OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.


Help: OASIS Mailing Lists Help | MarkMail Help

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index]
Re: [xml-dev] Lessons learned from the XML experiment

Michael Kay wrote:
"and they [namespaces] add myriad opportunities for doing things wrong."

You forgot an item in your list: they offer a unique possibility to get things right when things get really complicated.

For example when you do what I presently do: integrate schema information from 283 different schemas. Fortunately they use namespaces - 283 target namespaces - which enables me to keep everything clean without relying on document URIs and or any conventions, and without adding anything to existing structures (like marker attributes). I admit that I miss a lot of fun which I might have if there were no such thing as a target namespace (the shadow of namespaces).


Timothy W. Cook <tim@mlhim.org> schrieb am 10:48 Freitag, 15.November 2013:
Okay, I see a lot of 'complaining' about namespaces, even from people
that apparently do not use them. I have yet to see any actually
problem that the current implementation can cause.

An example or two would be very helpful for those (relative) newcomers
to using XML.

As far as the certainty and ambiguity argument goes.  Of course
everything has limits.  However, in healthcare information you cannot
afford ambiguity.  The lack of semantic interoperability in clinical
information is critical to cross application semantic continuity.
That is why controlled vocabularies like SNOMED-CT (and several
others) exist and must be used appropriately.

Why do I like XML technologies so much?  Because they are ubiquitous
in the computing world.  Previous efforts in multi-level modelling
based on a DSL have been in the works for over a decade but the slow
uptake is a circular problem with the lack of tools.

XML Schema provides the ability to very specifically define the syntax
and semantics of a clinical concept as the modeller intended and
transfer that information to the receiver, via the link in the data
instance, whether it is a human or a machine for further processing.
Specifically, XSD 1.1 with multiple substitution groups and asserts
provides a usable solution in a complex environment.

In the domain(s) that you work in you may leave it up to the receiver
to independently determine the meaning of some data that was recorded
20 years ago but in healthcare it is crucial to the health and safety
of the patient that the data is interpreted in the correct spatial,
temporal and ontological contexts in which it was recorded.

I do not have 15 - 20 years in XML.  But I do in healthcare IT.
Specifically in the challenges of semantic interoperability.  I fully
welcome XML experts in analyzing my work on this problem.  You can get
the latest version at http://launchpad.net/mlhim-specs


On Fri, Nov 15, 2013 at 2:37 AM, Kurt Cagle <kurt.cagle@gmail.com> wrote:
> On Thu, Nov 14, 2013 at 6:19 AM, Simon St.Laurent <simonstl@simonstl.com>
> wrote:
>> t's easy: I've learned over the years that people who believe in
>> certainty, especially those who believe that they can communicate certainty,
>> are dangerous.
>> Something goes deeply wrong when people assume that it is possible to know
>> things precisely, to name things precisely, and to communicate things
>> precisely.  (I'll grant that claims of precision are slightly less dangerous
>> than claims of accuracy.)
>> I sometimes call it naive positivism, but there are other philosophical
>> schools that lead to the same sad place.  Computers, of course, encourage
>> such delusions, but that is largely because they know so little about the
>> world.
> This is a philosophy to live by. RDF works when you keep to the open world
> assumption, works even better when you treat all assertions as provisional
> until proven otherwise. I've been modeling "assumption" frameworks, rather
> than assertion ones, precisely (pardon the pun) to provide a way of
> expressing risk, uncertainty and ambiguity.
>> Uche is extra-right about URIs, of course, a corner that painfully
>> demonstrates these limitations again and again.
> I've long felt that namespaces would have been a great deal less problematic
> if, early on, a formal protocol distinct from http: had been assigned. If
> you had a namespace uri of the form
> ns:com.example.foo.bar
> people wouldn't have tried to conflate them with http URLs, wouldn't have
> tried to treat them like web directory paths, and would probably have made
> them more readily adopted by developers who were already seeing other
> languages using similar notations for their namespace conventions.
> Surprisingly, I think that namespaces (and prefixes) work better in RDF, at
> least in Turtle or Sparql,
> Kurt Cagle
> Invited Expert, XForms Working Group, W3C
> Managing Editor, XMLToday.org
> kurt.cagle@gmail.com
> 443-837-8725


MLHIM VIP Signup: http://goo.gl/22B0U
Timothy Cook, MSc          +55 21 94711995
MLHIM http://www.mlhim.org
Like Us on FB: https://www.facebook.com/mlhim2
Circle us on G+: http://goo.gl/44EV5
Google Scholar: http://goo.gl/MMZ1o
LinkedIn Profile:http://www.linkedin.com/in/timothywaynecook


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

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index]

News | XML in Industry | Calendar | XML Registry
Marketplace | Resources | MyXML.org | Sponsors | Privacy Statement

Copyright 1993-2007 XML.org. This site is hosted by OASIS