> Are alpha characters used for field names valid in MARC records?
> When we do dumps of MARC records our ILS often dumps them with FMT and CAT
> field names. So not only do I have glorious 246 fields and 100 fields but I
> also have CAT fields and FMT fields. Are these features of my ILS --
> extensions of the standard -- or really a part of MARC? Moreover, does
> something like Marc4J or MARC::Batch and friends deal with these alpha field
> names correctly?

MARC is just a binary format that does not care what the field names
are, though they do need to be 3 characters long. Even within the
library community, there are multiple MARC flavors, and you'll find
elements in different data fields.

What you're thinking of as MARC are really bibliographic standards. To
address your question directly, these are peculiar to your ILS.

I would guess that any MARC processing tool would have no problem with
alpha fields because the index built into the MARC record indicates
the field names and where the data are and there would be no benefit
to having a parser evaluate the name of a tag.