RE: Enlightenment via avoiding the T-word

[Nicolas LEHUEN <nicolas.lehuen@ubicco.com>]

>Nicolas LEHUEN wrote:
>> Nope. I can forge any example where a carefully defined 
>global element name
>> within a particular namespace would require context inspection to
>> disambiguate its semantic.  If I give you a "person-name" 
>element, do you
>> know what to do with it immediatly ? No, you'll ask me for 
>the context.
>> Okay, the context is a bill. This does not give you any 
>information. Another
>> part of the context, is the "buyer". This gives you a lot 
>more information.
>
>I think that "global semantics" might mean something less here than you
>imagine. As an example, consider <Price>. What is meant by global
>semantics in this context is only what a <Price> element is -- that is,
>it is a price in money of an object, and it may always be in a
>particular currency or it may have an attribute denoting the currency.
>This is very different from what the price applies to, which is
>obviously context dependent.
>
>These kinds kinds of semantics are quite wide-spread and allow 
>you to do
>basic processing with context-insensitive routines before passing the
>result on to a context-sensitive routine.

See one of my recent posts. If you consider that you are ready to globally
define the semantic of the name "price", then go ahead. You'll get the
benefits of being able to globally process "price" elements in a
context-insensitive way, while still keeping the option of processing it in
a context-dependent way.

Maybe it is possible to define a global semantic for "price" (if you are in
a financial context ; in a school context it might mean something totally
different). The document context will not change this semantic, it will just
apply it to a particular object.

But I think the task is much more difficult for words like "name",
"location", "balance", etc. (as I'm French I can give you more example in
French than in English, but I'm sure there are as many in the two
languages). Does it means that we should not use such generic terms ?

I think not. Names are too precious things (when I'm designing or
programming, I find that one of the most difficult task is giving objects
the Right Name) to put some apart. I think that we should benefit from the
fact that XML enables us to build documents with rich contexts. Thus, let's
use "name", "location", "balance", etc. without defining those names
globally, but just as local element names, since a context is required to
refine their semantic until they are precise enough.

Regards,
Nicolas