Loading ...
Sorry, an error occurred while loading the content.

Adoption of VCARD in NewsML (was: If you don't want to re-invent the wheel, then you may have to use square wheels)

Expand Messages
  • Jo Rabin
    I ve been meaning to send a reply to this for quite a while and apologise for not doing so in a timely way. While the issue of adoption of vCard is significant
    Message 1 of 16 , May 12, 2003
    • 0 Attachment
      I've been meaning to send a reply to this for quite a while and apologise
      for not doing so in a timely way.

      While the issue of adoption of vCard is significant enough in its own right
      (one or more members has a need to do so), it has an additional importance
      in that it is a good stalking horse for the wider issue of adoption of
      external metadata standards in general.

      My personal summary of the issues follows.

      Jo

      A. Reasons for adopting vCard:

      1. It is expressive of concepts relating to personal contact information
      that are needed in NewsML.
      2. It is mature and has been through a number of revisions and refinements
      that make the IPTC confident that it is stable and able to cope with a wide
      range of requirements.
      3. It is processable by a wide range of applications and devices.
      4. It is formulated and maintained by a standards body "in good standing"
      which is able to draw on expertise beyond that which could be drawn upon by
      the IPTC.

      B. Problems with adopting vCard:

      1. There is no commonly accepted XML based representation.
      2. It is not clear how it should be included in NewsML, even given an XML
      based representation.
      3. Any representation that is derived by the IPTC will be superseded in due
      course by work in the vCard community itself, which is in the process of
      drawing up a standard XML based representation.

      C. Solutions

      1. Allow the incorporation of vCard in its present standard representation
      (i.e. non-xml) by extending the structure of the Metadata element
      2. Use NewsML properties to represent vCard information
      3. Model vCard in XML (using DTD or Schema) and either
      a. include in NewsML
      b. reference as an external DTD
      4. Wait for vCard community to specify in XML

      D. Issues

      The problem with adopting 1 is that it can't be parsed with an XML parser,
      and also requires extension, inter alia, of the Metadata element. Such
      extensions are probably needed anyway for other reasons, so this should not
      be a problem.

      The problem with both 2 and 3 is that it requires the IPTC to infer an
      object model from the existing vCard concrete syntax and define a
      representation of that model. There is certainly a reasonable amount of
      effort and expertise required to do this, especially as NewsML would
      presumably need to be able to transport vCards transparently.

      In future, as vCard evolves, the representation will need to keep track of
      such changes, imposing a maintenance burden on IPTC which it might not be
      able to carry out in a timely way. IPTC representation of vCard would thus
      become non-transparent to the transport of vCards.

      The problem with waiting for an XML vCard representation to become available
      is that at least one member wishes to use NewsML to transport it as soon as
      possible.

      E. Opinion

      The IPTC should wait till vCards are available in XML before adopting them
      as 'the' IPTC preferred representation for personal information. Any other
      course of action will leave IPTC in an evolutionary cul-de-sac, and commits
      it to defining standards which duplicate other efforts and which take it
      outside its member organisations' fields of expertise.

      Prior to this, individual members are at liberty to represent personal
      information in any way the see fit, within the rules of NewsML. IPTC should
      extend NewsML so that members, if they see fit, can use external metadata
      standards that are not based upon XML.

      The IPTC should decide what its policy is in general with regard to use of
      external metadata representations and how it approves such representations
      i.e. what the decision criteria should be.


      -----Original Message-----
      From: Laurent Le Meur [mailto:laurent.lemeur@...]
      Sent: 25 March 2003 08:42
      To: newsml@yahoogroups.com
      Subject: RE : [newsml] If you don't want to re-invent the wheel, then you
      may have to use square wheels.


      I must say that I wouldn't be comfortable with a non-xml vCard
      representation admitted as THE personal info representation recommended
      by IPTC. After all, we said that the most important part of the work
      should be to reach consensus on the object model of our futures
      standards: Misha's excel stylesheet presents a simple structure, which
      expresses most of the known vCard information, and we could agree upon
      various elements of this structure before we discuss their
      implementation.

      Missing in this xls in my opinion is a definition of the terms (but I
      should find them in the Dawson specs), the reason why 'N' sub-elements
      are multiple, the meaning of 'Constraint' for photos, and the usefulness
      of the explanatory types.

      As for the format, I agree with Misha that the main benefit of using
      vCard is that it is the current 'information structure' widely
      understood by pdas and the likes ; transporting it in its native format
      is more comfortable (this is why opening Metadata to any controlled
      structure is a positive thing in my mind), but as the native format is
      non-xml, I would prefer to choose between 1. following native NewsML
      rules (use of Property tags with formal names in upper camel case), and
      proposing xslt to transform it in whichever format is needed at the
      target end (transformation to caps is easy) or 2. following Jabber rules
      (DTD with element names in caps) if NewsML Metadata is voted to be
      'opened' to any metadata structure. As RDF use is a remote possibility
      in native NewsML, I wouldn't easily endorse its use for vCard info in
      the short term.

      Happy to get your opinions on this.

      Best Regards

      Laurent Le Meur
      AFP

      -----Message d'origine-----
      De : Jo Rabin [mailto:jo.rabin@...]
      Envoyé : lundi 24 mars 2003 11:10
      À : newsml@yahoogroups.com
      Objet : RE: [newsml] If you don't want to re-invent the wheel, then you
      may have to use square wheels.

      Hi Misha

      I guess my mail could have been more clearly expressed and with examples
      ...

      The benefit of using vCard is that it is the current _format_ which is
      widely understood, rather than any inferred object model or rendering of
      that model into XML. My suggestion amounts to saying that vCards should
      be
      embedded in (or referenced from) NewsML in that current format. When a
      clear
      XML based version emerges from the vCard community itself, that can be
      used.

      This suggestion implies that the content model of <Metadata> should
      extended
      to be similar to <ContentItem> - in that it is trying to achieve the
      same
      things as ContentItem, only on metadata, rather than content. Since
      metadata
      can describe other metadata as well as content it might be an idea to
      extend
      its content model to reflect this too.

      Hence, [given an extension to the Metadata element] a vCard would be
      embedded something like this:

      ...
      <Metadata MimeType="text/x-vcard">
      <DataContent>
      BEGIN:VCARD
      VERSION:2.1
      N:Rabin;Jo
      FN:Jo Rabin
      ORG:linguafranca.org
      TEL;WORK;VOICE:+44 (20) 9999 9999
      TEL;CELL;VOICE:+44 7999 9999999
      ADR;WORK:;;;;;;England
      LABEL;WORK:England
      EMAIL;PREF;INTERNET:jo@...
      REV:20020517T145752Z
      END:VCARD
      </DataContent>
      </Metadata>
      ...

      Regards
      Jo

      -----Original Message-----
      From: Misha Wolf [mailto:misha.wolf@...]
      Sent: 23 March 2003 12:24
      To: newsml@yahoogroups.com
      Subject: RE: [newsml] If you don't want to re-invent the wheel, then you
      may
      have to use square wheels.


      Hello Jo,

      Lots of good stuff, as usual.

      I probably didn't make it clear that there is no standard
      for representing vCard in XML. Instead there are a number
      of proposals (I'm aware of three), which differ in content
      models, case conventions, etc. In my analysis, I started
      with Frank Dawson's draft DTD, as he co-authored the vCard
      RFC, so he could be said to have some 'moral authority'.
      Where the other proposals differ from Dawson's and where
      these differences made sense to me, I adopted them. I'll
      write more about this when I have the time.

      I would very much welcome an effort to standardise an XML
      representation of vCard. For now, what I want is something
      that reflects the vCard object model and can easily be
      converted via XSLT to the standard non-XML vCard format, to
      any of the proposed XML-based formats, and to snippets of
      XHTML for embedding in XHTML pages.

      Regards,
      Misha


      > -----Original Message-----
      > From: Jo Rabin [mailto:jo.rabin@...]
      > Sent: 22 March 2003 12:29
      > To: newsml@yahoogroups.com
      > Subject: [newsml] If you don't want to re-invent the wheel,
      > then you may have to use square wheels.
      >
      >
      > It certainly does seem like a good idea to use proven methods
      > and standards
      > where they meet the requirements.
      >
      > Some questions are suggested by this philosophy:
      >
      > a) whether you have a clear idea of what your requirements are;
      > b) how to measure whether existing methods and standards meet the
      > requirements;
      > c) in what form to use them if they do;
      > d) how to track their subsequent evolution, after adoption.
      >
      > There's clearly a lot to say about a) and b) but assuming the
      > standards pass
      > the test we still have c) and d) to deal with.
      >
      > Where the originators of such standards have produced an XML
      > representation,
      > and no more is needed than to include 'foreign' or 'external'
      > metadata, it
      > seems there is little more to say than to refer to, or include via
      > namespace, such objects into the NewsML structure. Adoption
      > of Laurent le
      > Meur's proposal at the Nice meeting to extend the Metadata
      > element to allow
      > this would certainly make it clearer how this should be done.
      >
      > In the case where an XML representation of the standard does
      > not exist, or
      > where it is felt that the standard needs to be 'integrated'
      > into NewsML, it
      > is less obvious what should be done.
      >
      > Where NewsML is 'adopting' the standard, and it is not an XML based
      > standard, it would seem tempting to represent it in a NewsML
      > kind of way.
      > Hence, it is not really the standard that NewsML is adopting,
      > it is more the
      > object model behind the standard. In most cases this will not
      > be explicitly
      > stated, so it will be necessary i) to abstract an object
      > model from the
      > standard ii) to evolve a NewsML representation of the object
      > model iii) to
      > verify that there is a 1:1 mapping between the NewsML
      > representation of the
      > object model and the standard's original representation.
      >
      > The above methodology would also apply where the standard is
      > an XML standard
      > and there is the requirement to express it via NewsML's "property"
      > mechanism, rather than in its own schema.
      >
      > For most external metadata standards, in this case vCard, the
      > compelling
      > benefit and validation of the standard is the fact that PDAs
      > and so on can
      > "understand" the format. But what is proposed is abandoning
      > this format in
      > favour of a NewsML based representation of the standard. I am
      > very unsure
      > what is the benefit of adopting a format (rather than a
      > model), e.g. vCard,
      > because of its widespread adoption, but then abandoning the principal
      > benefit by translating it to another format.
      >
      > Additionally, where NewsML maintains its own representation
      > of the standard,
      > and as those standards evolve, NewsML needs to represent and
      > distinguish
      > between the evolving versions of the standard. This seems extremely
      > undesirable.
      >
      > On the whole, this suggests to me that adopted metadata
      > standards should not
      > be 'NewsML-ized'. And that if they are to be adopted as a
      > result of passing
      > test b) above - viz they meet the expressed requirement -
      > they should be
      > adopted "as is", warts and all, in whatever format they are natively
      > represented.
      >
      > Is this the price to be paid for not reinventing the wheel?
      >
      > Jo
      >
      > -----Original Message-----
      > From: Phil Wolff [mailto:philw@...]
      > Sent: 22 March 2003 00:21
      > To: newsml@yahoogroups.com
      > Subject: [newsml] How about iCal and GeoURL in NewsML?
      >
      >
      > Misha Wolf wrote:
      > >One of the outstanding issues for NewsML is how to incorporate the
      > >details of people who have created a piece of news or have
      > contributed
      > >to it's creation. At the IPTC meeting this week, I
      > proposed the use of
      > >vCard for this purpose. The idea is that we wrap a vCard
      > in a Party
      > >element, placing the latter in a Creator or Contributor element.
      >
      > vCard via NewsML is a great idea. Building on that, I'd
      > suggest adding
      > iCalendar for future events and ICBM (a very old standard)
      > addresses for
      > news about specific locations.
      >
      > The iCalendar standard is recognized by Microsoft Outlook
      > and most PDAs as
      > a description of an event or meeting. Sporting and
      > community events, news
      > conferences, embargoes and deadlines, invasion and protest
      > schedules can
      > be
      > distributed completely and exported to personal and portal calendar
      > software. vCal includes fields for event durations, the
      > better to expire
      > an
      > event. http://www.imc.org/pdi/. Browsers recognize the .cal
      > file extension
      > so a click saves the event straight into your calendar
      > without any typing.
      >
      > Geocoding lets you sort incoming news items for proximity
      > to your metro
      > area or in relation to another story ("news near Baghdad"). Rough
      > latitude
      > and longitude are fast and free to find from simple things
      > like US zip or
      > world postal or city codes. (Some ways to find lat and long:
      > http://geourl.org/resources.html). For web sites, you can
      > let me sort the
      > news by how far they are from my home, so the local crime
      > beat, local
      > entertainment coverage, and local events are customized.
      >
      > - phil
      >
      >
      >
      > Philip Wolff
      > evanwolf group
      >
      http://dijest.com/
      http://dijest.com/aka a klog apart
      http://www.ryze.org/?evanwolf
      http://edit.yahoo.com/config/send_webmesg?.target=philw&.src=pg
      groove http://dijest.com/aka/gems/vcards/evanwolf.vcg
      aim:addbuddy?screenname=evanwolff
      011 510.444.8234
      pwolff@...


      Yahoo! Groups Sponsor
      ADVERTISEMENT




      To Post a message, send it to: newsml@...

      To Unsubscribe, send a blank message to:
      newsml-unsubscribe@...

      Your use of Yahoo! Groups is subject to the Yahoo! Terms of Service.


      [Non-text portions of this message have been removed]



      To Post a message, send it to: newsml@...

      To Unsubscribe, send a blank message to: newsml-unsubscribe@...

      Your use of Yahoo! Groups is subject to
      http://docs.yahoo.com/info/terms/



      --------------------------------------------------------------- -
      Visit our Internet site at http://www.reuters.com

      Get closer to the financial markets with Reuters Messaging - for more
      information and to register, visit http://www.reuters.com/messaging

      Any views expressed in this message are those of the individual
      sender, except where the sender specifically states them to be
      the views of Reuters Ltd.


      Yahoo! Groups Sponsor
      ADVERTISEMENT




      To Post a message, send it to: newsml@...

      To Unsubscribe, send a blank message to: newsml-unsubscribe@...

      Your use of Yahoo! Groups is subject to the Yahoo! Terms of Service.



      To Post a message, send it to: newsml@...

      To Unsubscribe, send a blank message to: newsml-unsubscribe@...

      Your use of Yahoo! Groups is subject to
      http://docs.yahoo.com/info/terms/



      Yahoo! Groups Sponsor
      ADVERTISEMENT




      To Post a message, send it to: newsml@...

      To Unsubscribe, send a blank message to: newsml-unsubscribe@...

      Your use of Yahoo! Groups is subject to the Yahoo! Terms of Service.



      [Non-text portions of this message have been removed]
    • Laurent Le Meur
      Jo, Thanks for your clear synthesis; I agree with your conclusions. The only issue is that we have no assurance that a standard vCard will ever be specified be
      Message 2 of 16 , May 15, 2003
      • 0 Attachment
        Jo,

        Thanks for your clear synthesis; I agree with your conclusions.

        The only issue is that we have no assurance that a standard vCard will
        ever be specified be the XML generic standards organizations (W3C, OASIS
        ...). Ron Daniel, who followed the RDF vCard W3C note in 2001, is not
        aware of any new work that has happened with vCard. What could be fine
        is the creation of a XML "vCard Taskforce", like the "RDF Calendar
        taskforce" currently working on events under the auspices of the W3C
        Semantic Web. IPTC people could be part of it, so IPTC is participating,
        but not responsible of its maintenance. Warning: it would automatically
        be an RDF initiative (W3C!). Alternatively, such an initiative could
        perhaps be handled by OASIS, but their level of interest is "higher" in
        the XML stack.

        So, we could begin by opening NewsML to external metadata (and external
        Party meta-information), and let providers fulfill their needs as they
        choose, sharing results amongst members. Then, when a standard comes,
        IPTC could endorse it; we can discuss about the endorsement decision
        process in Aarhus.

        See
        Archived mail from R.Daniel about creating an XML vCard (1998)
        http://lists.xml.org/archives/xml-dev/199801/msg00405.html

        W3C note, 2001, RDF vCard, Renato Iannella (IPR Systems)
        http://www.w3.org/TR/vcard-rdf

        vCard in XML and RDF : links to other material
        http://xml.coverpages.org/vcard.html

        RDF Calendar Task Force
        http://www.w3.org/2002/12/cal/


        Laurent Le Meur


        -----Message d'origine-----
        De : Jo Rabin [mailto:jo.rabin@...]
        Envoyé : lundi 12 mai 2003 17:12
        À : newsml@yahoogroups.com
        Objet : [newsml] Adoption of VCARD in NewsML (was: If you don't want to
        re-invent the wheel, then you may have to use square wheels)

        I've been meaning to send a reply to this for quite a while and
        apologise
        for not doing so in a timely way.

        While the issue of adoption of vCard is significant enough in its own
        right
        (one or more members has a need to do so), it has an additional
        importance
        in that it is a good stalking horse for the wider issue of adoption of
        external metadata standards in general.

        My personal summary of the issues follows.

        Jo

        A. Reasons for adopting vCard:

        1. It is expressive of concepts relating to personal contact information
        that are needed in NewsML.
        2. It is mature and has been through a number of revisions and
        refinements
        that make the IPTC confident that it is stable and able to cope with a
        wide
        range of requirements.
        3. It is processable by a wide range of applications and devices.
        4. It is formulated and maintained by a standards body "in good
        standing"
        which is able to draw on expertise beyond that which could be drawn upon
        by
        the IPTC.

        B. Problems with adopting vCard:

        1. There is no commonly accepted XML based representation.
        2. It is not clear how it should be included in NewsML, even given an
        XML
        based representation.
        3. Any representation that is derived by the IPTC will be superseded in
        due
        course by work in the vCard community itself, which is in the process of
        drawing up a standard XML based representation.

        C. Solutions

        1. Allow the incorporation of vCard in its present standard
        representation
        (i.e. non-xml) by extending the structure of the Metadata element
        2. Use NewsML properties to represent vCard information
        3. Model vCard in XML (using DTD or Schema) and either
        a. include in NewsML
        b. reference as an external DTD
        4. Wait for vCard community to specify in XML

        D. Issues

        The problem with adopting 1 is that it can't be parsed with an XML
        parser,
        and also requires extension, inter alia, of the Metadata element. Such
        extensions are probably needed anyway for other reasons, so this should
        not
        be a problem.

        The problem with both 2 and 3 is that it requires the IPTC to infer an
        object model from the existing vCard concrete syntax and define a
        representation of that model. There is certainly a reasonable amount of
        effort and expertise required to do this, especially as NewsML would
        presumably need to be able to transport vCards transparently.

        In future, as vCard evolves, the representation will need to keep track
        of
        such changes, imposing a maintenance burden on IPTC which it might not
        be
        able to carry out in a timely way. IPTC representation of vCard would
        thus
        become non-transparent to the transport of vCards.

        The problem with waiting for an XML vCard representation to become
        available
        is that at least one member wishes to use NewsML to transport it as soon
        as
        possible.

        E. Opinion

        The IPTC should wait till vCards are available in XML before adopting
        them
        as 'the' IPTC preferred representation for personal information. Any
        other
        course of action will leave IPTC in an evolutionary cul-de-sac, and
        commits
        it to defining standards which duplicate other efforts and which take it
        outside its member organisations' fields of expertise.

        Prior to this, individual members are at liberty to represent personal
        information in any way the see fit, within the rules of NewsML. IPTC
        should
        extend NewsML so that members, if they see fit, can use external
        metadata
        standards that are not based upon XML.

        The IPTC should decide what its policy is in general with regard to use
        of
        external metadata representations and how it approves such
        representations
        i.e. what the decision criteria should be.


        -----Original Message-----
        From: Laurent Le Meur [mailto:laurent.lemeur@...]
        Sent: 25 March 2003 08:42
        To: newsml@yahoogroups.com
        Subject: RE : [newsml] If you don't want to re-invent the wheel, then
        you
        may have to use square wheels.


        I must say that I wouldn't be comfortable with a non-xml vCard
        representation admitted as THE personal info representation
        recommended
        by IPTC. After all, we said that the most important part of the work
        should be to reach consensus on the object model of our futures
        standards: Misha's excel stylesheet presents a simple structure, which
        expresses most of the known vCard information, and we could agree upon
        various elements of this structure before we discuss their
        implementation.

        Missing in this xls in my opinion is a definition of the terms (but I
        should find them in the Dawson specs), the reason why 'N' sub-elements
        are multiple, the meaning of 'Constraint' for photos, and the
        usefulness
        of the explanatory types.

        As for the format, I agree with Misha that the main benefit of using
        vCard is that it is the current 'information structure' widely
        understood by pdas and the likes ; transporting it in its native
        format
        is more comfortable (this is why opening Metadata to any controlled
        structure is a positive thing in my mind), but as the native format is
        non-xml, I would prefer to choose between 1. following native NewsML
        rules (use of Property tags with formal names in upper camel case),
        and
        proposing xslt to transform it in whichever format is needed at the
        target end (transformation to caps is easy) or 2. following Jabber
        rules
        (DTD with element names in caps) if NewsML Metadata is voted to be
        'opened' to any metadata structure. As RDF use is a remote possibility
        in native NewsML, I wouldn't easily endorse its use for vCard info in
        the short term.

        Happy to get your opinions on this.

        Best Regards

        Laurent Le Meur
        AFP

        -----Message d'origine-----
        De : Jo Rabin [mailto:jo.rabin@...]
        Envoyé : lundi 24 mars 2003 11:10
        À : newsml@yahoogroups.com
        Objet : RE: [newsml] If you don't want to re-invent the wheel, then
        you
        may have to use square wheels.

        Hi Misha

        I guess my mail could have been more clearly expressed and with
        examples
        ...

        The benefit of using vCard is that it is the current _format_ which is
        widely understood, rather than any inferred object model or rendering
        of
        that model into XML. My suggestion amounts to saying that vCards
        should
        be
        embedded in (or referenced from) NewsML in that current format. When a
        clear
        XML based version emerges from the vCard community itself, that can be
        used.

        This suggestion implies that the content model of <Metadata> should
        extended
        to be similar to <ContentItem> - in that it is trying to achieve the
        same
        things as ContentItem, only on metadata, rather than content. Since
        metadata
        can describe other metadata as well as content it might be an idea to
        extend
        its content model to reflect this too.

        Hence, [given an extension to the Metadata element] a vCard would be
        embedded something like this:

        ...
        <Metadata MimeType="text/x-vcard">
        <DataContent>
        BEGIN:VCARD
        VERSION:2.1
        N:Rabin;Jo
        FN:Jo Rabin
        ORG:linguafranca.org
        TEL;WORK;VOICE:+44 (20) 9999 9999
        TEL;CELL;VOICE:+44 7999 9999999
        ADR;WORK:;;;;;;England
        LABEL;WORK:England
        EMAIL;PREF;INTERNET:jo@...
        REV:20020517T145752Z
        END:VCARD
        </DataContent>
        </Metadata>
        ...

        Regards
        Jo

        -----Original Message-----
        From: Misha Wolf [mailto:misha.wolf@...]
        Sent: 23 March 2003 12:24
        To: newsml@yahoogroups.com
        Subject: RE: [newsml] If you don't want to re-invent the wheel, then
        you
        may
        have to use square wheels.


        Hello Jo,

        Lots of good stuff, as usual.

        I probably didn't make it clear that there is no standard
        for representing vCard in XML. Instead there are a number
        of proposals (I'm aware of three), which differ in content
        models, case conventions, etc. In my analysis, I started
        with Frank Dawson's draft DTD, as he co-authored the vCard
        RFC, so he could be said to have some 'moral authority'.
        Where the other proposals differ from Dawson's and where
        these differences made sense to me, I adopted them. I'll
        write more about this when I have the time.

        I would very much welcome an effort to standardise an XML
        representation of vCard. For now, what I want is something
        that reflects the vCard object model and can easily be
        converted via XSLT to the standard non-XML vCard format, to
        any of the proposed XML-based formats, and to snippets of
        XHTML for embedding in XHTML pages.

        Regards,
        Misha


        > -----Original Message-----
        > From: Jo Rabin [mailto:jo.rabin@...]
        > Sent: 22 March 2003 12:29
        > To: newsml@yahoogroups.com
        > Subject: [newsml] If you don't want to re-invent the wheel,
        > then you may have to use square wheels.
        >
        >
        > It certainly does seem like a good idea to use proven methods
        > and standards
        > where they meet the requirements.
        >
        > Some questions are suggested by this philosophy:
        >
        > a) whether you have a clear idea of what your requirements are;
        > b) how to measure whether existing methods and standards meet the
        > requirements;
        > c) in what form to use them if they do;
        > d) how to track their subsequent evolution, after adoption.
        >
        > There's clearly a lot to say about a) and b) but assuming the
        > standards pass
        > the test we still have c) and d) to deal with.
        >
        > Where the originators of such standards have produced an XML
        > representation,
        > and no more is needed than to include 'foreign' or 'external'
        > metadata, it
        > seems there is little more to say than to refer to, or include via
        > namespace, such objects into the NewsML structure. Adoption
        > of Laurent le
        > Meur's proposal at the Nice meeting to extend the Metadata
        > element to allow
        > this would certainly make it clearer how this should be done.
        >
        > In the case where an XML representation of the standard does
        > not exist, or
        > where it is felt that the standard needs to be 'integrated'
        > into NewsML, it
        > is less obvious what should be done.
        >
        > Where NewsML is 'adopting' the standard, and it is not an XML based
        > standard, it would seem tempting to represent it in a NewsML
        > kind of way.
        > Hence, it is not really the standard that NewsML is adopting,
        > it is more the
        > object model behind the standard. In most cases this will not
        > be explicitly
        > stated, so it will be necessary i) to abstract an object
        > model from the
        > standard ii) to evolve a NewsML representation of the object
        > model iii) to
        > verify that there is a 1:1 mapping between the NewsML
        > representation of the
        > object model and the standard's original representation.
        >
        > The above methodology would also apply where the standard is
        > an XML standard
        > and there is the requirement to express it via NewsML's "property"
        > mechanism, rather than in its own schema.
        >
        > For most external metadata standards, in this case vCard, the
        > compelling
        > benefit and validation of the standard is the fact that PDAs
        > and so on can
        > "understand" the format. But what is proposed is abandoning
        > this format in
        > favour of a NewsML based representation of the standard. I am
        > very unsure
        > what is the benefit of adopting a format (rather than a
        > model), e.g. vCard,
        > because of its widespread adoption, but then abandoning the
        principal
        > benefit by translating it to another format.
        >
        > Additionally, where NewsML maintains its own representation
        > of the standard,
        > and as those standards evolve, NewsML needs to represent and
        > distinguish
        > between the evolving versions of the standard. This seems extremely
        > undesirable.
        >
        > On the whole, this suggests to me that adopted metadata
        > standards should not
        > be 'NewsML-ized'. And that if they are to be adopted as a
        > result of passing
        > test b) above - viz they meet the expressed requirement -
        > they should be
        > adopted "as is", warts and all, in whatever format they are natively
        > represented.
        >
        > Is this the price to be paid for not reinventing the wheel?
        >
        > Jo
        >
        > -----Original Message-----
        > From: Phil Wolff [mailto:philw@...]
        > Sent: 22 March 2003 00:21
        > To: newsml@yahoogroups.com
        > Subject: [newsml] How about iCal and GeoURL in NewsML?
        >
        >
        > Misha Wolf wrote:
        > >One of the outstanding issues for NewsML is how to incorporate
        the
        > >details of people who have created a piece of news or have
        > contributed
        > >to it's creation. At the IPTC meeting this week, I
        > proposed the use of
        > >vCard for this purpose. The idea is that we wrap a vCard
        > in a Party
        > >element, placing the latter in a Creator or Contributor element.
        >
        > vCard via NewsML is a great idea. Building on that, I'd
        > suggest adding
        > iCalendar for future events and ICBM (a very old standard)
        > addresses for
        > news about specific locations.
        >
        > The iCalendar standard is recognized by Microsoft Outlook
        > and most PDAs as
        > a description of an event or meeting. Sporting and
        > community events, news
        > conferences, embargoes and deadlines, invasion and protest
        > schedules can
        > be
        > distributed completely and exported to personal and portal
        calendar
        > software. vCal includes fields for event durations, the
        > better to expire
        > an
        > event. http://www.imc.org/pdi/. Browsers recognize the .cal
        > file extension
        > so a click saves the event straight into your calendar
        > without any typing.
        >
        > Geocoding lets you sort incoming news items for proximity
        > to your metro
        > area or in relation to another story ("news near Baghdad"). Rough
        > latitude
        > and longitude are fast and free to find from simple things
        > like US zip or
        > world postal or city codes. (Some ways to find lat and long:
        > http://geourl.org/resources.html). For web sites, you can
        > let me sort the
        > news by how far they are from my home, so the local crime
        > beat, local
        > entertainment coverage, and local events are customized.
        >
        > - phil
        >
        >
        >
        > Philip Wolff
        > evanwolf group
        >
        http://dijest.com/
        http://dijest.com/aka a klog apart
        http://www.ryze.org/?evanwolf
        http://edit.yahoo.com/config/send_webmesg?.target=philw&.src=pg
        groove http://dijest.com/aka/gems/vcards/evanwolf.vcg
        aim:addbuddy?screenname=evanwolff
        011 510.444.8234
        pwolff@...


        Yahoo! Groups Sponsor
        ADVERTISEMENT




        To Post a message, send it to: newsml@...

        To Unsubscribe, send a blank message to:
        newsml-unsubscribe@...

        Your use of Yahoo! Groups is subject to the Yahoo! Terms of Service.


        [Non-text portions of this message have been removed]



        To Post a message, send it to: newsml@...

        To Unsubscribe, send a blank message to:
        newsml-unsubscribe@...

        Your use of Yahoo! Groups is subject to
        http://docs.yahoo.com/info/terms/



        --------------------------------------------------------------- -
        Visit our Internet site at http://www.reuters.com

        Get closer to the financial markets with Reuters Messaging - for more
        information and to register, visit http://www.reuters.com/messaging

        Any views expressed in this message are those of the individual
        sender, except where the sender specifically states them to be
        the views of Reuters Ltd.


        Yahoo! Groups Sponsor
        ADVERTISEMENT




        To Post a message, send it to: newsml@...

        To Unsubscribe, send a blank message to:
        newsml-unsubscribe@...

        Your use of Yahoo! Groups is subject to the Yahoo! Terms of Service.



        To Post a message, send it to: newsml@...

        To Unsubscribe, send a blank message to:
        newsml-unsubscribe@...

        Your use of Yahoo! Groups is subject to
        http://docs.yahoo.com/info/terms/



        Yahoo! Groups Sponsor
        ADVERTISEMENT




        To Post a message, send it to: newsml@...

        To Unsubscribe, send a blank message to:
        newsml-unsubscribe@...

        Your use of Yahoo! Groups is subject to the Yahoo! Terms of Service.



        [Non-text portions of this message have been removed]



        To Post a message, send it to: newsml@...

        To Unsubscribe, send a blank message to: newsml-unsubscribe@...

        Your use of Yahoo! Groups is subject to
        http://docs.yahoo.com/info/terms/
      • Misha Wolf
        Hello Jo, You wrote: [...] ... information ... refinements ... wide ... standing ... upon by ... Where is the standards body which formulated and maintains
        Message 3 of 16 , May 30, 2003
        • 0 Attachment
          Hello Jo,

          You wrote:

          [...]

          > A. Reasons for adopting vCard:
          >
          > 1. It is expressive of concepts relating to personal contact
          information
          > that are needed in NewsML.
          > 2. It is mature and has been through a number of revisions and
          refinements
          > that make the IPTC confident that it is stable and able to cope with a
          wide
          > range of requirements.
          > 3. It is processable by a wide range of applications and devices.
          > 4. It is formulated and maintained by a standards body "in good
          standing"
          > which is able to draw on expertise beyond that which could be drawn
          upon by
          > the IPTC.

          Where is the standards body which formulated and maintains the vCard
          spec? My research didn't find one. This is not an argument against
          vCard, but rather an expression of my pedantry :-)

          > B. Problems with adopting vCard:
          >
          > 1. There is no commonly accepted XML based representation.
          > 2. It is not clear how it should be included in NewsML, even given an
          XML
          > based representation.
          > 3. Any representation that is derived by the IPTC will be superseded
          in due
          > course by work in the vCard community itself, which is in the process
          of
          > drawing up a standard XML based representation.

          Where is this work taking place? I found three rival proposals, with
          no indication of collaboration. I wrote to the vcard mailing list,
          asking who would be interested in collaborating on an XML-based
          representation, and I got just one response, from an individual not
          connected with any of the three existing proposals :-(

          [...]

          Regards,
          Misha



          -------------------------------------------------------------- --
          Visit our Internet site at http://www.reuters.com

          Get closer to the financial markets with Reuters Messaging - for more
          information and to register, visit http://www.reuters.com/messaging

          Any views expressed in this message are those of the individual
          sender, except where the sender specifically states them to be
          the views of Reuters Ltd.
        • Misha Wolf
          Hi Laurent, You wrote: [...] ... OASIS ... participating, ... automatically ... in ... I ve just had a chat with Scott McGrath at OASIS. He was positive about
          Message 4 of 16 , May 30, 2003
          • 0 Attachment
            Hi Laurent,

            You wrote:

            [...]

            > The only issue is that we have no assurance that a standard vCard will
            > ever be specified be the XML generic standards organizations (W3C,
            OASIS
            > ...). Ron Daniel, who followed the RDF vCard W3C note in 2001, is not
            > aware of any new work that has happened with vCard. What could be fine
            > is the creation of a XML "vCard Taskforce", like the "RDF Calendar
            > taskforce" currently working on events under the auspices of the W3C
            > Semantic Web. IPTC people could be part of it, so IPTC is
            participating,
            > but not responsible of its maintenance. Warning: it would
            automatically
            > be an RDF initiative (W3C!). Alternatively, such an initiative could
            > perhaps be handled by OASIS, but their level of interest is "higher"
            in
            > the XML stack.

            I've just had a chat with Scott McGrath at OASIS. He was positive
            about the idea and suggested we take the TC Discussion Lists route:
            http://www.oasis-open.org/committees/process.php#disc_lists
            as a first step towards encouraging the formation of a vCard-XML TC
            (Technical Committee). I would be happy to be one of the three
            required proposers, representing Reuters, which is an OASIS member.

            Regards,
            Misha

            [...]



            --------------------------------------------------------------- -
            Visit our Internet site at http://www.reuters.com

            Get closer to the financial markets with Reuters Messaging - for more
            information and to register, visit http://www.reuters.com/messaging

            Any views expressed in this message are those of the individual
            sender, except where the sender specifically states them to be
            the views of Reuters Ltd.
          • Misha Wolf
            Hello Jo, ... - they ... December, ... and ... dare ... Hmmm. The IMC hosts the vcard mailing list, but doesn t seem to be at all (pro)active. ... list of ...
            Message 5 of 16 , May 31, 2003
            • 0 Attachment
              Hello Jo,

              > Misha
              >
              > I thought that IMC (http://www.imc.org) was the relevant organization
              - they
              > seem to think they are (http://www.imc.org/pdi/ - "Beginning in
              December,
              > 1996, the Internet Mail Consortium took on responsibility for the
              > development and promotion of these two important technologies [vcard
              and
              > vcalendar]") - but if you are saying they are not then I would not
              dare
              > challenge you!

              Hmmm. The IMC hosts the vcard mailing list, but doesn't seem to be at
              all (pro)active.

              > In any case, it seems to me that you can take my A. below more as a
              list of
              > requirements than statement of fact. In general, if there is no
              'responsible
              > body' for a standard that is a whole different ball-game.
              >
              > My comment on there being work in progress on an XML version was based
              upon
              > conversation with you. I must have misunderstood what you said. How
              about
              > approaching IMC (pdi-info@...) formally, as IPTC, and asking them?
              Maybe
              > Reuters would do this in its own right?

              I did ask Paul Hoffman (the IMC director) and got the same response
              that Ron Daniel got [1].

              [1] http://lists.xml.org/archives/xml-dev/199801/msg00405.html

              Regards,
              Misha



              --------------------------------------------------------------- -
              Visit our Internet site at http://www.reuters.com

              Get closer to the financial markets with Reuters Messaging - for more
              information and to register, visit http://www.reuters.com/messaging

              Any views expressed in this message are those of the individual
              sender, except where the sender specifically states them to be
              the views of Reuters Ltd.
            Your message has been successfully submitted and would be delivered to recipients shortly.