yak
[Top] [All Lists]

[yak@collab] Re: [tools-yak@collab] Re: Are XML Namespaces necessary?

To: tools-yak@xxxxxxxxxxxxxxxxxxx
Cc: yak@xxxxxxxxxxxxxxxxxxx
From: "Peter P. Jones" <ppj@xxxxxxxxxxxxxxxxxxxxx>
Date: Tue, 16 Mar 2004 12:18:53 -0000
Message-id: <4056F0AD.7653.55FA35@localhost>
Hi,
Before I answer what you've asked I would just like to invoke the 
IANAL (I am not a lawyer) clause, because I think the answers to 
these questions might turn on a combination of the processes by which 
XML documents are handled and the (new?) legal twists that XML 
Namespaces might give rise to in that area.
Anyone who can inject better knowledge into this discussion on the 
question of intellectual property (IP) is welcome to do so.    (01)

On 15 Mar 2004 at 21:45, Murray Altheim wrote:    (02)

[...]
> While I'm no fan of the W3C's Recommendation, exactly how do you
> consider that XML Namespaces has a moral consequence?
I am thinking it has opened the path to an enclosure of a vital commons.
I don't think XML Namespaces merely disambiguate names, I think they 
simultaneously assert property rights (as you have mentioned below) 
in scenarios where dynamic aggregation of chunks of content (possibly 
from providers who are not recognised in advance) is at a premium.
Is that a bad thing?
It depends on how you see the enclosure of this commons playing out. 
At the moment it looks like it's following a very traditional pattern 
in some respects.
Three issues are vital, as I see it:
1) International law covering public interest exceptions on the 
withholding of information from the public domain.
2) International law covering the filtering/editing of information 
and its relationship to public interest.
3) What it means to nest information from one namespace within 
another (or not) in the tree of markup elements, and who gets to play 
at the top of the heap.
In effect, where does interoperability go when, once past what few 
social contracts exist, the zone beneath it is a big free-for-all?
Given a zone like that, what happens when payment for viewing content 
under a particular namespace is demanded?    (03)

> If anything, I
> kinda like the fact that the ability to declare a namespace using a
> non-legislated URI means that anyone can do so.
It looks neat doesn't it, and I'm as guilty as anyone of thinking 
that it is a quick and easy way of protecting an interest. I'm not so 
sure it protects wider public interest though.    (04)

> They could have
> required that the namespaces were registered with an authority, as is
> the case with URNs. Now there's a place to be pissed off. I.e., if you
> or I want to create a URN namespace, we have to apply to an
> organization for a top-level domain, and they can turn you down.
They could have. But they didn't, and I'm just paranoid enough to ask 
why. The problem with an authority is not just selectivity, but lack 
of speed and greater fixity. Bureaucracy usually only attaches to 
commerce when there is a greater profit to be had out of doing so.    (05)

> 
> Now, XML Namespaces (as I mentioned in my note) have their syntactic
> and interpretive problems, but they at least allow for a level of
> egalitarianism for disambiguated names, so that you and I can have our
> own HTMLs and DocBooks and whatever we like, merely by declaring a
> namespace using a domain we control. So for me, I created an XML
> markup language called XNode whose XML Namespace is mine. There's
> probably even a legal protection available for that. The W3C is only
> responsible for creating the spec, they don't have anything to say
> about namespaces that you or I might create. There's no sense of a
> "blessed" namespace or anything, e.g., my namespace is just as good as
> IBM's in terms of XML Namespaces. (This is not the case with URNs.)    (06)

Instead I would like to ask the question: If there are other ways of 
achieving disambiguation that would not break the commons, why are 
these methods not more prominent?
For example, control over how W3C specs could fit together could have 
been achieved for detecting document conformance - the XML Namespace 
mechanism was not essential to that, nor does it aid it particularly.
And the XML Namespace, as you right say, opens up the labelling of 
content with a namespace to anyone, *in principle*.
Whether such namespaces would be allowed to play freely in the pool 
in a substantive sense is another matter.    (07)

Those are my concerns. Perhaps they are too cynical, and perhaps not.     (08)

-- 
Peter
P.S. I've sent this to yak too as it really crosses more into that 
zone - but do the two lists have all the same participants anyway?    (09)

-- 
This message is archived at:    (010)

http://collab.blueoxen.net/forums/cgi-bin/mesg.cgi?a=yak&i=4056F0AD.7653.55FA35@localhost    (011)
<Prev in Thread] Current Thread [Next in Thread>