Re: [xml-dev] A question of necessity

[James Bryce Clark OASIS <jamie.clark@oasis-open.org>]

Personal opinions only:  
*  There certainly is such a thing as namespaces.
*   That device was overused and misused in some XML practices, but those outcomes do not disprove its utility in some other cases.
*   As it (maybe) becomes more clear that the sweet spot of XML is in certain use cases (e.g. complex content), but not others (e.g. early-2000s middleware messages), perhaps the overcomplexity bad rap will dissipate.

Cordially, JBC

Sent from my mobile

On Apr 21, 2016, at 9:02 PM, Arjun Ray <arjun.ray@verizon.net> wrote:

On Thu, 21 Apr 2016 13:32:21 +1000, Alexander Johannesen
<alexander.johannesen@gmail.com> wrote:

| Arjun Ray <arjun.ray@verizon.net> wrote:
| >
| > Materially implied; sine qua non.

| Unfortunately that doesn't answer the question.

You may have misread mine.

| That's just defining what necessary means technically, not what it
| means in our context.

I don't know about your context; in mine, I intended the plain meaning
of the word.

| So what I'm really after is what you mean with necessary in the
| context of namespaces in XML.

I was talking about the syntactic device of colonified names.  Permit
me to requote a passage, which you quoted too:

:> In other words, can you formulate a use case where the syntactic 
:> device of colonified names - and the processing burden that comes 
:> with it - is necessary?

The _concept_ of "XML Namespaces" - assuming there is such a thing -
would be a wider issue, and it could be that you have equated notion
and implementation. 

| Anything I can think of, I can do it differently without namespaces.

In what ways would treating colonified names as atomic names have made
things difficult for you?

| But the same applies to attributes. To entities. To whitespace nodes.
| To Unicode support and encodings. To CDATA. Lack of empty-element tags.
| To angle brackets. And on and on.

Sorry, I have no idea what you're talking about.

| An attribute isn't as good a solution as a namespace to keep mixed
| content models sane. They're awesome for separation of domains and
| concerns. 

Please define, or at least clarify, mixed content models and sanity
thereof. 

| In our case, without it makes for less flexible are more verbose code. 
| Is that necessary enough, though?

Specifics or a concrete example would help.

I suspect, however, that no one would have been any worse off had
colonified names never been foisted on XML.  All the plausible and
semi-plausible "justifications" advanced by the proponents bit the
dust on the old WG/SIG lists. So the goalposts shifted and we got some
bombast about "universal names" somehow being "necessary" (which at
that time really meant a non-negotiable "requirement" imposed on the
XML Activity, which is ultimately how these beasties came to be).

That is, the necessity of "universal names" was simply laid down by
fiat.  I still don't think that was good enough.  Here's someone who
agrees:

http://www.ibm.com/developerworks/library/x-abolns/

_______________________________________________________________________

XML-DEV is a publicly archived, unmoderated list hosted by OASIS
to support XML implementation and development. To minimize
spam in the archives, you must subscribe before posting.

[Un]Subscribe/change address: http://www.oasis-open.org/mlmanage/
Or unsubscribe: xml-dev-unsubscribe@lists.xml.org
subscribe: xml-dev-subscribe@lists.xml.org
List archive: http://lists.xml.org/archives/xml-dev/
List Guidelines: http://www.oasis-open.org/maillists/guidelines.php