Bullard and Megginson Musing over Namespaces [and (non-)Centralized Namespace Registration]
Date: Wed, 15 Dec 1999 19:01:17 -0600 From: Len Bullard <cbullard@hiwaay.net> To: David Megginson <david@megginson.com> Subject: Re: Musing over Namespaces David Megginson wrote: > > It's messy, but it's the only standards path that really seems to > work. At least with Namespaces we can remove 50% of the messiness > (there's no chance of confusing different party's extensions) on the > way to standards Nirvana. I think not. The standards path is complete once the means to create and identify the namespace is possible. What now needs to happen is for the technologists to create means by which company registries of namesspaces can be accessed and negotiated with such that for any contract, the ROA namespaces can be declared. OASIS and BIZTALK may be good libraries, may provide examples, but trying even at the UN level to standardize everyone's local names or names of exchange is a foolish if goodhearted errand. It is not the way to do it right. It is not the path to success. It is the way that standards wonks think, but I doubt it works as well as registry negotiation. If Microsoft wants to be a good partner, provide technology and some startup templates, then get out of the way. If OASIS wants to thrive, teach the means, do not arbitrate. Powertrips are not attractive. Education is seductive. Influence is more powerful than rules. > I'm going to write a paper with a catchy title on this topic some day, > so that I can make US$45M like Eric Raymond just did from "The > Cathedral and the Bazaar." "Green green it's green they say on the far side of the hill. Green green I'm going away to where the grass is greener still." len xml-dev: A list for W3C XML Developers. To post, mailto:xml-dev@ic.ac.uk Archived as: http://www.lists.ic.ac.uk/hypermail/xml-dev/ and on CD-ROM/ISBN 981-02-3594-1
Date: Wed, 15 Dec 1999 19:13:01 -0600 From: Len Bullard <cbullard@hiwaay.net> Cc: xml-dev@ic.ac.uk Subject: Re: Musing over Namespaces W. E. Perry wrote: > > Only the receiving > node defines the particular processing which it would like to initiate upon instantiation of > that element. And if in time the nature of that processing changes, that node's correspondents > will have no way to know that, nor to know how their <purchase> elements are now being > differently instantiated at that node. Yes! And the receiving node gets to change its mind when it needs to. It gets to renegotiate and it gets to advertise the interfaces by which it does both to configure the means and expressions of the transactions. len xml-dev: A list for W3C XML Developers. To post, mailto:xml-dev@ic.ac.uk Archived as: http://www.lists.ic.ac.uk/hypermail/xml-dev/ and on CD-ROM/ISBN 981-02-3594-1
Date: Wed, 15 Dec 1999 19:09:34 -0600
From: Len Bullard <cbullard@hiwaay.net>
To: Tim Bray <tbray@textuality.com>
Cc: "Clark C. Evans" <clark.evans@manhattanproject.com>,
David Megginson <david@megginson.com>, xml-dev@ic.ac.uk
Subject: Re: Musing over Namespaces
Tim Bray wrote:
>
> At 11:04 AM 12/14/99 -0500, Clark C. Evans wrote:
> >It'd still be nice to have a single database with
> >everyone's namespace definitions in one place though...
> >perhaps even a DTD to help describe them. I'm sure
> >there are organizations doing this... are there?
>
> Yes, lots. That's the problem. -Tim
No Tim, that's the answer. Common goals achieved by
common means, not common control. Put the definitions in
place by performance, not uberNamespace. It has to
breathe, my friends, like a good well performed piece
of music, it has to breathe.
The standards wonks really need to study management
and contract negotiation. OASIS and BIZTALK are
the absolute wrong way to do this. This means a
marvelous lucrative opportunity is sitting there being
ignored. Someone or some company will get rich
off this if they go to these companies and demonstrate
the right way to do this.
len
xml-dev: A list for W3C XML Developers. To post, mailto:xml-dev@ic.ac.uk
Archived as: http://www.lists.ic.ac.uk/hypermail/xml-dev/ and on CD-ROM/ISBN 981-02-3594-1
Date: Thu, 16 Dec 1999 16:57:46 -0500 (EST)
From: David Megginson <david@megginson.com>
To: xml-dev@ic.ac.uk
Subject: Re: Musing over Namespaces
Len Bullard writes:
> David Megginson wrote:
> >
> > It's messy, but it's the only standards path that really seems to
> > work. At least with Namespaces we can remove 50% of the messiness
> > (there's no chance of confusing different party's extensions) on the
> > way to standards Nirvana.
>
> I think not. The standards path is complete once the means to
> create and identify the namespace is possible. What now needs to
> happen is for the technologists to create means by which company
> registries of namesspaces can be accessed and negotiated with such
> that for any contract, the ROA namespaces can be declared.
I've noticed this assumption behind a lot of the Namespace-related
postings (not just Len's) -- I guess it would be a good thing to have
a global resolution mechanism of some kind, but I'm surprised that
people consider Namespaces incomplete without this.
After all, there's nothing similar for (say) Java or Perl package
names, and while having something like that would be convenient (and
CPAN is very nice as far as it goes), the absence of a global
resolution mechanism for figuring out what org.xml.sax or XML::Parser
means doesn't seem to inhibit useful work in Java or Perl.
Why is XML different? Is it just that we come from the SGML
background, where we consider structural validation to be part of a
document rather than a process applied to it, or is there some kind of
a fundamental difference between naming code and naming document
nodes that no one has articulated yet?
> > I'm going to write a paper with a catchy title on this topic some
> > day, so that I can make US$45M like Eric Raymond just did from
> > "The Cathedral and the Bazaar."
>
> "Green green it's green they say on the far side of the hill.
> Green green I'm going away to where the grass is greener still."
No, that's too long -- I need something shorter and catchier.
All the best,
David
David Megginson david@megginson.com
http://www.megginson.com/
Date: Thu, 16 Dec 1999 18:23:17 -0600 From: Len Bullard <cbullard@hiwaay.net> To: David Megginson <david@megginson.com> Cc: xml-dev@ic.ac.uk Subject: Re: Musing over Namespaces David Megginson wrote: > > I've noticed this assumption behind a lot of the Namespace-related > postings (not just Len's) -- I guess it would be a good thing to have > a global resolution mechanism of some kind, but I'm surprised that > people consider Namespaces incomplete without this. My point is not that that namespaces are incomplete but that they are complete. The next step is not necessarily to provide uber registration agnencies such as BizTalk or OASIS although nothing prevents such things. I think the better solution is local registries which the businesses create and administer. It assuredly takes longer for such things to be created, but by interfaces to such, the noise levels are encapsulated and localized. It is by process and definition of how to open and close and structure processes, what I once called views in the older days, that businesses execute and perform. It is necessary to tune these. What we have with wall-to-wall markup systems is the loose coupling by which such message based systems can both operate openly but control visibility. This is necessary because all business processes and instruments have a degree of slop which has to be there to cope with what the Boeing trainers called the unknown-unknowns. [...] len xml-dev: A list for W3C XML Developers. To post, mailto:xml-dev@ic.ac.uk Archived as: http://www.lists.ic.ac.uk/hypermail/xml-dev/ and on CD-ROM/ISBN 981-02-3594-1
Date: Fri, 17 Dec 1999 14:29:30 +0000 (GMT)
From: Dan Brickley <Daniel.Brickley@bristol.ac.uk>
To: xml-dev@ic.ac.uk
Subject: Re: Musing over Namespaces
Len Bullard <cbullard@hiwaay.net>
12/15/99 07:09 PM
To: Tim Bray <tbray@textuality.com>@SMTP@Exchange
Tim Bray wrote:
> At 11:04 AM 12/14/99 -0500, Clark C. Evans wrote:
> >It'd still be nice to have a single database with
> >everyone's namespace definitions in one place though...
> >perhaps even a DTD to help describe them. I'm sure
> >there are organizations doing this... are there?
>
> Yes, lots. That's the problem. -Tim
> No Tim, that's the answer. Common goals achieved by
> common means, not common control. Put the definitions in
I didn't take Tim to be arguing against decentralisation, but against
the multitude of companies/organisations who both promote the
centralised monolithic (meta)data registry approach, and
(coincidentally) attempt to promote themselves as providing that
service.
One of the supposed big wins of using the same syntax/model for our
meta-languages and our instance data is re-use, synergy.
Eg. if RDF and XML vocabularies are themselves described using RDF and
XML, then generic discovery/indexing/trust systems applicable to _all_
XML/RDF content should be equally applicable to schemas. Why then
promote centralised registries for all schemas? Surely there will be
search engines and 'trusted third parties' for schema data, as there
will be for other applications of XML and RDF. By defining schema
languages in instance syntax, we implicitly promote the idea that there
will be some big payoff for doing so (otherwise, lets stick with
DTDs). Some synergy that means generic tools will be applicable to
schemata. I find this impossible to reconcile with the
www.really-important-trusted-metaregistry.[com|org] approach that seems
popular in the industry. I'm banking on doing schema searches at the
mainstream search engines in 2-3 years time...
Dan
daniel.brickley@bristol.ac.uk
xml-dev: A list for W3C XML Developers. To post, mailto:xml-dev@ic.ac.uk
Archived as: http://www.lists.ic.ac.uk/hypermail/xml-dev/ and on CD-ROM/ISBN 981-02-3594-1
Prepared by Robin Cover for the The SGML/XML Web Page archive.
