pointing to the first item about marcspec

[cKlee]

Another possibility is to use a more XLSTish syntax like 100[0] for the first author, 100[1] for the second author and so on...

[cKlee]

But what is about repeatable subfields? Pointing to an index of a subfield in an index of a field will become very verbose.

100[0]a[0]

Maybe not a good idea?

[nichtich]

If aimed at trained programmers, the index should be zero-based, if aimed at non-programming people, one should consider 1 to refer to the first element. Indexes, such as 100[1], 100[2], 100a[1], and even 100[1]a[1] are fine IMHO - if this feature is important enough to be added to MARCspec. One could also allow ranges, such as [1-2], [2-], and [-3], but use cases are probably limited. The data is further processed anyway, so more complex processing can also be done by other means instead of MARCspec.

[cKlee]

Yes 1 for the first element seems more obvious. But I think librarians familiar with MARC already know about the character positions and that the first index is always 0. I think we should keep this consitent: First character and first element should both use the index 0. Otherwise it gets confusing.

[cKlee]

Done with tag v0.4draft