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

Re: [TaxoCoP] Specialty/Other Categories

Expand Messages
  • John O'Gorman
    Patrick, Matt good morning; A few points in rebuttal: First, Scerri does not propose an entirely new periodic table, but a different representation based on
    Message 1 of 46 , May 5, 2010
    • 0 Attachment
       
      Patrick, Matt good morning;
       
      A few points in rebuttal: First, Scerri does not propose an entirely new periodic table, but a different representation based on new information. The on-going debate about the relationship of the chemistry to quantum physics (a la Niels Bohr) is an evolutionary one, not a foundational one.
       
      Point taken on the 'identify, do not classify' debate: as someone pointed out most people think a tomatoe is a vegetable. However, these are not mutually exclusive exercises and one might ask themselves that once having identified everything and creating huge long lists of 'things' what techniques might be available to aggregate them into something more manageable. Hmmm...
       
      Third, the English alphabetic is frozen for a good reason: no one could think of any more letters to add without creating a rather large refactoring challenge. Ask your local database administrator what that feels like. One should point out that spoken language came first, and that while the alphabet is a contractual necessity for publishing a book or composing a response on the taxonomy community of practice, it is a convention that does not constrain which letters you put together to do so. This is also true of the relationship between inorganic elements (however they are 'organized') and living things. You can't have one without the other, but the degrees of freedom available to the latter based on the foundational laws of the former are nothing short of remarkable.
       
      Fourth, the aim of taxonomy or ontology or any other systematic approach is to make the foundational knowledge accessible as a predecessor to more evolved concepts. The extent to which an organizational principle does its job is measured by how many times you need to go back and reorganize. My point was that the three systems I mentioned are very stable, boast an almost universal adoption and as many curricula around the world attest, eminently teachable.
       
      The framework I use is simple a testment to the fact that a universal taxonomy is not only attainable, it is already in use.
       
      John O'
       
       
       
      -----Original Message-----
      From: Matt Moore [mailto:innotecture@...]
      Sent: Tuesday, May 4, 2010 11:27 PM
      To: TaxoCoP@yahoogroups.com
      Subject: Re: [TaxoCoP] Specialty/Other Categories

       
      John,

      Patrick's tackled the changeable nature of frameworks such as the periodic table (BTW the English alphabet is an example of a system that was frozen at a particular point in its history and has since parted ways with the spoken reality of its users).

      I would add that all of these systems (esp. the correct use of the alphabet) require a massive training and enforcement apparatus in order to be effective. If people need 12 years of school to use your taxonomy then you may face adoption issues.

      Matt


      From: John O'Gorman <jogorman@tiberon- ia.com>
      To: TaxoCoP@yahoogroups .com
      Sent: Wed, May 5, 2010 1:21:13 PM
      Subject: Re: [TaxoCoP] Specialty/Other Categories

       

       

       
      Hi Patrick;
       
      I can't help but agree with you again, but only to a certain point. Absolutely correct about people messing with the most elegant taxonomies. Also correct about the need and the utility of the 'other' category, but I said that before. Where we part company is the idea that there is no possibility of a perfectly stable and adaptive framework that humans won't break.
       
      The fixed-facet framework is built on the same principles as the periodic table, the English alphabet and Arabic numbering system: there are a fixed number of elements all 'behaving' in predictable and stable ways. and all capable of describing or as you say representing the most complex and elegant combinations.
       
      I'm not saying that 'other' words or objects can't exist temporarily; just that they can't stay unclassified very long if someone else agrees to use the same label. The introduction of such a term, like misprision, even while suspended in a status of 'unknown' should not break the model or force it change.  Once I discover that 'misprision' is a noun, and a conceptual term at that I'm back to zero in my 'other' class.
       
      The word 'pump', because I can use it as a noun AND a verb is also a concept, but of a special type: 'pump' as a noun in my world is a Function of a physical thing.   That means, among other things, that I can use it as a class in its own right to build a collection of 'pumps' - which are physical things organized by function.
       
      From a taxonomic view, 'pump' would then be instance on the Function facet, just like 'Honda model 342-h78-4-W' is a conceptual instantiation (on the Logical Asset facet) of the Functional concept. Similarly, the one in my garage (John O'Gorman's Honda model 342-h78-4-W serial number 12345) represents (at least to my insurance agent) an intersection of the concept, the function and the digital record in a database.
       
      I figure this group is the correct forum to put these ideas out to...I'm just saying it's a good framework to take to work along with the other taxonomic devices we already use. 
       
      John O'
       
       
      -----Original Message-----
      From: Patrick Lambe [mailto:plambe@ straitsknowledge .com]
      Sent: Tuesday, May 4, 2010 07:06 PM
      To: TaxoCoP@yahoogroups .com
      Subject: Re: [TaxoCoP] Specialty/Other Categories

       

      Hi John


      Well you may think you agree with me, but I think I disagree with you :) - specifically on the suggestion that if you have a fixed set of classified and intersecting facets, you can eliminate the "other".

      My point was that however good your taxonomy design is, and however comprehensive you manage to make it, you cannot control for the user end of the equation - and we human beings have an apparently limitless capacity to "mis"-understand or creatively engage with terms and concepts in ways that seem completely bizarre to others (especially to taxonomy designers). Your examples have used objects, which are easier to pin down though they also fall prey to creative misprision; but in the information and knowledge management world, information artefacts are extremely slippery beasts when it comes to variable human interpretation. Don't get me wrong, well designed facets that aim for comprehensiveness and completeness and clarity can do wonders for enhancing the organisation and finding of content. But human capacity for misprision exceeds the cleverest design. And that's my main reason for defending the "other".

      As Kevin says, the utility of the "other" bucket depends on the quality of your governance and maintenance processes.

      I'm pressing the point because as taxonomists we do tend to be drawn into the taxonomy design side of our system and forget that we are designing for a human system. The idea of a perfect and complete taxonomy design is a holy grail - and a mirage. If we just think of the taxonomy as a representational artefact, of course it looks possible. If we look at the world it looks difficult but aspirational. Only when we look at humans do we see it as the myth that it is. In human terms, taxonomies should be organic, negotiated ways of maintaining common ground among groups of people to help them work together. As their work shifts and the people change, shared understandings constantly break down. Taxonomies stabilise this, and slow it down, if they are well designed, but they never achieve the goal of complete and perfect describability for anyone other than the taxonomist who is satisfied with what they have done.

      P


      Patrick Lambe

      website: http://www.straitsknowledge.com/

      Have you seen our KM Method Cards or
      Organisation Culture Cards?  





      On May 4, 2010, at 7:13 PM, John O'Gorman wrote:


       
      Good morning, all;
       
      While I agree with Patrick's comment that the existence of an 'Other' / 'Unspecified' category can act as a kind of safety valve, it is also a vestigal part of any taxonomic activity. Taxonomies are, of necessity, binary in nature. At each node, the question is asked: "Does this thing that I am classifying fit here?" Yes or No are the only available answers.
       
      In Julie's case for example, the first question is: "Is this thing a pump?", and beyond that the next question is "What kind of pump is it?" with the answer presumably being any number of available "types" in the hierarchy. If she does not have a branch on the tree that a user can point to and say "Yes, that's what type it is.", then the pump falls into the 'Other' category. If a user only has one question to ask - as is the case in all single-dimension taxonomies - you had best have an exhaustive list to accommodate any and all 'types', because "I don't know what type it is." is not really a category.
       
      As I said, I agree with all of Patrick's reasons for having an 'Other' category, but there are ways to mitigate and even eliminate the "other" without a lot of pain. As I have said in previous posts, the answer lies in a fixed set of classified - and intersecting (mutually exclusive) - facets.
       
      John O'
       
      -----Original Message-----
      From: Patrick Lambe [mailto:plambe@ straitsknowledge .com]
      Sent: Monday, May 3, 2010 09:20 PM
      To: TaxoCoP@yahoogroups .com
      Subject: Re: [TaxoCoP] Specialty/Other Categories

       

      I know this is belated, but wanted to throw in a comment on the "let's banish Other/Miscellaneous /Not Otherwise Categorised" recommendation.


      I think in the ideal world "other" would be redundant. But its utility is not simply a function of the imperfection/ incompleteness of the taxonomy design (which is a challenge in itself) - it is also a function of the comprehension and perspective of the content/taxonomy users - ie content classified in "other" is a signal of a usability issue or a completeness issue or an ambiguity issue, or an issue of lack of knowledge in the user.

      Removing the "other" removes that information from your monitoring/feedback system - and then people start forcefitting content into the most convenient categories, or simply not tagging at all.

      So I think despite its lack of purity, "Other" performs a useful feedback function, and so long as it is closely monitored and investigated for what posting there actually mean, it can help to improve the usability and precision of the taxonomy system. If we see it as a temporary bucket that drives improvements, it's fine, and several other commenters talked about that.

      P


      Patrick Lambe

      website: http://www.straitsknowledge.com/

      Have you seen our KM Method Cards or
      Organisation Culture Cards?  





      On Apr 27, 2010, at 4:36 AM, Julie Vittengl wrote:

      John,

      I like the way you think, having no "others" would be ideal.  I don't know if it is practical for us in the current way we gather the product information and the way it is stored.  There are limits on how many values we can have.  So to answer your first question I meant values for an existing facet.  I'm trying to think of a case where we might add new facets with new values, perhaps in rapidly changing product technologies or if we chose to expand or refine the way we describe a product (changing the scope of the taxonomy area).

      Thanks again,
      Julie


      To: TaxoCoP@yahoogroups .com
      From: jogorman@tiberon- ia.com
      Date: Mon, 26 Apr 2010 20:21:39 +0000
      Subject: Re: [TaxoCoP] Specialty/Other Categories

       

      Hey Julie; 
       
      What do they say about 'assuming' something?  My bad, but I think you just hit on one of the reasons my model might be having trouble...when you say you need to 'add five more facets' to accommodate the new 'types' do you mean five more values into an existing facet or five more brand new facets? 
       
      In my model there are only nineteen facets / classes / dimensions of information, so when you ask "...when do we take a look at the 'other' bin?..", the answer (if you want the new pump to play nicely with the rest of the kids) I would say is "always".  In other words, don't have an 'other' category at all.
       
      John O'
       
      -----Original Message-----
      From: Julie Vittengl [mailto:julievitten gl@hotmail. com]
      Sent: Monday, April 26, 2010 02:10 PM
      To: taxocop@yahoogroups .com
      Subject: RE: [TaxoCoP] Specialty/Other Categories

       
      John,

      Thank you for the reply.  We actually do have a faceted classification system for classifying the products once they are in the product category of pumps (in addition to maintaining separate categories for the different types of pumps).  An issue we run into though is for each of the facets (what, action, states, who, discipline, etc) we have an "Other" which ends up collecting a various assortment of things.  So when do we take a look at that "other" bin for the "What" and say ok we need to add 5 more facets to this classification because there are now too many "others".

      Thanks again for any insight,
      Julie



      To: TaxoCoP@yahoogroups .com
      From: jogorman@tiberon- ia.com
      Date: Mon, 26 Apr 2010 19:54:33 +0000
      Subject: Re: [TaxoCoP] Specialty/Other Categories

       

      Hi Julie;
       
      A couple of things come to mind...one is to separate the taxonomy for 'pumps' from the taxonomy for the content/records that go with the pumps; two, consider going fractal - I mean faceted on both, then intersect them as combinations. So, for the pump (which is a functional concept) I would ask the following questions:
       
      pump - what (gas,water, oil, gingerale, beer, breast milk, concrete)
      pump - action (reciprocating,
      pump - states (submersible, in-line, three-phase, heavy-duty, corrosion resistant)
      pump - who (names of competitors, suppliers, installers, if applicable)
      pump - discipline (Oil and Gas, Agriculture, Mining, Medicine, Sports)
       
      Content and records are normally classified by funtion (doc type/record type) alone, but you could easily add others. So, in addition to specification, manual, catalogue and bulletin you could add:
       
      document - states (hard-copy, digital, on-line, archived, recent)
      document - activity (service, maintain, repair, install)
      document - event (publish, update, archive)
       
      You can make a fairly extensive collection just by combining the values I've listed here, and I'm certain that your whole collection could be classified the same way. 
       
      Nothing can escape into the miscellaneous bucket of a good faceted classification system - sort of like accounting: no dollar goes unclassified.
       
      John O'
       
      -----Original Message-----
      From: julierae22 [mailto:julievitten gl@hotmail. com]
      Sent: Monday, April 26, 2010 12:56 PM
      To: TaxoCoP@yahoogroups .com
      Subject: [TaxoCoP] Specialty/Other Categories

       
      Our current product taxonomy has a handful of categories that are "Other" or "Specialty" categories which have essentially turned into junk drawers for records/content that doesn't fit nicely into an existing more specific category. Our taxonomy is a technical products taxonomy so an example would be a pump type that doesn't fit into one of our 20+ pumps categories would end up in "Specialty Pumps." As we continue to add records to these specialty categories they become less and less useful to our users.

      I'm wondering if anyone knows of any existing resources or has any best practices on when to review these types of categories and when to break up them up into new categories.





      Hotmail has tools for the New Busy. Search, chat and e-mail from your inbox. Learn more.
       



      The New Busy think 9 to 5 is a cute idea. Combine multiple calendars with Hotmail. Get busy.



       


       


       

    • John O'Gorman
      Patrick, Matt good morning; A few points in rebuttal: First, Scerri does not propose an entirely new periodic table, but a different representation based on
      Message 46 of 46 , May 5, 2010
      • 0 Attachment
         
        Patrick, Matt good morning;
         
        A few points in rebuttal: First, Scerri does not propose an entirely new periodic table, but a different representation based on new information. The on-going debate about the relationship of the chemistry to quantum physics (a la Niels Bohr) is an evolutionary one, not a foundational one.
         
        Point taken on the 'identify, do not classify' debate: as someone pointed out most people think a tomatoe is a vegetable. However, these are not mutually exclusive exercises and one might ask themselves that once having identified everything and creating huge long lists of 'things' what techniques might be available to aggregate them into something more manageable. Hmmm...
         
        Third, the English alphabetic is frozen for a good reason: no one could think of any more letters to add without creating a rather large refactoring challenge. Ask your local database administrator what that feels like. One should point out that spoken language came first, and that while the alphabet is a contractual necessity for publishing a book or composing a response on the taxonomy community of practice, it is a convention that does not constrain which letters you put together to do so. This is also true of the relationship between inorganic elements (however they are 'organized') and living things. You can't have one without the other, but the degrees of freedom available to the latter based on the foundational laws of the former are nothing short of remarkable.
         
        Fourth, the aim of taxonomy or ontology or any other systematic approach is to make the foundational knowledge accessible as a predecessor to more evolved concepts. The extent to which an organizational principle does its job is measured by how many times you need to go back and reorganize. My point was that the three systems I mentioned are very stable, boast an almost universal adoption and as many curricula around the world attest, eminently teachable.
         
        The framework I use is simple a testment to the fact that a universal taxonomy is not only attainable, it is already in use.
         
        John O'
         
         
         
        -----Original Message-----
        From: Matt Moore [mailto:innotecture@...]
        Sent: Tuesday, May 4, 2010 11:27 PM
        To: TaxoCoP@yahoogroups.com
        Subject: Re: [TaxoCoP] Specialty/Other Categories

         
        John,

        Patrick's tackled the changeable nature of frameworks such as the periodic table (BTW the English alphabet is an example of a system that was frozen at a particular point in its history and has since parted ways with the spoken reality of its users).

        I would add that all of these systems (esp. the correct use of the alphabet) require a massive training and enforcement apparatus in order to be effective. If people need 12 years of school to use your taxonomy then you may face adoption issues.

        Matt


        From: John O'Gorman <jogorman@tiberon- ia.com>
        To: TaxoCoP@yahoogroups .com
        Sent: Wed, May 5, 2010 1:21:13 PM
        Subject: Re: [TaxoCoP] Specialty/Other Categories

         

         

         
        Hi Patrick;
         
        I can't help but agree with you again, but only to a certain point. Absolutely correct about people messing with the most elegant taxonomies. Also correct about the need and the utility of the 'other' category, but I said that before. Where we part company is the idea that there is no possibility of a perfectly stable and adaptive framework that humans won't break.
         
        The fixed-facet framework is built on the same principles as the periodic table, the English alphabet and Arabic numbering system: there are a fixed number of elements all 'behaving' in predictable and stable ways. and all capable of describing or as you say representing the most complex and elegant combinations.
         
        I'm not saying that 'other' words or objects can't exist temporarily; just that they can't stay unclassified very long if someone else agrees to use the same label. The introduction of such a term, like misprision, even while suspended in a status of 'unknown' should not break the model or force it change.  Once I discover that 'misprision' is a noun, and a conceptual term at that I'm back to zero in my 'other' class.
         
        The word 'pump', because I can use it as a noun AND a verb is also a concept, but of a special type: 'pump' as a noun in my world is a Function of a physical thing.   That means, among other things, that I can use it as a class in its own right to build a collection of 'pumps' - which are physical things organized by function.
         
        From a taxonomic view, 'pump' would then be instance on the Function facet, just like 'Honda model 342-h78-4-W' is a conceptual instantiation (on the Logical Asset facet) of the Functional concept. Similarly, the one in my garage (John O'Gorman's Honda model 342-h78-4-W serial number 12345) represents (at least to my insurance agent) an intersection of the concept, the function and the digital record in a database.
         
        I figure this group is the correct forum to put these ideas out to...I'm just saying it's a good framework to take to work along with the other taxonomic devices we already use. 
         
        John O'
         
         
        -----Original Message-----
        From: Patrick Lambe [mailto:plambe@ straitsknowledge .com]
        Sent: Tuesday, May 4, 2010 07:06 PM
        To: TaxoCoP@yahoogroups .com
        Subject: Re: [TaxoCoP] Specialty/Other Categories

         

        Hi John


        Well you may think you agree with me, but I think I disagree with you :) - specifically on the suggestion that if you have a fixed set of classified and intersecting facets, you can eliminate the "other".

        My point was that however good your taxonomy design is, and however comprehensive you manage to make it, you cannot control for the user end of the equation - and we human beings have an apparently limitless capacity to "mis"-understand or creatively engage with terms and concepts in ways that seem completely bizarre to others (especially to taxonomy designers). Your examples have used objects, which are easier to pin down though they also fall prey to creative misprision; but in the information and knowledge management world, information artefacts are extremely slippery beasts when it comes to variable human interpretation. Don't get me wrong, well designed facets that aim for comprehensiveness and completeness and clarity can do wonders for enhancing the organisation and finding of content. But human capacity for misprision exceeds the cleverest design. And that's my main reason for defending the "other".

        As Kevin says, the utility of the "other" bucket depends on the quality of your governance and maintenance processes.

        I'm pressing the point because as taxonomists we do tend to be drawn into the taxonomy design side of our system and forget that we are designing for a human system. The idea of a perfect and complete taxonomy design is a holy grail - and a mirage. If we just think of the taxonomy as a representational artefact, of course it looks possible. If we look at the world it looks difficult but aspirational. Only when we look at humans do we see it as the myth that it is. In human terms, taxonomies should be organic, negotiated ways of maintaining common ground among groups of people to help them work together. As their work shifts and the people change, shared understandings constantly break down. Taxonomies stabilise this, and slow it down, if they are well designed, but they never achieve the goal of complete and perfect describability for anyone other than the taxonomist who is satisfied with what they have done.

        P


        Patrick Lambe

        website: http://www.straitsknowledge.com/

        Have you seen our KM Method Cards or
        Organisation Culture Cards?  





        On May 4, 2010, at 7:13 PM, John O'Gorman wrote:


         
        Good morning, all;
         
        While I agree with Patrick's comment that the existence of an 'Other' / 'Unspecified' category can act as a kind of safety valve, it is also a vestigal part of any taxonomic activity. Taxonomies are, of necessity, binary in nature. At each node, the question is asked: "Does this thing that I am classifying fit here?" Yes or No are the only available answers.
         
        In Julie's case for example, the first question is: "Is this thing a pump?", and beyond that the next question is "What kind of pump is it?" with the answer presumably being any number of available "types" in the hierarchy. If she does not have a branch on the tree that a user can point to and say "Yes, that's what type it is.", then the pump falls into the 'Other' category. If a user only has one question to ask - as is the case in all single-dimension taxonomies - you had best have an exhaustive list to accommodate any and all 'types', because "I don't know what type it is." is not really a category.
         
        As I said, I agree with all of Patrick's reasons for having an 'Other' category, but there are ways to mitigate and even eliminate the "other" without a lot of pain. As I have said in previous posts, the answer lies in a fixed set of classified - and intersecting (mutually exclusive) - facets.
         
        John O'
         
        -----Original Message-----
        From: Patrick Lambe [mailto:plambe@ straitsknowledge .com]
        Sent: Monday, May 3, 2010 09:20 PM
        To: TaxoCoP@yahoogroups .com
        Subject: Re: [TaxoCoP] Specialty/Other Categories

         

        I know this is belated, but wanted to throw in a comment on the "let's banish Other/Miscellaneous /Not Otherwise Categorised" recommendation.


        I think in the ideal world "other" would be redundant. But its utility is not simply a function of the imperfection/ incompleteness of the taxonomy design (which is a challenge in itself) - it is also a function of the comprehension and perspective of the content/taxonomy users - ie content classified in "other" is a signal of a usability issue or a completeness issue or an ambiguity issue, or an issue of lack of knowledge in the user.

        Removing the "other" removes that information from your monitoring/feedback system - and then people start forcefitting content into the most convenient categories, or simply not tagging at all.

        So I think despite its lack of purity, "Other" performs a useful feedback function, and so long as it is closely monitored and investigated for what posting there actually mean, it can help to improve the usability and precision of the taxonomy system. If we see it as a temporary bucket that drives improvements, it's fine, and several other commenters talked about that.

        P


        Patrick Lambe

        website: http://www.straitsknowledge.com/

        Have you seen our KM Method Cards or
        Organisation Culture Cards?  





        On Apr 27, 2010, at 4:36 AM, Julie Vittengl wrote:

        John,

        I like the way you think, having no "others" would be ideal.  I don't know if it is practical for us in the current way we gather the product information and the way it is stored.  There are limits on how many values we can have.  So to answer your first question I meant values for an existing facet.  I'm trying to think of a case where we might add new facets with new values, perhaps in rapidly changing product technologies or if we chose to expand or refine the way we describe a product (changing the scope of the taxonomy area).

        Thanks again,
        Julie


        To: TaxoCoP@yahoogroups .com
        From: jogorman@tiberon- ia.com
        Date: Mon, 26 Apr 2010 20:21:39 +0000
        Subject: Re: [TaxoCoP] Specialty/Other Categories

         

        Hey Julie; 
         
        What do they say about 'assuming' something?  My bad, but I think you just hit on one of the reasons my model might be having trouble...when you say you need to 'add five more facets' to accommodate the new 'types' do you mean five more values into an existing facet or five more brand new facets? 
         
        In my model there are only nineteen facets / classes / dimensions of information, so when you ask "...when do we take a look at the 'other' bin?..", the answer (if you want the new pump to play nicely with the rest of the kids) I would say is "always".  In other words, don't have an 'other' category at all.
         
        John O'
         
        -----Original Message-----
        From: Julie Vittengl [mailto:julievitten gl@hotmail. com]
        Sent: Monday, April 26, 2010 02:10 PM
        To: taxocop@yahoogroups .com
        Subject: RE: [TaxoCoP] Specialty/Other Categories

         
        John,

        Thank you for the reply.  We actually do have a faceted classification system for classifying the products once they are in the product category of pumps (in addition to maintaining separate categories for the different types of pumps).  An issue we run into though is for each of the facets (what, action, states, who, discipline, etc) we have an "Other" which ends up collecting a various assortment of things.  So when do we take a look at that "other" bin for the "What" and say ok we need to add 5 more facets to this classification because there are now too many "others".

        Thanks again for any insight,
        Julie



        To: TaxoCoP@yahoogroups .com
        From: jogorman@tiberon- ia.com
        Date: Mon, 26 Apr 2010 19:54:33 +0000
        Subject: Re: [TaxoCoP] Specialty/Other Categories

         

        Hi Julie;
         
        A couple of things come to mind...one is to separate the taxonomy for 'pumps' from the taxonomy for the content/records that go with the pumps; two, consider going fractal - I mean faceted on both, then intersect them as combinations. So, for the pump (which is a functional concept) I would ask the following questions:
         
        pump - what (gas,water, oil, gingerale, beer, breast milk, concrete)
        pump - action (reciprocating,
        pump - states (submersible, in-line, three-phase, heavy-duty, corrosion resistant)
        pump - who (names of competitors, suppliers, installers, if applicable)
        pump - discipline (Oil and Gas, Agriculture, Mining, Medicine, Sports)
         
        Content and records are normally classified by funtion (doc type/record type) alone, but you could easily add others. So, in addition to specification, manual, catalogue and bulletin you could add:
         
        document - states (hard-copy, digital, on-line, archived, recent)
        document - activity (service, maintain, repair, install)
        document - event (publish, update, archive)
         
        You can make a fairly extensive collection just by combining the values I've listed here, and I'm certain that your whole collection could be classified the same way. 
         
        Nothing can escape into the miscellaneous bucket of a good faceted classification system - sort of like accounting: no dollar goes unclassified.
         
        John O'
         
        -----Original Message-----
        From: julierae22 [mailto:julievitten gl@hotmail. com]
        Sent: Monday, April 26, 2010 12:56 PM
        To: TaxoCoP@yahoogroups .com
        Subject: [TaxoCoP] Specialty/Other Categories

         
        Our current product taxonomy has a handful of categories that are "Other" or "Specialty" categories which have essentially turned into junk drawers for records/content that doesn't fit nicely into an existing more specific category. Our taxonomy is a technical products taxonomy so an example would be a pump type that doesn't fit into one of our 20+ pumps categories would end up in "Specialty Pumps." As we continue to add records to these specialty categories they become less and less useful to our users.

        I'm wondering if anyone knows of any existing resources or has any best practices on when to review these types of categories and when to break up them up into new categories.





        Hotmail has tools for the New Busy. Search, chat and e-mail from your inbox. Learn more.
         



        The New Busy think 9 to 5 is a cute idea. Combine multiple calendars with Hotmail. Get busy.



         


         


         

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