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

Re: Re: [rest-discuss] Addressing context?

Expand Messages
  • jalgermissen@topicmapping.com
    ... In fact, it was rather difficult to grasp this part of Web architecture because we (1) really have in our bones that the *use* of a URI determines what it
    Message 1 of 5 , Oct 1, 2003
    • 0 Attachment
      Mark Baker <distobj@...> schrieb am 01.10.2003, 04:32:37:
      > On Mon, Sep 29, 2003 at 07:34:59AM -0400, algermissen1971 wrote:
      > > I have understood that Web architecture is grounded in the existence
      > > of a single,
      > > unique addressing context (URI allways identifies resource) and
      > > vaguely think that
      > > I see some technical/conceptual advanteges of this, well, axiom.
      >
      > I hope so! 8-)

      In fact, it was rather difficult to grasp this part of Web architecture
      because
      we (1) really have in our bones that the *use* of a URI determines what
      it
      actually identifies. IOW: the use provides addressing context.

      (1) 'we': Topic Map community, but I think also a substantial portion
      of knowledge representation communities in general.

      Sorry again, to raise these issues on this list. While I believe that
      the
      use of names (e.g. URIs) in combination with context is the superior
      choice for knowledge representation, I also understand that a
      knowledge representation system can only be "on the Web" if it uses
      a unique addressing context.

      I am bringing this up here, because I am looking for the technical
      arguments/benefits for a knowledge representation system to be
      "on the Web" as much as possible and, well, I expect to get the
      most qualified answers on this list.

      One argument I know about is that unique addressing context enables
      intermediaries to 'add-in additional information' because they need
      not deal with any addressing ambiguity of the URIs they 'see'. (Hope
      I recalled that correctly).

      Are there other arguments along these lines?



      >
      > > The distinction can also bee seen this way: In TM-land the author of a
      > > Topic Map
      > > decides in what a given URI identifies, in Web-land the controling
      > > authority
      > > makes this decison.
      > >
      > > My particular questions are:
      > >
      > > - Is the idea that an authority controls the 'meaning of URI' a REST
      > > principle or
      > > 'just' a Web principle?
      >
      > I'd say the latter. I don't recall REST talking at all about ownership,
      > though it's possible some degree of that is implicit in the "meaning
      > via use" aspect of REST.

      Mark, do you have any pointers regarding this "meaning via use"
      aspect of REST?

      And, uh, to me this sounds like the exact opposite to "meaning
      via authoritative assignment"...?!?!?

      >
      > > - What exactly are the technical advantages of using a unique
      > > addressing context
      > > when building distributes systems based on REST? Has anybody
      > > analized that
      > > in detail or can help me with some pointers?
      >
      > Well, independent of REST, the benefit of global naming has been
      > discussed for many many years. Not that I'm very familiar with the
      > history, but you might look up the work of Engelbart (Augment/NLS), or
      > Nelson (Xanadu). Sorry I can't be more specific. I'm sure somebody
      > here who's more familiar with hypertext would have a good pointer.
      >
      > Mark.

      Thanks,

      Jan
      > --
      > Mark Baker. Ottawa, Ontario, CANADA. http://www.markbaker.ca
      --
      Jan Algermissen <algermissen@...>
      Consultant & Programmer

      http://www.topicmapping.com
      http://www.gooseworks.org
    • S. Alexander Jacobson
      ... If the question is whether URIs generically have meaning outside of the context in which they are delivered, the answer is no. The most general URI
      Message 2 of 5 , Oct 1, 2003
      • 0 Attachment
        > > > - Is the idea that an authority controls the 'meaning of URI' a REST
        > > > principle or
        > > > 'just' a Web principle?

        If the question is whether URIs generically have
        "meaning" outside of the context in which they
        are delivered, the answer is no.

        The most general URI scheme is the "data:" URI
        scheme (RFC2397) which allows delivery of raw data
        via URI e.g.

        data:,I%20agree%20with%20you.

        Since data: URIs have no authoity but do have
        meaning (via context), I think we can trivially
        assert that the meaning of at least some URIs
        comes from use-context rather than any authority.

        If the question is whether the meaning of URIs
        with authorities e.g. HTTP URIs is controlled by
        those authorities, the answer is still no.
        If I send you an HTTP URI that returns the
        sentence "I agree with you", only context
        tells you who "I" is in the sentence, not the
        authority.

        Alternatively, what do you mean by "meaning of a
        URI"?

        -Alex-

        ___________________________________________________________________
        S. Alexander Jacobson Check out my new blog!!!
        1-212-787-1914 voice http://alexjacobson.com


        On Wed, 1 Oct 2003 jalgermissen@... wrote:

        >
        > Mark Baker <distobj@...> schrieb am 01.10.2003, 04:32:37:
        > > On Mon, Sep 29, 2003 at 07:34:59AM -0400, algermissen1971 wrote:
        > > > I have understood that Web architecture is grounded in the existence
        > > > of a single,
        > > > unique addressing context (URI allways identifies resource) and
        > > > vaguely think that
        > > > I see some technical/conceptual advanteges of this, well, axiom.
        > >
        > > I hope so! 8-)
        >
        > In fact, it was rather difficult to grasp this part of Web architecture
        > because
        > we (1) really have in our bones that the *use* of a URI determines what
        > it
        > actually identifies. IOW: the use provides addressing context.
        >
        > (1) 'we': Topic Map community, but I think also a substantial portion
        > of knowledge representation communities in general.
        >
        > Sorry again, to raise these issues on this list. While I believe that
        > the
        > use of names (e.g. URIs) in combination with context is the superior
        > choice for knowledge representation, I also understand that a
        > knowledge representation system can only be "on the Web" if it uses
        > a unique addressing context.
        >
        > I am bringing this up here, because I am looking for the technical
        > arguments/benefits for a knowledge representation system to be
        > "on the Web" as much as possible and, well, I expect to get the
        > most qualified answers on this list.
        >
        > One argument I know about is that unique addressing context enables
        > intermediaries to 'add-in additional information' because they need
        > not deal with any addressing ambiguity of the URIs they 'see'. (Hope
        > I recalled that correctly).
        >
        > Are there other arguments along these lines?
        >
        >
        >
        > >
        > > > The distinction can also bee seen this way: In TM-land the author of a
        > > > Topic Map
        > > > decides in what a given URI identifies, in Web-land the controling
        > > > authority
        > > > makes this decison.
        > > >
        > > > My particular questions are:
        > > >
        > > > - Is the idea that an authority controls the 'meaning of URI' a REST
        > > > principle or
        > > > 'just' a Web principle?
        > >
        > > I'd say the latter. I don't recall REST talking at all about ownership,
        > > though it's possible some degree of that is implicit in the "meaning
        > > via use" aspect of REST.
        >
        > Mark, do you have any pointers regarding this "meaning via use"
        > aspect of REST?
        >
        > And, uh, to me this sounds like the exact opposite to "meaning
        > via authoritative assignment"...?!?!?
        >
        > >
        > > > - What exactly are the technical advantages of using a unique
        > > > addressing context
        > > > when building distributes systems based on REST? Has anybody
        > > > analized that
        > > > in detail or can help me with some pointers?
        > >
        > > Well, independent of REST, the benefit of global naming has been
        > > discussed for many many years. Not that I'm very familiar with the
        > > history, but you might look up the work of Engelbart (Augment/NLS), or
        > > Nelson (Xanadu). Sorry I can't be more specific. I'm sure somebody
        > > here who's more familiar with hypertext would have a good pointer.
        > >
        > > Mark.
        >
        > Thanks,
        >
        > Jan
        > > --
        > > Mark Baker. Ottawa, Ontario, CANADA. http://www.markbaker.ca
        > --
        > Jan Algermissen <algermissen@...>
        > Consultant & Programmer
        >
        > http://www.topicmapping.com
        > http://www.gooseworks.org
        >
        >
        > To unsubscribe from this group, send an email to:
        > rest-discuss-unsubscribe@yahoogroups.com
        >
        >
        >
        > Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
        >
        >
      • jalgermissen@topicmapping.com
        ... Arggr - I meant to ask: - Is the idea that an authority controls what the semantics of a resource are a REST principle or just a Web principle? Sorry.
        Message 3 of 5 , Oct 1, 2003
        • 0 Attachment
          "S. Alexander Jacobson" <alex@...> schrieb am 01.10.2003, 17:08:07:
          > > > > - Is the idea that an authority controls the 'meaning of URI' a REST
          > > > > principle or
          > > > > 'just' a Web principle?

          Arggr - I meant to ask:

          - Is the idea that an authority controls what the semantics of a
          resource
          are a REST principle or 'just' a Web principle?

          Sorry.

          Jan

          > If the question is whether URIs generically have
          > "meaning" outside of the context in which they
          > are delivered, the answer is no.
          >
          > The most general URI scheme is the "data:" URI
          > scheme (RFC2397) which allows delivery of raw data
          > via URI e.g.
          >
          > data:,I%20agree%20with%20you.
          >
          > Since data: URIs have no authoity but do have
          > meaning (via context), I think we can trivially
          > assert that the meaning of at least some URIs
          > comes from use-context rather than any authority.
          >
          > If the question is whether the meaning of URIs
          > with authorities e.g. HTTP URIs is controlled by
          > those authorities, the answer is still no.
          > If I send you an HTTP URI that returns the
          > sentence "I agree with you", only context
          > tells you who "I" is in the sentence, not the
          > authority.
          >
          > Alternatively, what do you mean by "meaning of a
          > URI"?
          >
          > -Alex-
          >
          > ___________________________________________________________________
          > S. Alexander Jacobson Check out my new blog!!!
          > 1-212-787-1914 voice http://alexjacobson.com
          >
          >
          > On Wed, 1 Oct 2003 jalgermissen@... wrote:
          >
          > >
          > > Mark Baker schrieb am 01.10.2003, 04:32:37:
          > > > On Mon, Sep 29, 2003 at 07:34:59AM -0400, algermissen1971 wrote:
          > > > > I have understood that Web architecture is grounded in the existence
          > > > > of a single,
          > > > > unique addressing context (URI allways identifies resource) and
          > > > > vaguely think that
          > > > > I see some technical/conceptual advanteges of this, well, axiom.
          > > >
          > > > I hope so! 8-)
          > >
          > > In fact, it was rather difficult to grasp this part of Web architecture
          > > because
          > > we (1) really have in our bones that the *use* of a URI determines what
          > > it
          > > actually identifies. IOW: the use provides addressing context.
          > >
          > > (1) 'we': Topic Map community, but I think also a substantial portion
          > > of knowledge representation communities in general.
          > >
          > > Sorry again, to raise these issues on this list. While I believe that
          > > the
          > > use of names (e.g. URIs) in combination with context is the superior
          > > choice for knowledge representation, I also understand that a
          > > knowledge representation system can only be "on the Web" if it uses
          > > a unique addressing context.
          > >
          > > I am bringing this up here, because I am looking for the technical
          > > arguments/benefits for a knowledge representation system to be
          > > "on the Web" as much as possible and, well, I expect to get the
          > > most qualified answers on this list.
          > >
          > > One argument I know about is that unique addressing context enables
          > > intermediaries to 'add-in additional information' because they need
          > > not deal with any addressing ambiguity of the URIs they 'see'. (Hope
          > > I recalled that correctly).
          > >
          > > Are there other arguments along these lines?
          > >
          > >
          > >
          > > >
          > > > > The distinction can also bee seen this way: In TM-land the author of a
          > > > > Topic Map
          > > > > decides in what a given URI identifies, in Web-land the controling
          > > > > authority
          > > > > makes this decison.
          > > > >
          > > > > My particular questions are:
          > > > >
          > > > > - Is the idea that an authority controls the 'meaning of URI' a REST
          > > > > principle or
          > > > > 'just' a Web principle?
          > > >
          > > > I'd say the latter. I don't recall REST talking at all about ownership,
          > > > though it's possible some degree of that is implicit in the "meaning
          > > > via use" aspect of REST.
          > >
          > > Mark, do you have any pointers regarding this "meaning via use"
          > > aspect of REST?
          > >
          > > And, uh, to me this sounds like the exact opposite to "meaning
          > > via authoritative assignment"...?!?!?
          > >
          > > >
          > > > > - What exactly are the technical advantages of using a unique
          > > > > addressing context
          > > > > when building distributes systems based on REST? Has anybody
          > > > > analized that
          > > > > in detail or can help me with some pointers?
          > > >
          > > > Well, independent of REST, the benefit of global naming has been
          > > > discussed for many many years. Not that I'm very familiar with the
          > > > history, but you might look up the work of Engelbart (Augment/NLS), or
          > > > Nelson (Xanadu). Sorry I can't be more specific. I'm sure somebody
          > > > here who's more familiar with hypertext would have a good pointer.
          > > >
          > > > Mark.
          > >
          > > Thanks,
          > >
          > > Jan
          > > > --
          > > > Mark Baker. Ottawa, Ontario, CANADA. http://www.markbaker.ca
          > > --
          > > Jan Algermissen
          > > Consultant & Programmer
          > >
          > > http://www.topicmapping.com
          > > http://www.gooseworks.org
          > >
          > >
          > > To unsubscribe from this group, send an email to:
          > > rest-discuss-unsubscribe@yahoogroups.com
          > >
          > >
          > >
          > > Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
          > >
          > >
          >
          >
          >
          >
          > To unsubscribe from this group, send an email to:
          > rest-discuss-unsubscribe@yahoogroups.com
          >
          >
          >
          > Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
          --
          Jan Algermissen <algermissen@...>
          Consultant & Programmer

          http://www.topicmapping.com
          http://www.gooseworks.org
        Your message has been successfully submitted and would be delivered to recipients shortly.