[Home] [By Thread] [By Date] [Recent Entries]


Agree.  My first inkling that this 
was going off track was the incessant beating on DTDs.  One 
reputable and experienced contributor objected and said, "I 
thought we were doing DTDs++" and instead, it became "OOP--" 
with some overbuilt bits for RDBs, yet DTDs had been quite 
serviceable for some years to the majority of markup designs. 
On the other hand, I did call for it being used until we 
had a better option.  We have one now:  RelaxNG, but as 
it usually happens, we also now have legacy.

It would have been a good idea to note up front that there 
are multiple kinds of databases and they are not united by 
a common schema language, and that forcing one to do that 
would be an unnatural act.  In markup design, it is usually 
a bad idea to attempt to smooth over organic irregularities. 
That is why mixed models are provided despite their theoretical 
inelegance.   XML succeeded initially by removing features some 
thought not useful for the majority of applications, but the 
minority continues to exist.  Designers have to make the 
tradeoffs and that is always a political process.  No free lunch.

Any size that fits all is always uglier than any thing to be fit.

len


From: Robin Berjon [mailto:robin.berjon@e...]

There's probably more, but that's all I'm thinking of right now. I think 
it's an interesting experience to do. But then I guess I should just 
quit whining and get a job that doesn't require me to deal with that 
specific gorgon (so long as we can keep it from polluting other specs at 
random) :)

Site Map | Privacy Policy | Terms of Use | Trademarks
Free Stylus Studio XML Training:
W3C Member