tab delimited
file _because_ it is so verbose. If the original architects choose tag
names which are clear, then when someone approaches the data 10 years
later, and the original authors are long gone, the chance of this new-comer
understanding the data format increases significantly. This is what Steven
Newcomb calls self-descriptive documents. (Steven/Peter, did I get that
right?)
I have been bitten by problem that I write a quick and dirty data-dump tool
which dumps out to a tab-delimited file and then, a year later I can't
remember exactly what all the fields were. XML can help. It is not a
perfect solution, but it beats re-engineering software (esp if you don't
have source any more....)
but again, it all goes back to your requirements. If your data is only
going to be used by 3 programs you wrote, and the data has a short life
expectancy, then tab-delimited files are a good choice.
-derek
Derek E. Denny-Brown II || ddb@criinc.com
"Reality is that which, || Seattle, WA USA
when you stop believing in it, || WWW/SGML/HyTime/XML
doesn't go away." -- P. K. Dick || Java/Perl/Scheme/C/C++