[
Lists Home |
Date Index |
Thread Index
]
Precisely. Schema/DTD evolution is hard across
the scale of the number of implementors who can
make changes to the code that implements it.
Maintaining the schema itself is easy. So,
again, the issues of evolution will be more
intense in the system libraries. We spend
a lot more time sorting out dll hell than
bad schema design.
What we can live without are the pronouncements
from the Beltway saying things like, avoid the
use of attributes. That sort of policy driven
schema design predicated on superstition is
harmful. Now the schema design can hurt the
software design because the social hierarchy
actively supports bad decisions about fundamentals
of markup.
len
-----Original Message-----
From: Uche Ogbuji [mailto:uche.ogbuji@fourthought.com]
Rick says:
> There are many ways that a schema may evolve, and different mechanisms
> may be suitable for each way. For example, change over time,
> change between locales, change between processes in pipelines,
> change depending on what the user is interested in, minor changes/major
> changes.
<AOL>Very well said</AOL>
I'm surprised to come across the idea of schema evolution as a monolithic
problem. I am doubly surprised to see RSS held up as an example of the
difficulty of schema evolution. RSS is but example of politics and
carelessness. All social science, and not a drop of computer science to spare.
|
