On Fri, 16 Mar 2001 17:37:32 -0800, Bill Frantz wrote:
>><!-- A person has a name and one or more roles -->
>><!ELEMENT Person (Name, Role*)>
>>
>><!-- The people that worked on this survey. If a Personnel element is
>>present, there must
>> be one or more Persons defined -->
>><!ELEMENT Personnel (Person+)>
>I have some survey data lying around which lists names, but not roles. In
>addition, in most surveys I have worked on, the roles shift based on
>circumstances. For example, the time when the sketcher felt ill and handed
>the book off to an instrument person. Or more commonly, when an instrument
>person was fooling with a light and a tape person or sketcher grabbed the
>instruments for a shot.
Valid point. That _could_ be done by adding that person with the role or adding a comment at that station. That is if it's
deemed something that needs to be documented of course. On the surveys I've done, I only record the person doing the
majority of the job. A one or two exception is not recorded. Is that your experience, too?
>For the unknown role, perhaps an empty role is the correct solution making
>it zero or more roles. I have never seen these other switches documented
>in the book, so I guess CaveXML doesn't have to represent them. :-)
It currently is zero or more. To make it:
- one or more, use '+';
- zero or more, use '*';
- zero or one, use '?'.
FWIW, there is also a CDATA role available. If the list does not fully cover what you want to document, you can make one
up. However, the point of using a defined list is to allow managers (ie government people) to query the data for information.
With defined roles, it makes it easier to limit the possibilities in a search.
Thanks for the feedback.
- Richard Knapp
_________________________________________________________
Do You Yahoo!?
Get your free @yahoo.com address at http://mail.yahoo.com
This archive was generated by hypermail 2b30 : Mon Apr 02 2001 - 18:00:01 CEST