Fwd: Metadata in DITA

Some thoughts from Fabrice at Antidot ...

    

    

    
Best,

      Kris

      

      Kristen James Eberlein

      Chair, OASIS DITA Technical Committee

      Principal consultant, Eberlein Consulting

      www.eberleinconsulting.com

      +1 919 682-2290; kriseberlein (skype)

      

    

    

      

      -------- Forwarded Message --------
      
        
          

            Subject:
            
            Metadata in DITA
          

          

            Date: 
            Thu, 6 Oct 2016 08:04:57 +0200
          

          

            From: 
            Fabrice LACROIX <>
          

          

            To: 
            
          

        
      
      

      

      
      Hello Kristen,
      
I hope you had a safe trip back.
        
Following our rapid conversation in Montreal, here
          are some notes and personal insights about metadata and DITA.

        

        

        
Real use cases and demands from our clients (even
          if names used here for examples are fictional):

        

        

        
- As a user, I am reading the book « Installation
          Manual of DITA-CMS version 4.2 » and I want to switch to the
          same manual for version 4.3.

        

          

          

          
- As a user, I am reading the Japanese version
            of « DITA-CMSバージョン4.2のインストールマニュアル » and I want to switch to
            the English version of the manual.

        

        

        

        
Simple question: how does Fluid Topics know with
          standard DITA that two books (maps) are different variants of
          the same « conceptual » manual?

        

        

        
We miss in DITA a conceptual model for describing
          how documents relate with each other. The particular case
          described above makes me think about FRBR (https://en.wikipedia.org/wiki/Functional_Requirements_for_Bibliographic_Records).
          We can see here that DITA was design with static publication
          in mind and that « topics and maps » would not be used
          otherwise. 

        
It’s interesting to see that CCMSes like IXIASOFT
          have this information (particularly with DRM in their case)
          but don’t export this information in maps because nothing
          allows to do so. If CCMSes were to export this information,
          they would all do it in a specific way, each one using a
          personal set of metadata, which is exactly what we don’t want.

        

        

        

        

        
I have also an other remark on the subject:  Why
          does DITA reinvent its own set of metadata and taxonomy
          mechanism (namely subject scheme).

        
It would make more sense to leverage existing
          standard vocabularies (such as Dublin Core), and SKOS for
          taxonomies.

        
It’s not a good sign for a standard to show a
          « not invented here » syndrome :-)

        

        

        
I hope that the DITA TC will pay attention to this
          critical subject and we would be happy to help.

        

        

        
Kind regards,

        
Fabrice

        

        

        
        
  
   
   
   
   
   
   
   
   
   
   
  
  
   
   
   
   
   
   
   
   
   
   
   
  

        
        
        

          

            

              
-- 

              
Fabrice
                    Lacroix  |
                      |  Cell: +33 607
                  408 771 

              
Fluid
                    Topics - Enter the age of
                  limitless technical content delivery | www.fluidtopics.com | @fluidtopics | Linkedin