Am 21.11.2011 17:14, schrieb J. McRee Elrod:
>> Usefulness
>>    Cover all usage requirements
>>    Form supports function: Minimize effort of the user
>>    Flexibility: Openness for future requirements and new options
> Later you address interfaces, but do not explicitly mention *display*,
> a function well met by Avrom's orriginal MARC design (but ignored by recent
> additions); and met by AACR2, but ignored by RDA.
In Avram's time, display was the only task for the data, card
production was MARC's raison d'Ítre.
And for the cataloging rules, card layout was eminently important
because the card was at once the product of cataloging *and* the user

Now, the product of cataloging has to fit into many more environments
but no longer into card drawers. Accordingly, data have to meet many
new requirements. Beside human-readable displays, we have to create
services that deliver machine-actionable data. This is where our
legacy stuff is very deficient and inconsistent, and RDA hasn't quite
arrived in that world either.
RDA does not ignore ISBD altogether, but it assumes that catalogers
have no direct business any more with the user interface. The pasting
together of an ISBD display, in other words (or any other, for that
matter), ought to be a purely automatic process, and that includes the
insertion of punctuation.

And LC's 2008 report on the Future of Bibliographic Control
clearly stated:
   "4.1 Design for Today's and Tomorrow's User"
which may be understood to imply less or no regard for yesterday's
user's expectations.

And yet, for all the new qualities and functionality our data are
supposed to have, the end products will have to satisfy the needs of
humans - as long as machines don't replace the readers.
I.e., something will always have to be displayed for human consumption.
However, that includes a lot more than just displays of single catalog
records, like search results in various stages, index listings,
navigation through subject headings and classifications ... all
derived from catalog data but not guided by RDA (or AACR) either. And
then add to all that all sorts of enrichment stuff in the widest sense,
not covered by any cataloging rules at all.

The display issue itself is rather old:
The loss of a standardized user interface in new OPACs, and the loss
of influence of catalogers on that interface,  was the subject of a
series of postings to Autocat in about 1994, under the
title "Face the Interface". Nothing much ensued back then but
around 2003, IFLA ran a Task Force on Guidelines for OPAC displays, and
Martha Yee came up with a still recommendable report. Search for
"Martha Yee" "opac displays" to find out more, or 
"YeeOPACGuidelines.pdf" (containing a list of 37 principles). Even at
that time, machine actionability was not a subject, nor web services.
But by no means do those new subjects make the old issues of human
interpretability of displays go away. And yes, it should still serve
libraries well to have a recognizable standard for bibliographic data
display, applied and understood the world over.