Re: [xml-dev] Lesson Learned: Use namespaces for both markup anddata

[Michael Ludwig <milu71@gmx.de>]

Glidden, Douglass A schrieb am 11.08.2009 um 17:44:19 (-0400):
> I have to say that, all in all, I don't understand why working with
> namespaces is supposedly so painful, so I've stayed out of this
> discussion; however, as an "application developer" who works with
> quite a bit of XML data, I have to disagree with you here.

Working with namespaces presents three or more burdens of varying
importance:

(1) It may be painful for novice users, until they figure it out.

(2) It frequently imposes a certain tedium on the reader and sometimes
traps even experienced users in silly errors because the established
convention of choosing unnecessarily complicated URIs or URLs to
identify namespaces is so horrible:

http://markmail.org/message/3yq7mei2bzbnzmbs

(3) It is a problem in the data model and in implementations because
of the context dependency that the mere possibilty of namespace
declarations in mid-markup introduces:

http://markmail.org/message/utae2qaibwkky7gr
http://markmail.org/message/mciccvc4wtqwforo

Well, the novice will have to do some grokking as per burden (1).

As an XML user past the novice stage, I find the current system quite
okay, and prefixes even good as they tell me what namespace an element
belongs to. We should just start using friendlier namespace identifiers.
Then, burden (2) would be alleviated.

But it seems there is more to prefixes than meets the eye, they cause a
lot of complexity in the data model and limit your ability to transplant
nodes from one location to another. (3) Things could be made to work in
a simpler fashion. A hypothetical better system might benefit both users
and implementors.

-- 
Michael Ludwig