> to define values for fuller sequencing of the display of names.
I have been thinking about this possibility too, but I did not implement 
such a thing and chose to follow the MODS schema. The question is 
interesting but I consider that the answer is not trivial. My aim with 
this e-mail is make a try to identify some of the difficulties that such 
an effort would encounter.

If you deal with ONE kind of records (for example scientific articles from 
ONE periodical, within ONE narrow discipline and ONE period of time), you 
may be able to find one rather consistent way to (group and) sequence the 
several authors of each article.

Dealing with different journals or disciplines, you will soon notice that 
different strategies are used to group and / or order the authors: Authors 
may be grouped on different lines (according to some criteria) and 
alphabetically ordered within each line (or paragraph if there are dozens 
of authors) but completely different strategies (affiliation, seniority, 
etc.) may be used to group and sort authors in other publications.

Finding a consistent and accurate way to deal with all that is rather 
difficult if MODS is to be useful for a wide range of scientific articles, 
novels, books for children and much more.

Of course, we could consider different classes for "author grouping":

- typographic according to the font characteristics used to write the name 
of each author

- "topographic" according to where the name of each author occur on the 

- relational to the other authors' names

- cluster-wise (for example if several names are on the same line)

Let us also consider the fact that different authors may have had 
(slightly) different roles.

Maybe other people on the list have other ideas or experience about those