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

Re: [TaxoCoP] hierarchical or associated relationships between company names and industries

Expand Messages
  • Patrick Lambe
    Regulators would also be a part of the financial industry, and they are not (usually) companies. My feeling is that you re dealing with a part-whole
    Message 1 of 15 , Apr 26, 2008
    • 0 Attachment
      Regulators would also be a part of the financial industry, and they are not (usually) companies.

      My feeling is that you're dealing with a part-whole relationship here.

      But a single-hierarchical view is not going to satisfy all needs. You will probably need - at the least - a polyhierarchical view, for those companies that span or support different industries. Health insurers, for example.

      Patrick


      On 27 Apr 2008, at 4:17 AM, Heather Hedden wrote:

      Thanks for your quick reply, Fred.
      It isn't obvious.

      The Baltic Sea "is a" sea or is an instance of a sea. But you cannot say 
      that Citibank "is a" financial industry, so I wasn't sure I could say it 
      is an instance of the industry.

      Yes, Citibank is an instance of "Financial companies" and "Financial 
      companies" is narrower if not a synonym for Financial industry.

      -- Heather

      Fred Leise wrote:
      > Heather,
      >
      > I would consider the relationship between a company name and its 
      > industry to be hierarchical. But rather than representing a strict 
      > whole-part relationship (automobile NT steering wheel), it seems to me 
      > to be a combination of whole-part and the instance relationship. Jean 
      > Aitchison notes that this is "the relationship between a general 
      > category of things and events, expressed by a common noun and an 
      > individual instance of that category, the instance then forming a class 
      > of one which is represented by a proper name. The example she gives is 
      > seas NT Baltic Sea.
      >
      > One could argue that Citibank is a particular instance, or at least a 
      > part instance of "financial industry," for example.
      >
      > To be really strict, the hierarchy should probably be financial industry 
      > NT financial companies NT Citibank. So the industry-company relationship 
      > is essentially eliding that middle term.
      >
      > Fred
      >
      > Fred Leise
      > Principal and Owner
      > ContextualAnalysis, LLC
      > www.contextualanaly sis.com
      > 773-764-2588
      >
      >
      >
      > Heather Hedden wrote:
      > 
      >> Here is a question for thesaurus standards experts: Is the relationship 
      >> between a company name and its corresponding industry considered to be 
      >> hierarchical (broader term whole-part) or rather an associative 
      >> relationship (with a corporate entity type being the broader term 
      >> generic for company name entities)?
      >>
      >> -- Heather
      >>
      >>
      >> 


    • Lisa
      Hi Heather, I can see how it s not obvious, because it seems that, this combined notion conflates set membership ( isa ) with a partonomic relationship which
      Message 2 of 15 , Apr 26, 2008
      • 0 Attachment
        Hi Heather,

        I can see how it's not obvious, because it seems that, this combined notion conflates set
        membership ("isa") with a partonomic relationship which is illogical, but may be necessary if you
        want to represent both notions in a single hierarchy.

        From a logical point-of-view, there is a distinction between being an instance of an organization
        vs. an instance of a set of organizations. For example, we could say that Citibank is an instance
        of a "Financial Corporation", but it's not an instance of "Financial Industry/Corporations"
        (plural)- as "Financial Industry" denotes a collection/set of entities, some of which are each
        instances of "Financial Corporation".

        An industry represents a collection of entities/parties (e.g. organizations and individuals) as
        well as may have other things attached to it (like events, standards, information about what the
        industry does, etc.). A single company is not an industry. Even in the case of monopolies, the
        "industry" would still be a set with one member, the monopoly.

        At least that's how I see it. Interesting issue. Thanks,

        - lisa

        --- Heather Hedden <heather@...> wrote:

        > Thanks for your quick reply, Fred.
        > It isn't obvious.
        >
        > The Baltic Sea "is a" sea or is an instance of a sea. But you cannot say
        > that Citibank "is a" financial industry, so I wasn't sure I could say it
        > is an instance of the industry.
        >
        > Yes, Citibank is an instance of "Financial companies" and "Financial
        > companies" is narrower if not a synonym for Financial industry.
        >
        > -- Heather
        >
        > Fred Leise wrote:
        > > Heather,
        > >
        > > I would consider the relationship between a company name and its
        > > industry to be hierarchical. But rather than representing a strict
        > > whole-part relationship (automobile NT steering wheel), it seems to me
        > > to be a combination of whole-part and the instance relationship. Jean
        > > Aitchison notes that this is "the relationship between a general
        > > category of things and events, expressed by a common noun and an
        > > individual instance of that category, the instance then forming a class
        > > of one which is represented by a proper name. The example she gives is
        > > seas NT Baltic Sea.
        > >
        > > One could argue that Citibank is a particular instance, or at least a
        > > part instance of "financial industry," for example.
        > >
        > > To be really strict, the hierarchy should probably be financial industry
        > > NT financial companies NT Citibank. So the industry-company relationship
        > > is essentially eliding that middle term.
        > >
        > > Fred
        > >
        > > Fred Leise
        > > Principal and Owner
        > > ContextualAnalysis, LLC
        > > www.contextualanalysis.com
        > > 773-764-2588
        > >
        > >
        > >
        > > Heather Hedden wrote:
        > >
        > >> Here is a question for thesaurus standards experts: Is the relationship
        > >> between a company name and its corresponding industry considered to be
        > >> hierarchical (broader term whole-part) or rather an associative
        > >> relationship (with a corporate entity type being the broader term
        > >> generic for company name entities)?
        > >>
        > >> -- Heather
        > >>
        > >>
        > >>
        >
      • Patrick Lambe
        This is why I argued for a distinction between hierarchical and tree taxonomies in my book. In biology, hierarchies follow consistent rules about what the
        Message 3 of 15 , Apr 26, 2008
        • 0 Attachment
          This is why I argued for a distinction between "hierarchical" and "tree" taxonomies in my book.

          In biology, hierarchies follow consistent rules about what the superordinate/subordinate relationship means. In human systems, they don't. It's a mistake I think to assume that hierarchies used to organise human systems and knowledge objects can be "logical" and consistent all the way through. Personally I prefer to use the term "tree" to clarify the difference and to focus on the fact that what counts is the form of the structure, containing subordinate/superordinate entities in relationship to each other, where the nature of the relationships between levels can shift. Using associative relationships doesn't seem to capture this subordinate/superordinate function, unless you define each relationship type specifically, as in an ontology.

          I accept however that "hierarchy" is embedded in common language about taxonomies and thesauri - if we must use it, I think it's critical to recognise that it's used loosely to describe just the form in which the relationships are represented and navigated, and it does not (for human systems) also embody a single consistent meaning of what the relationship entails as it does in biology. From a pragmatic point of view, I think it should be fine to express BT/NT relationships where the nature of the relationship varies, so long as that reflects common categorisation assumptions in the user population.

          Best

          Patrick


          On 27 Apr 2008, at 9:54 AM, Lisa wrote:

          Hi Heather,

          I can see how it's not obvious, because it seems that, this combined notion conflates set
          membership ("isa") with a partonomic relationship which is illogical, but may be necessary if you
          want to represent both notions in a single hierarchy.

          From a logical point-of-view, there is a distinction between being an instance of an organization
          vs. an instance of a set of organizations. For example, we could say that Citibank is an instance
          of a "Financial Corporation" , but it's not an instance of "Financial Industry/Corporatio ns"
          (plural)- as "Financial Industry" denotes a collection/set of entities, some of which are each
          instances of "Financial Corporation" . 

          An industry represents a collection of entities/parties (e.g. organizations and individuals) as
          well as may have other things attached to it (like events, standards, information about what the
          industry does, etc.). A single company is not an industry. Even in the case of monopolies, the
          "industry" would still be a set with one member, the monopoly.

          At least that's how I see it. Interesting issue. Thanks,

          - lisa

          --- Heather Hedden <heather@hedden. net> wrote:

          > Thanks for your quick reply, Fred.
          > It isn't obvious.
          > 
          > The Baltic Sea "is a" sea or is an instance of a sea. But you cannot say 
          > that Citibank "is a" financial industry, so I wasn't sure I could say it 
          > is an instance of the industry.
          > 
          > Yes, Citibank is an instance of "Financial companies" and "Financial 
          > companies" is narrower if not a synonym for Financial industry.
          > 
          > -- Heather
          > 
          > Fred Leise wrote:
          > > Heather,
          > >
          > > I would consider the relationship between a company name and its 
          > > industry to be hierarchical. But rather than representing a strict 
          > > whole-part relationship (automobile NT steering wheel), it seems to me 
          > > to be a combination of whole-part and the instance relationship. Jean 
          > > Aitchison notes that this is "the relationship between a general 
          > > category of things and events, expressed by a common noun and an 
          > > individual instance of that category, the instance then forming a class 
          > > of one which is represented by a proper name. The example she gives is 
          > > seas NT Baltic Sea.
          > >
          > > One could argue that Citibank is a particular instance, or at least a 
          > > part instance of "financial industry," for example.
          > >
          > > To be really strict, the hierarchy should probably be financial industry 
          > > NT financial companies NT Citibank. So the industry-company relationship 
          > > is essentially eliding that middle term.
          > >
          > > Fred
          > >
          > > Fred Leise
          > > Principal and Owner
          > > ContextualAnalysis, LLC
          > > www.contextualanaly sis.com
          > > 773-764-2588
          > >
          > >
          > >
          > > Heather Hedden wrote:
          > > 
          > >> Here is a question for thesaurus standards experts: Is the relationship 
          > >> between a company name and its corresponding industry considered to be 
          > >> hierarchical (broader term whole-part) or rather an associative 
          > >> relationship (with a corporate entity type being the broader term 
          > >> generic for company name entities)?
          > >>
          > >> -- Heather
          > >>
          > >>
          > >> 
          > 


        • Matt Moore
          Hi, At the risk of being prosaic, to what extent does this depend on the purpose of the taxonomy & the knowledge base of its users? One issue with sitting a
          Message 4 of 15 , Apr 26, 2008
          • 0 Attachment
            Hi,

            At the risk of being prosaic, to what extent does this
            depend on the purpose of the taxonomy & the knowledge
            base of its users?

            One issue with sitting a company under an industry
            heading is that it may be a comglomerate (i.e. work in
            several different industries) - so it fits in several
            industry categories.

            As Patrick notes, there are a whole bunch of
            organisations who operate in the financial services
            sector who are not companies. And those companies can
            probably be sub-divided into Banks, Insurers, Brokers,
            etc.

            As a first cut, industry (parent) -> company (child)
            may be "good enough". But it also depends on the
            entities the thesaurus is being applied to. If it's
            companies themselves, then you want something robust
            and more accurate than that. If it's only an
            incidental area then probably not...

            Matt


            ____________________________________________________________________________________
            Be a better friend, newshound, and
            know-it-all with Yahoo! Mobile. Try it now. http://mobile.yahoo.com/;_ylt=Ahu06i62sR8HDtDypao8Wcj9tAcJ
          • Lisa
            Thanks, Patrick. I appreciate your definition of tree . I represent an ontological/logical point-of-view which I originally thought would be helpful in
            Message 5 of 15 , Apr 26, 2008
            • 0 Attachment
              Thanks, Patrick. I appreciate your definition of 'tree'. I represent an ontological/logical
              point-of-view which I originally thought would be helpful in explaining why the initial reluctance
              to make a particular company an instance of an industry seems counter-intuitive.

              As others have mentioned, there are other probably more relevant distinctions such as companies
              that may be situated in multiple industries which shift over time, based on mergers and
              marketplace shifts. Some technology companies are really advertising companies or perhaps even
              just branding organizations. The organizational structures themselves are often more complex to
              represent when one has to think of the various partnerships and matrices. It really depends on
              what type of problem one needs to solve. No need to over-design for a simple problem.


              lisa

              --- Patrick Lambe <plambe@...> wrote:

              > This is why I argued for a distinction between "hierarchical" and
              > "tree" taxonomies in my book.
              >
              > In biology, hierarchies follow consistent rules about what the
              > superordinate/subordinate relationship means. In human systems, they
              > don't. It's a mistake I think to assume that hierarchies used to
              > organise human systems and knowledge objects can be "logical" and
              > consistent all the way through. Personally I prefer to use the term
              > "tree" to clarify the difference and to focus on the fact that what
              > counts is the form of the structure, containing subordinate/
              > superordinate entities in relationship to each other, where the
              > nature of the relationships between levels can shift. Using
              > associative relationships doesn't seem to capture this subordinate/
              > superordinate function, unless you define each relationship type
              > specifically, as in an ontology.
              >
              > I accept however that "hierarchy" is embedded in common language
              > about taxonomies and thesauri - if we must use it, I think it's
              > critical to recognise that it's used loosely to describe just the
              > form in which the relationships are represented and navigated, and it
              > does not (for human systems) also embody a single consistent meaning
              > of what the relationship entails as it does in biology. From a
              > pragmatic point of view, I think it should be fine to express BT/NT
              > relationships where the nature of the relationship varies, so long as
              > that reflects common categorisation assumptions in the user population.
              >
              > Best
              >
              > Patrick
              >
              > Patrick Lambe
              >
              > weblog: www.greenchameleon.com
              > website: www.straitsknowledge.com
              > book: www.organisingknowledge.com
              >
              >
              >
              > On 27 Apr 2008, at 9:54 AM, Lisa wrote:
              > > Hi Heather,
              > >
              > > I can see how it's not obvious, because it seems that, this
              > > combined notion conflates set
              > > membership ("isa") with a partonomic relationship which is
              > > illogical, but may be necessary if you
              > > want to represent both notions in a single hierarchy.
              > >
              > > From a logical point-of-view, there is a distinction between being
              > > an instance of an organization
              > > vs. an instance of a set of organizations. For example, we could
              > > say that Citibank is an instance
              > > of a "Financial Corporation", but it's not an instance of
              > > "Financial Industry/Corporations"
              > > (plural)- as "Financial Industry" denotes a collection/set of
              > > entities, some of which are each
              > > instances of "Financial Corporation".
              > >
              > > An industry represents a collection of entities/parties (e.g.
              > > organizations and individuals) as
              > > well as may have other things attached to it (like events,
              > > standards, information about what the
              > > industry does, etc.). A single company is not an industry. Even in
              > > the case of monopolies, the
              > > "industry" would still be a set with one member, the monopoly.
              > >
              > > At least that's how I see it. Interesting issue. Thanks,
              > >
              > > - lisa
              > >
              > > --- Heather Hedden <heather@...> wrote:
              > >
              > > > Thanks for your quick reply, Fred.
              > > > It isn't obvious.
              > > >
              > > > The Baltic Sea "is a" sea or is an instance of a sea. But you
              > > cannot say
              > > > that Citibank "is a" financial industry, so I wasn't sure I could
              > > say it
              > > > is an instance of the industry.
              > > >
              > > > Yes, Citibank is an instance of "Financial companies" and "Financial
              > > > companies" is narrower if not a synonym for Financial industry.
              > > >
              > > > -- Heather
              > > >
              > > > Fred Leise wrote:
              > > > > Heather,
              > > > >
              > > > > I would consider the relationship between a company name and its
              > > > > industry to be hierarchical. But rather than representing a strict
              > > > > whole-part relationship (automobile NT steering wheel), it
              > > seems to me
              > > > > to be a combination of whole-part and the instance
              > > relationship. Jean
              > > > > Aitchison notes that this is "the relationship between a general
              > > > > category of things and events, expressed by a common noun and an
              > > > > individual instance of that category, the instance then forming
              > > a class
              > > > > of one which is represented by a proper name. The example she
              > > gives is
              > > > > seas NT Baltic Sea.
              > > > >
              > > > > One could argue that Citibank is a particular instance, or at
              > > least a
              > > > > part instance of "financial industry," for example.
              > > > >
              > > > > To be really strict, the hierarchy should probably be financial
              > > industry
              > > > > NT financial companies NT Citibank. So the industry-company
              > > relationship
              > > > > is essentially eliding that middle term.
              > > > >
              > > > > Fred
              > > > >
              > > > > Fred Leise
              > > > > Principal and Owner
              > > > > ContextualAnalysis, LLC
              > > > > www.contextualanalysis.com
              > > > > 773-764-2588
              > > > >
              > > > >
              > > > >
              > > > > Heather Hedden wrote:
              > > > >
              > > > >> Here is a question for thesaurus standards experts: Is the
              > > relationship
              > > > >> between a company name and its corresponding industry
              > > considered to be
              > > > >> hierarchical (broader term whole-part) or rather an associative
              > > > >> relationship (with a corporate entity type being the broader term
              > > > >> generic for company name entities)?
              > > > >>
              > > > >> -- Heather
              > > > >>
              > > > >>
              > > > >>
              > > >
              > >
              > >
              > >
              >
              >
            • Patrick Lambe
              I agree with this Lisa. I think one important factor needs to be the utility of the thesaurus/taxonomy being designed, which goes to Matt s question about the
              Message 6 of 15 , Apr 26, 2008
              • 0 Attachment
                I agree with this Lisa. I think one important factor needs to be the utility of the thesaurus/taxonomy being designed, which goes to Matt's question about the intended audience and purpose. 

                Of course, without knowing Heather's context, we don't know whether it's a simple one or a fiendishly complex one! :)

                Best

                Patrick


                On 27 Apr 2008, at 11:49 AM, Lisa wrote:

                No need to over-design for a simple problem.

                lisa


              • Ken Megill
                I agree with the general drift of the discussion about hierarchies. As I understand it, the conclusions being drawn is that the question of how to design any
                Message 7 of 15 , Apr 27, 2008
                • 0 Attachment

                  I agree with the general drift of the discussion about hierarchies.

                   

                  As I understand it, the conclusions being drawn is that the question of how to design any kind of authorities depends on use….or to be more specific, on the community of practice (although this term is not used in the discussion) which uses the authorities and is bound together by common work.

                   

                  Language grows out of and lives in communities that grow and live through common work and life.  Identifying the community and the nature of the work done by the community is the key to developing effective and efficient authorities, such as taxonomies.

                   

                  There are many kinds of “logic” – and the logic of language in a community is what we are trying to capture when we develop authorities.

                   

                  Any particular person or institution lives in multiple communities – but not an infinite number of communities.  Communities overlap and are defined by the nature of the work they do and they knowledge they produce and the information they use to make that knowledge.

                   

                  When I use the term knowledge, I mean the traditional and time-honored definition of knowledge as “justified true belief”.  One person’s knowledge becomes someone else’s information – we can only share information, not knowledge.

                   

                  Developing hierarchical or other relationships in taxonomy needs to be rooted in an analysis of the work being done by a community.

                   

                  Until we identify the communities and work we cannot develop a good taxonomy.

                   

                  At least that is what I think and what I think I have learned trying to do this kind of work for quite a while.

                   

                  A good discussion.

                   

                  _

                • Gardner, Mike
                  I would agree that it depends on the context of the objects you are trying to model. If you are trying to model from an internal corporate perspective then
                  Message 8 of 15 , Apr 28, 2008
                  • 0 Attachment
                    I would agree that it depends on the context of the objects you are trying to model.
                     
                    If you are trying to model from an internal corporate perspective then your view may be very different from modelling within the objects themselves. For instance, as a business we may work with individual companies as partners, suppliers, competitors, customers, but we may also work with Industry bodies. The overall Industry bodies are not a company as such, so where do they fit? Some companies (large corporations) may fit in to multiple Industries so how would these be modelled? My belief would be to model the objects within the context of the business situation. There are obviously multiple and complex relationships that could be implied here
                     

                    Mike Gardner
                    EDS CIO EKM Team - EDS Taxonomist & Content Rationalization Leader
                    Telephone: +44 (0) 1332 227367 (Hudson House) - Mon
                    +44 (0)1332 663964 (Home Office) - Tue - Fri
                    Mobile: +44 (0)7790 492991
                    Hudson House, Derby, UK
                    micheal.gardner@...

                    We deliver on our commitments so you can deliver on yours.

                    This email contains information which is confidential and may be privileged. Unless you are the intended addressee (or authorised to receive for the addressee) you may not use, forward, copy or disclose to anyone this email or any information contained in this email. If you have received this email in error, please advise the sender by reply email immediately and delete this email.

                    Electronic Data Systems Ltd
                    Registered Office:, Lansdowne House, Berkeley Square, London  W1J 6ER
                    Registered in England no: 53419
                    VAT number: 432 99 5915

                     


                    From: TaxoCoP@yahoogroups.com [mailto:TaxoCoP@yahoogroups.com] On Behalf Of Patrick Lambe
                    Sent: 27 April 2008 07:35
                    To: TaxoCoP@yahoogroups.com
                    Subject: Re: [TaxoCoP] hierarchical or associated relationships between company names and industries

                    I agree with this Lisa. I think one important factor needs to be the utility of the thesaurus/taxonomy being designed, which goes to Matt's question about the intended audience and purpose. 


                    Of course, without knowing Heather's context, we don't know whether it's a simple one or a fiendishly complex one! :)

                    Best

                    Patrick


                    On 27 Apr 2008, at 11:49 AM, Lisa wrote:

                    No need to over-design for a simple problem.

                    lisa


                  • Marc Shimpeno
                    I would definitely consider the relationship to be hierarchical and do in my current Industry/Company taxonomy. Think of it this way: Microsoft is a company
                    Message 9 of 15 , May 1 10:58 AM
                    • 0 Attachment
                      I would definitely consider the relationship to be hierarchical and do in my current Industry/Company taxonomy.  Think of it this way: "Microsoft" is a company that is a part of the "Applications Software" sector.
                       
                      Associative implies a See Also, which doesn't really make sense between companies and the sector they belong to.  Again, notice the ownership aspects in that last sentence?
                       
                      On this note, about the relationships between subjects and the industry sector.  Using the same example, let's say you have the term "Software" and you want to link it with the sector "Application Software".  You have several options, none of which seems better than any of the others.
                       
                      1. Sectors are a different class of information and are NTs to Topics
                      2. Sectors are in the Topical class and are NTs to Topics
                      3. Maybe this is where the associative relationship comes in?  Sector is a RT of Topic
                       
                      Thoughts?

                      Heather Hedden <heather@...> wrote:
                      Here is a question for thesaurus standards experts: Is the relationship
                      between a company name and its corresponding industry considered to be
                      hierarchical (broader term whole-part) or rather an associative
                      relationship (with a corporate entity type being the broader term
                      generic for company name entities)?

                      -- Heather



                      Be a better friend, newshound, and know-it-all with Yahoo! Mobile. Try it now.

                    • Dalbin
                      ... do in my current Industry/Company taxonomy. ... ... US experts have already worked : From ANSI/NISO Z39.19-2005 ***8.3.2 Instance Relationships*** This
                      Message 10 of 15 , May 1 11:58 AM
                      • 0 Attachment
                        >>--- In TaxoCoP@yahoogroups.com, Marc Shimpeno <marcshimp@...> wrote:
                        >> I would definitely consider the relationship to be hierarchical and
                        do in my current Industry/Company taxonomy. ...
                        >>
                        >> Heather Hedden <heather@...> wrote:
                        >> Here is a question for thesaurus standards experts


                        US experts have already worked :
                        From ANSI/NISO Z39.19-2005
                        ***8.3.2 Instance Relationships***
                        This relationship identifies the link between a general category of
                        things or events, expressed by a common noun, and an individual
                        instance of that category, often a proper name. This type of
                        relationship is also known as an "IsA" relationship.
                        8.3.2.1 Codes for the Instance Relationship
                        The hierarchical instance relationship may be indicated by the
                        following abbreviations:
                        • BTI = Broader term (instance)
                        • NTI = Narrower term (instance)

                        Don't hesitate : it's free on ANSI website : http://www.niso.org/home
                        --
                        Sylvie Dalbin
                        Tel : +33 (0)1 42 77 18 10
                        -----------------------------------------------------------
                        Assistance & Techniques Documentaires
                        127 Rue Amelot F-75011 Paris
                        Sites : www.descripteurs.net - www.ATD-doc.com
                        ------------------------------------------------------------
                      • Heather Hedden
                        Hi Sylvie, I have a copy of the ANSI/NISO standard, and it isn t that obvious in this case. Microsoft Corp. is a Software industry. This does not sound
                        Message 11 of 15 , May 1 1:16 PM
                        • 0 Attachment
                          Hi Sylvie,

                          I have a copy of the ANSI/NISO standard, and it isn't that obvious in this case.

                          Microsoft Corp. "is a" Software industry.
                          This does not sound correct.

                          -- Heather

                          On Thu, May 1, 2008 at 2:58 PM, Dalbin <ForumATD@...> wrote:
                          >>--- In TaxoCoP@yahoogroups.com, Marc Shimpeno <marcshimp@...> wrote:
                          >> I would definitely consider the relationship to be hierarchical and
                          do in my current Industry/Company taxonomy. ...
                          >>
                          >> Heather Hedden <heather@...> wrote:
                          >> Here is a question for thesaurus standards experts


                          US experts have already worked :
                          From ANSI/NISO Z39.19-2005
                          ***8.3.2 Instance Relationships***
                          This relationship identifies the link between a general category of
                          things or events, expressed by a common noun, and an individual
                          instance of that category, often a proper name. This type of
                          relationship is also known as an "IsA" relationship.
                          8.3.2.1 Codes for the Instance Relationship
                          The hierarchical instance relationship may be indicated by the
                          following abbreviations:
                          • BTI = Broader term (instance)
                          • NTI = Narrower term (instance)

                          Don't hesitate : it's free on ANSI website : http://www.niso.org/home
                          --
                          Sylvie Dalbin
                          Tel : +33 (0)1 42 77 18 10
                          -----------------------------------------------------------
                          Assistance & Techniques Documentaires
                          127 Rue Amelot F-75011 Paris
                          Sites : www.descripteurs.net - www.ATD-doc.com
                          ------------------------------------------------------------


                        • Banan, Gabrielle
                          Isn t that one of the purposes of a faceted taxonomy? If you know what facet you are in, you would know the nature of the relationship between set and
                          Message 12 of 15 , May 7 7:35 AM
                          • 0 Attachment
                            Isn't that one of the purposes of a faceted taxonomy?  If you know what facet you are in, you would know the nature of the relationship between set and super-set. So you wouldn't be confused as to whether the relationship implied or ignored the issue of ownership - you'd know from the context of the facet.
                            An organizational facet would imply that the set was a member of the super-set in the sense of being part of it through reporting structure. A subject or discipline facet would imply that the set was a member of the super-set in the sense of being an instance of it. For people who use taxonomies for practical purposes such as finding or navigating content, I don't think there's that much confusion.
                             

                            Gabrielle Banan
                            Information Architect
                            Delta Technology
                            gabrielle.banan@...



                            From: TaxoCoP@yahoogroups.com [mailto:TaxoCoP@yahoogroups.com] On Behalf Of Dalbin
                            Sent: Thursday, May 01, 2008 2:59 PM
                            To: TaxoCoP@yahoogroups.com
                            Subject: [TaxoCoP] Re: hierarchical or associated relationships between company names and industrie

                            >>--- In

                            href="mailto:TaxoCoP%40yahoogroups.com">TaxoCoP@yahoogroups .com, Marc Shimpeno <marcshimp@. ..> wrote:
                            >> I would definitely
                            consider the relationship to be hierarchical and
                            do in my current Industry/Company taxonomy. ...
                            >>
                            >> Heather Hedden
                            <heather@... > wrote:
                            >> Here is a question for thesaurus
                            standards experts

                            US experts have already worked :
                            From ANSI/NISO Z39.19-2005
                            ***8.3.2 Instance Relationships* **
                            This relationship identifies the link between a general category of
                            things or events, expressed by a common noun, and an individual
                            instance of that category, often a proper name. This type of
                            relationship is also known as an "IsA" relationship.
                            8.3.2.1 Codes for the Instance Relationship
                            The hierarchical instance relationship may be indicated by the
                            following abbreviations:
                            • BTI = Broader term (instance)
                            • NTI = Narrower term (instance)

                            Don't hesitate : it's free on ANSI website : http://www.niso. org/home
                            --
                            Sylvie Dalbin
                            Tel : +33 (0)1 42 77 18 10
                            ------------ --------- --------- --------- --------- --------- -
                            Assistance & Techniques Documentaires
                            127 Rue Amelot F-75011 Paris
                            Sites : www.descripteurs. net - www.ATD-doc. com
                            ------------ --------- --------- --------- --------- --------- -

                          Your message has been successfully submitted and would be delivered to recipients shortly.