I agree that the community should agree on best practices. I for one will
make children of the verso and recto of photos when the verson contains
captioning information, and treat the image solely as object only when the
verso is blank.
Sue
At 12:27 PM 1/11/99 -0800, you wrote:
>As the tools person--the one who is going to have to cope will all of the
>variations in some fashion for presentation purposes--I am all in favor of
>Johns best practices suggestion. I agree will Merrilee's and Jerry's
>analysis of how Sue's case should be handled. I don't think that Sue's
>alternate handling would throw the tools for a loop as they are currently
>constructed; but I am expecting a lot of loops in any event, as the DTD
>evolves and as it becomes clearer how people are actually going to make use
>of the DTD.
>
>With respect to the last point: Sue seems to be moving along. I wonder if
>she (or anyone) has enough done that they might be willing to share their
>database in it its current state with us just for informational purposes?
>Probably its premature; but I can hope, yes....?
>
>Rick
>
>
> At 11:02 AM 01/11/1999 -0800, John Hassan wrote:
>>i agree with m and jerry. this is another case, though, where there
>>probably should be some kind of best practices about how to do it, so that
>>there isn't a lot of variation when we look at each other's objects.
>>
>>
>>At 04:36 PM 01/08/1999 -0800, Jerome McDonough wrote:
>>>At 03:58 PM 1/8/1999 -0800, Merrilee Proffitt wrote:
>>>>I think, actually, that the latter is correct; the _object_ is the photo,
>>>>with _subobjects_ recto and verso. If you had made a decision to only
>>>>record the print side (leaving out all mention of the verso), however, I
>>>>think you would do well to just record the information at the object
>>>>level, and not go down another level. Which seems a bit paradoxical, and
>>>>makes me unsure of my answer. Which is why I'm bouncing it out there for
>>>>all of you.
>>>>
>>>
>>>Merrilee has it right (at least, it matches with my understanding).
>>>In the case of only recording the print side, you can consider the image
to
>>>stand in for the entire object; in that case, there's no need for child
>>>subobjects. If you're going to image both sides, however, it's better to
>>>assign these images to two subobjects of the photo object, one for the
recto
>>>and one for the verso. That way, when people examine the navigation
>>hierarchy
>>>for the object, they can see that the two images are at the same level of
>>>the tree and are basically paired. Making the verso a child of the
>>>photo object where the photo object is matched with the recto image implies
>>>that
>>>the verso image is a subsidiary part of the recto image. I think it's
>>>much clearer to have the tree balanced when you're going to image both
>>>sides, with both recto and verso subobjects for the photo object.
>>>
>>>This is not that huge a deal; the sky will not collapse if you make the
>verso
>>>of a photo an only child under the photo object. But it will look a little
>>>odd to have a navigation hierarchy for the object in which you have a root
>>>node for the whole photo, a branch for the verso, and no branch for the
>>recto.
>>>
>>>
>>>Jerome McDonough -- jmcdonou@library.Berkeley.EDU | (......)
>>>Library Systems Office, 386 Doe, U.C. Berkeley | \ * * /
>>>Berkeley, CA 94720-6000 (510) 642-5168 | \ <> /
>>>"Well, it looks easy enough...." | \ -- / SGNORMPF!!!
>>> -- From the Famous Last Words file | ||||
>>>
>>>
>>
>>
>
>-----------------------------------------------------
>Rick Beaubien
>
>Software Engineer: Research and Development
>Library Systems Office
>Rm 386 Doe Library
>University of California
>Berkeley, CA 94720-6000
>510-643-9776
>