[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]
Subject: RE: [oasis-member-discuss] RE: RDDL use in ASIS
Thanks for all your comments on RDDL, Chris; I think I agree
with all of them, and personally have urged the use of
RDDL (experimental or not) for a long time. I can't think
of (m)any reason(s) one would:
* publish an HTTP scheme URI namespace that intentionally
dereferences to a 404
* create an HTTP scheme URI namespace that dereferences
to a specific instance [particular version] of a
versioned schema at a version-specific URI
* design and deploy ad hoc namespace document formats
when RDDL provides a reasonable candidate for current
use
The thread on RDDL [1] in connection with the ASIS review
motivates me to be more attentive to news about RDDL,
especially about the experimental uses that are now
in wide evidence. I have a document subsection here:
RDDL Newsbits
http://xml.coverpages.org/rddl.html#RDDL-newsbits
Examples
http://xml.coverpages.org/rddl.html#examples
Please send me references, and I'll try to provide
citations.
It's a bit disconcerting to discover that a lot of
people simply don't know about RDDL, and that some
are unclear on the concept of a namespace document.
Here are two key references:
Associating Resources with Namespaces
Draft TAG Finding 13 December 2005
http://www.w3.org/2001/tag/doc/nsDocuments/
Namespace documents
http://www.w3.org/TR/webarch/#namespace-document
In another posting (to follow), I want to ask the readers
for ideas about optimum designs for a file system which
[transparently exposes paths + filenames as URIs and]
clearly separates the notion of a namespace URI from other
related URIs [2], and manages user expectations with respect
to server behavior when a namespace URI corresponds to
the URI of a directory which contains related URI-addressable
resources.
Robin Cover
[1] Some postings on RDDL (ASIS review)
http://lists.oasis-open.org/archives/oasis-member-discuss/200602/msg00001.html
Chris Ferris
http://lists.oasis-open.org/archives/oasis-member-discuss/200602/msg00007.html
Chris Ferris
http://lists.oasis-open.org/archives/oasis-member-discuss/200602/msg00010.html
Marc Goodner
http://lists.oasis-open.org/archives/oasis-member-discuss/200602/msg00012.html
Bill Cox
http://lists.oasis-open.org/archives/oasis-member-discuss/200602/msg00017.html
Bill Cox
http://lists.oasis-open.org/archives/oasis-member-discuss/200602/msg00018.html
Paul Fremantle
http://lists.oasis-open.org/archives/oasis-member-discuss/200602/msg00019.html
Chris Ferris
http://lists.oasis-open.org/archives/oasis-member-discuss/200602/msg00022.html
Chris Ferris
http://lists.oasis-open.org/archives/oasis-member-discuss/200602/msg00024.html
Marc Goodner
http://lists.oasis-open.org/archives/oasis-member-discuss/200602/msg00025.html
Bill Cox
http://lists.oasis-open.org/archives/oasis-member-discuss/200602/msg00027.html
Gabe Wachob
http://lists.oasis-open.org/archives/oasis-member-discuss/200602/msg00028.html
Hal Lockhart
http://lists.oasis-open.org/archives/oasis-member-discuss/200602/msg00029.html
Chris Ferris
http://lists.oasis-open.org/archives/oasis-member-discuss/200602/msg00031.html
Marc Goodner
http://lists.oasis-open.org/archives/oasis-member-discuss/200602/msg00032.html
OASIS Staff
http://lists.oasis-open.org/archives/oasis-member-discuss/200603/msg00000.html
Chris Ferris
http://lists.oasis-open.org/archives/oasis-member-discuss/200603/msg00002.html
Norman Walsh
http://lists.oasis-open.org/archives/oasis-member-discuss/200603/msg00009.html
Hal Lockhart
http://lists.oasis-open.org/archives/oasis-member-discuss/200603/msg00010.html
Chris Ferris
http://lists.oasis-open.org/archives/oasis-member-discuss/200603/msg00011.html
Bill Cox
http://lists.oasis-open.org/archives/oasis-member-discuss/200603/msg00012.html
Bill Cox
http://lists.oasis-open.org/archives/oasis-member-discuss/200603/msg00014.html
Chris Ferris
http://lists.oasis-open.org/archives/oasis-member-discuss/200603/msg00016.html
Hal Lockhart
http://lists.oasis-open.org/archives/oasis-member-discuss/200603/msg00018.html
Chris Ferris
[2] Chris Ferris on separate concerns: namespace URI vs
URI(s) of resource(s) describing a namespace URI
http://lists.oasis-open.org/archives/oasis-member-discuss/200602/msg00022.html
[someone said:]
> Original motivation, I think, was to allow a single schema
> namespace URI without requiring a change for each developement
> version of a spec. This is existing practices in a number of TCs.
[Chris says:]
IMO, this is a classic example of where it makes sense to
separate out the namespace URI from the URI of the resource(s)
that it represents and why it is considered good practice to
have a namespace document (RDDL) that describes the namespace
as opposed to having the schema or WSDL document at the end
of the namespace URI. Please DO keep in mind that there
is not necessarily a 1:1 correspondence between a namespace
and a physical schema document.
However, IMO, it also points out the folly of forcing the
metadata to be expressed in the filename.
IMO, the practice adopted by the W3C for providing:
This version: [dated URI]
Latest version: [persistent URI]
Previous version: [dated URI]
is something that OASIS should long ago have adopted as well.
If I embed the metadata within the artifact, yet make it
available via some persistent URI (regardless of what its
current status is), then it means that in such cases where
ONLY the metadata has changed, that the resource's URI need
not change. This is "A Good Thing(tm)".
On Sat, 4 Mar 2006, Christopher B Ferris wrote:
> Hal,
>
> Please see my CIL below.
>
> Cheers,
>
> Christopher Ferris
> STSM, Software Group Standards Strategy
> email: chrisfer@us.ibm.com
> blog: http://www.ibm.com/developerworks/blogs/dw_blog.jspa?blog=440
> phone: +1 508 377 9295
>
> "Hal Lockhart" <hlockhar@bea.com> wrote on 03/03/2006 10:08:48 AM:
>
> > Thought it was obvious.
> >
> > 1. We never documented why we dropped RDDL. It was not because we
> > didn?t like the concept of RDDL.
>
> That wasn't the impression I got, but I am glad to hear that.
>
> > 2. Because RDDL consists of a bunch of different documents scattered
> > around on various web sites with no clear home, it could disappear
> tomorrow.
>
> So what? There are two (or more) versions of lots of specs/standards out
> there.
> RSS0.9, RSS1.0, RSS2.0 (which actually predates 1.0), Atom... That does
> not
> seem to have had a detrimental effect on the adoption of syndication feeds
> as a general concept, and in many cases, there are tools that can
> effectively
> consume and produce multiple versions (e.g. SharpReader, RSSBandit, etc.
> etc.)
>
> Only one of these technologies has any standards status (Atom) and that is
> a
> relatively recent development. There were plenty of tools that implemented
> v0.3 well before the IETF ratified 1.0 as an standards track RFC.
>
> > 3. It is not only unclear what version of RDDL to cite, it is
> > unclear even what version represents the best and latest. The tree
> > appears to have forked.
>
> All of the versions have merit, IMO. None, to my knowledge, require any
> tooling upgrades to support. Why should it matter, then, which is chosen?
> Pick the one you like the best. Pick the one that others have used.
> Flip a coin. Whatever.
>
> However, IMO, you threw the baby out with the bath-water
> in this case and provided no rationale as to why you did so.
>
> Did anyone contact Jonathan Borden or Tim Bray and ask them? Did
> anyone notice that the specs/schema published at schemas.xmlsoap.org
> now use RDDL and ask Microsoft which version they used (although it
> is obvious by doing a "view as source")?
>
> There was active discussion on the WS-RX TC about adopting RDDL
> for its "namespace documents", you might have asked the TC
> administrator what they were doing. I know for a fact that at least
> a couple of members of the TAB are also members of WS-RX TC.
>
> > 4. Although every version is labeled as a non-final draft, no work
> > has been done for two years. This suggests that the spec is an
> > orphan and not something OASIS should cite either normatively or as
> > a best practice.
>
> No work has been done on the RSS0.9 spec for years and years either.
> I don't see how that makes it any less valuable.
>
> Just because something is a standard-with-a-capital-S, doesn't make
> it good. Just because something isn't, doesn't make it worthless.
>
> In the context of the use for which we have for the metadata,
> IMO, any of the versions would satisfy OASIS's requirements.
>
> >
> > Hal
> >
> >
> > From: Christopher B Ferris [mailto:chrisfer@us.ibm.com]
> > Sent: Thursday, March 02, 2006 9:25 PM
> > To: Hal Lockhart
> > Cc: James Bryce Clark; Norman Walsh; oasis-member-discuss@lists.
> > oasis-open.org; William Cox
> > Subject: [oasis-member-discuss] RE: RDDL use in ASIS
> >
> >
> > And the point being?
> >
> > Christopher Ferris
> > STSM, Software Group Standards Strategy
> > email: chrisfer@us.ibm.com
> > blog: http://www.ibm.com/developerworks/blogs/dw_blog.jspa?blog=440
> > phone: +1 508 377 9295
> >
> > "Hal Lockhart" <hlockhar@bea.com> wrote on 03/02/2006 05:12:51 PM:
> >
> > > > | What do others think? As I said, there was a lot of pushback on
> > > RDDL.
> > > >
> > > > I don't recall reading the pushback on RDDL, but "and preference to
> > > > have an index.html or one of the other default HTML pages" isn't
> > > > related. RDDL is a mechanism for placing metadata in HTML.
> > >
> > >
> > > Ok, for the record, last summer Bill and I tried to figure out the
> > > standards pedigree of RDDL so we could cite it.
> > >
> > > No sign of it at W3C, not even published as a Note.
> > >
> > > At www.rddl.org, there is a spec dated Feb 18, 2002, no version
> > > specified. I guess this is version 1.
> > >
> > > At http://www.openhealth.org/RDDL/20040118/rddl-20040118.html there is
> a
> > > spec dated Jan 18, 2004 marked as version 2.0. It describes itself as
> "a
> > > draft".
> > >
> > > At http://www.tbray.org/tag/rddl/rddl3.html there is a document dated
> > > June 1, 2003, with no version. Not sure what version this is. Perhaps
> > > Tim's private version? If it is RDDL 3, it is older than RDDL 2.0.
> > >
> > > All of these contain the sentence "This document has no official
> > > standing and has not been considered nor approved by any
> organization."
> > >
> > > There are also a number of articles, implementations and even a
> > > Wikipedia article (which points to the 2002 version). The 2004 version
> > > says "While this document has no official standing, it is the
> intention
> > > of the TAG to seek guidance from the W3C membership and the larger
> > > community on the question of whether and how to progress this document
> > > and the use of RDDL." As far as I can see there has been no work done
> on
> > > RDDL in 2 years.
> > >
> > > Will the real RDDL please stand up? If this is as great stuff as you
> all
> > > say it is, can't somebody put in enough time to push it thru an OASIS
> TC
> > > or publish it as a TAG Finding? If I put a normative reference to
> > > something with a pedigree like this in an OASIS Committee Spec and
> > > submitted it for member approval, I would end up with a bunch of
> arrows
> > > sticking out of me.
> > >
> > > Hal
> > >
> _______________________________________________________________________
> > > Notice: This email message, together with any attachments, may
> contain
> > > information of BEA Systems, Inc., its subsidiaries and affiliated
> > > entities, that may be confidential, proprietary, copyrighted and/or
> > > legally privileged, and is intended solely for the use of the
> individual
> > > or entity named in this message. If you are not the intended
> recipient,
> > > and have received this message in error, please immediately return
> this
> > > by email and then delete it.
> > _______________________________________________________________________
> > Notice: This email message, together with any attachments, may contain
> > information of BEA Systems, Inc., its subsidiaries and affiliated
> > entities, that may be confidential, proprietary, copyrighted and/or
> > legally privileged, and is intended solely for the use of the individual
> > or entity named in this message. If you are not the intended recipient,
> > and have received this message in error, please immediately return this
> > by email and then delete it.
>
--
[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]