From: Tatu S. <cow...@ya...> - 2007-01-27 05:22:13
|
--- Boris Kolpackov <bo...@co...> wrote: > Hi Jimmy, > > Jimmy Zhang <cra...@co...> writes: > > > do you see this as an essential feature of XSD? > > Looking at the wide use of default attribute values > in public > schemas we can conclude that the users have decided > it is an > essential feature ;-). I personally also think that Actually, I'm not sure that is conclusive proof -- what matters more is how many actual document instances make use of such features. However, it is interesting finding that so many schemas do make use of this feature. Personally I find it depressing... since: > it is a > handy feature (just as default argument values in > modern > programming languages). It may be just me, having to write parsers, but I think inclusion of attribute defaulting in XML Schema was one of its biggest blunders. It has nothing to do with either validation or data typing -- such business-logic definitions belong to another layer, not to static-by-nature structural (and coarse data typing) constraints model as schemas. And as to programming languages, some have it (c++, languages with named parameters), others not (java, and most scripting langauges). I prefer not to have them, as kind of syntactic sugar I can live without; but at least it's less fragile as defaulting actual data in documents. I can of course see practical convenience value, but I would lump it together with similar features like xml general entities or C preprocessor macros. Anynway, to each his own, -+ Tatu +- ____________________________________________________________________________________ Food fight? Enjoy some healthy debate in the Yahoo! Answers Food & Drink Q&A. http://answers.yahoo.com/dir/?link=list&sid=396545367 |