Cart
[Home] [By Thread] [By Date] [Recent Entries]
Frank Boumphrey writes:
> Writing a new DTD is (should be) very little different from any other
> research activity. There is always the temptation to follow established
> guidelines, but such a process will only propagate any faults in the
> existing corpus of work. (If any exist). Researchers should be familiar with
> previous work, but should not let that work inform the detail of their new
> efforts.
What is the first thing we get a research student to do? We tell them
to do a literature search, and prove that their proposed subject is
indeed new, and important. If you find the subject is already covered,
you find a new research project
> To take an innovative approach it is necessary to take a fresh look at the
> base material.
yes. but where is the justification for an "innovative approach"? what
is _innovative_ about Gutenberg?
> 1. Write a set of requirements for a DTD
> 2. Investigate possible component sets for the material in question.
> 3. Model a provisional DTD
> 4. Compare the provisional DTD with established DTD's
> 5. As far as possible reconcile the new DTD with the established DTD's
I think you forgot
1 (a). If these requirements exactly match the description of an
existing DTD, use it....
Sebastian
***************************************************************************
This is xml-dev, the mailing list for XML developers.
To unsubscribe, mailto:majordomo@x...&BODY=unsubscribe%20xml-dev
List archives are available at http://xml.org/archives/xml-dev/
***************************************************************************
|
