> From: Geoff Mottram [mailto:[log in to unmask]]
> Sent: Wednesday, April 03, 2002 5:25 PM
> I will have to disagree with you here. I don't see "displayForm"
> as presentation in the same way that font specifications and
> layout commands are. I also think you are expecting too much of
> your data consumers. If you want to subfield your data extensively,
> it should still be easy for a user of your data to display it
> without having to understand your content model.
More markup makes it _easier_ for data consumers. Because they can
rearrange the pieces so that they make sense to their community.
Your markup already includes the required base elements of <first>
and <last>. From those two elements I can put the names from your
records in any order I desire. From a technical standpoint, to make
<first> and <last> into <displayForm> I'm looking at a simple four
line XSLT transform, that I probably could collapse into one line.
> At the simplest level, MODS should contain the filing version
> of a field and, if its different, a display version.
Filing can easily be achieved by appropriate markup, you suggested
for example <nonsort> someone else suggested <article>. By taking
textual information from the appropriate tags and then running it
through the NISO filing rules algorithm, filing can be easily
derived. Again from a technical standpoint, I'm looking at a few
or java depending upon which XSLT processor I'm using.
> I shouldn't have to figure out what's a first name, a last name
I'm a little confused here since your example already marks up
<first> and <last> so how are you trying to figure it out? One
of the reasons for my refinement proposal was that if MODS or
someone defined <name> and all they did was:
then this is probably not going to give you the flexibility in
extracting the pieces that will allow you to display it to your
community in a different manner. So if you refine the above
<name> to be:
you now have the ability to collapse the refinement to the
original non-refined or to display it to your community in a
different manner. This is the essence of markup. The comma
in the non-refined is actually taking the place of markup.
Something, markup designers need to be aware of when defining
a content model.
> What's wrong with using whatever punctuation is appropriate
> for the language in use in a given record? A Greek record
> would use whatever punctuation they would normally use.
Imposing AACR2 on non-English countries just doesn't work. Many
countries have different rules for displaying dates, times, names,
numbers, etc. All of those items I just mentioned use different
punctuation for separating parts within the item. This the reason
why you have a Regional Setting Icon in the Control Panel of
Windows. So instead of providing a date string of 04/03/2002,
you should be marking up the day, month, and year. In many
European countries the year comes before the month and day, in
other countries, the day comes before the month. So in my example
is that month 04 day 03 or month 03 day 04?
> If you remove punctuation, I fear an extreme amount of markup
> required and make the standard too difficult to implement.
LC may decide, for example that they don't want to specifically
define <month>, <day>, <year> in my date example above. They
may feel that AACR2 rules are sufficient. After all LC is a U.S.
based entity concerned mostly about U.S. based standards for U.S.
based libraries. For organization that have international
members AACR2 may not be sufficient if you want to think globally
and act locally. Through my refinement proposal, my community
can take LC's records transform them though XSLT into a more
refined, e.g. more marked up, record.
I'm not proposing that LC develop a deeper level markup [although
they might do so through my refinement proposal (non-normative)
to roundtrip MARC21 to MODS to MARC21], just a framework were
refinements can be made and a way for my community or my data
consumers to collapse my refinements back to something sensible.
> > elements. Obviously, refinement elements and "smart" XLST
> > transforms could, and probably should, be shared across
> > communities.
> I am concerned with any proposal that requires transforms in
> order to use the data. A well designed schema should suffice
> for most users.
I think you missed the point of my refinement proposal. It's
a standardized mechanism to allow different communities to
further refine the base MODS standard. The "smart" XSLT
transforms are a mechanism for the community whom made the
refinements to dumb down their refinements for interoperability.
All refinements revert back to #PCDATA. If you choose not to
make refinements to the base MODS standard then my proposal
doesn't come into play. If you do, then it does, e.g. it's
In the simplest form the MODS standard could be a set of 19
elements who's content model is #PCDATA. Everything, else
could be handled through my refinement proposal. Although,
I suspect, LC probably would define the 19 elements a little
deeper. But the point is that the base standard could be
19 elements of #PCDATA just like Dublin Core is. You cannot
get any simpler than that.