[
Lists Home |
Date Index |
Thread Index
]
Dare Obasanjo wrote:
>
> It has no discernable bias if you're not versed in the subject
> matter.
He couldn't disagree with you unless he is biased. I couldn't disagree
with you about his bias unless I am lacking in knowledge about XML
Schema.
I've been doing "XML" since around 1996 when it was (more or less)
called "best practices for SGML". I wrote one of the first proposals for
how type inheritance might work (or, I concluded, finally, probably
would NOT WORK). When he says type inheritance is a bad idea, that
resonates with me for a variety of reasons that would take many hours to
go into. When he says that "first-class" complex types are not
particularly interesting outside of inheritance, that also resonates
with me.
>...
> "To be precise, you can always write it without understanding complex
> types, but unfortunately you have to type <complexType> elements."
Makes sense to me!
> followed by
>
> "Just consider a <complexType> as something you have to write as a
> sole child of the <element> element. That is, you write element declarations as follows:
Ditto.
> ...
> "Inheritance is a complex type's only advantage, but you really
> don't want to use it."
Yep! After a year of off-and-on research I concluded that trying to
import OO-style inheritance to schemas was a bad idea.
>...
> Loosely translated, don't use this feature because 1 month after
> the technology became a W3C recommendation implementations are
> still buggy and I believe developers are too lazy and too dumb
> to implement it anyway.
There are many features of various technologies that developers never
get around to implementing properly. Perhaps the long term proves him
wrong about this one.
> The article is a FUD fest. Every genuine concern is exaggerated
> while phantom almost nonsensical concerns are constructed in an
> attempt to discredit the technology.
I choose to interpret as a guide to avoiding the parts of the technology
that will cause frustration.
--
Paul Prescod
|
