#344 Split off 'Dates and Times' to new TEI chapter

AMBER
closed-wont-fix
5
2012-06-03
2012-02-21
James Cummings
No

The reason that 'Names, Dates, People, and Places' are all one chapter is one of historical legacy. The names/people/places sections have so developed as to swamp the minor discussion of dates.

I propose that we:
a) split off Dates http://www.tei-c.org/release/doc/tei-p5-doc/en/html/ND.html#NDDATE to its own separate small chapter.
b) move the creation of att.datable.custom att.datable.iso att.duration.iso there instead of ND
c) rename the chapter 'NP' instead of 'ND'
d) reference the new chapter from http://www.tei-c.org/release/doc/tei-p5-doc/en/html/CO.html#CONADA
e) expand new chapter to give more examples of @calendar and dating edge-cases
f) consider merging timeline information into new chapter and/or referencing from chapter what is at http://www.tei-c.org/release/doc/tei-p5-doc/en/html/SA.html#SASYMP

I recognise there are many reasons for not doing this (persistent URLs, etc.) But I think separating out date/time related markup into one place might make more sense. Arguably it might make more sense to undertake this simultaneously to doing the same kind of things with sections in other chapters.

Just thinking out loud.

-James

Discussion

  • Martin Holmes
    Martin Holmes
    2012-02-21

    Good idea -- +1 from me. It would serve as a model for breaking up some of the other chapters too. Chapter 11 (Representation of Primary Sources) is pretty big these days. And Chapter 3 (Elements Available in All TEI Documents) is a grab-bag running to over 80 pages.

     
  • Lou Burnard
    Lou Burnard
    2012-03-18

    This seems fairly pointless to me. It would be different if dating info was scattered all over the place, but it isn't. Or are you suggesting that the datable attributes should be in their own *module* ?
    That's quite a different proposition, for which a more plausible argument might be made. However, idoes it make sense to fragment just a single case like this? Wouldn't it be better (as I think Martin is implying) to come up with a list of several such modules we'd like to see separated out?

     
  • Laurent Romary
    Laurent Romary
    2012-03-18

    I would concur with Martin here and would happily follow the discussions related to making focussed modules (bib? ;-)) in the future.

     
  • Martin Holmes
    Martin Holmes
    2012-03-18

    I thought James was talking about the CHAPTER in the Guidelines, not modules in the schema system. I think the point he was trying to make (and the point I agreed with) was that there could be more chapters which are shorter, each dealing with a more discrete topic, making the Guidelines more readable. I think that's definitely an issue -- think how few of us actually got all the way through the chapter 11 proofing last year -- and I think it would make the Guidelines as a whole much more user-friendly (especially for non-native speakers) if it were presented in shorter chunks.

    The issue of schema modules is completely different.

     
  • BODARD Gabriel
    BODARD Gabriel
    2012-03-20

    I think this is a very good idea. I also think coming up with proposals for other chapters/sections that would benefit from similar treatment would be very valuable, but I don't think the latter should be a blocker for the former. (In fact, turning it into such a big job would almost certainly become a blocker in its own right...) The only requirement is to have thought it through enough that later changes don't impact in new and disruptive ways on the new chapter.

     
  • James Cummings
    James Cummings
    2012-04-15

    If we just do this to the chapters, and not the underlying modules, then we would be making a conscious decision to decouple the relationship between modules and chapters.

    I was, in fact, proposing both splitting the prose chapter in to two _and_ the underlying module. In answer to Lou's criticism, yes, I would expect this new module to have all the non-trivial dating attributes, things like timeline, etc. in it. Basically any date-time related things except the most basic <date>. But just splitting the output is one possibility.

     
  • James Cummings
    James Cummings
    2012-06-03

    It was decided in Ann Arbor face to face meeting 2012 that module fragmentation such as this should be actively reviewed as part of a roadmap to P6 but not as part of P5 development.

     
  • James Cummings
    James Cummings
    2012-06-03

    • assigned_to: nobody --> jcummings
    • status: open --> closed-wont-fix