Just a friendly remember that I would like (read: need) comments about the
Linkbat data model. In reviewing the descriptions I have come up with several
questions that need to be discussed and answered:
- Does this data model make sense?
- Are the other KU types that w should include?
- Are there any relationships that its missing?
- Do we need to have an ID within the <KnowledgeUnit> tag?
Originally, it was intended to make sure each KU was unique and to use as a
reference point. However, a unique ID can be generated when the XML files are
parsed during the conversion to CSV or database.
- Do we need to include a definition for the Page KU within the same files as
the other KUs?
Eventually we will be converting the HTML pages to XML and all of the
necessary tags can be included in that file. At this point, I think it would
be a lot of work to convert the HTML files to XML and parse them. Since we
can automate the creation of Page KUs, I think we should go ahead and have
the Page KU definition seperate from the actual content for the time being.
That is, we should include the definition with the other KUs **for now**.
- How do we define KU sub-types?
For example, the "man-page" sub-type to the MoreInfo KU.
Should we have a <Type> and <SubType> tag or is the sub-type an attribute of
the Type tag: <Type sub-type="">? We also need to account for the specific
location and other info. For other sites we need a location; for books, we
need the ISBN, author and perhaps publisher? Should these be attributes of
the <Type> tag or seperate tags?
- Should be include a <Language> tag?
Eventually we will have everything translated into other languages. We could
create a special tag to identify the language the would allow us to have a
single knowledge base in multiple language. Or we simply keep the languages
- Should be include a <SkillLevel> tag?
The content is getting to the point where is a lot of it might be overwhelming
to newbies. Therefore, it would be nice to be able to configure the system to
only display KUs that are appropriate for their skill level. This could be
accomplished with a <SkillLevel> tag. Questions could also be marked with
this tag and only questions appropriate for that skill level would be asked.
- Do command KU need a <SeeAlso> tag?
Like the man-pages the point to more information, but that's the point of
simply referencing another KU. If it is important enough to mention as a "See
Also", then it should have a KU, right?
- When we create the Glossary KUs, should we include the page references?
Currently, I have a script that searches through the content pages and creates
a list of all pages that contain each glossary term. This is more efficient
than including the PageRef within the Glossary KU. I see no need and it it is
one more place to forget something. For example, if you add a new content
page we could forget to update the appropriate glossary KUs. Scanning the
content pages seams more efficient.
MAYBE it would be necessary to create a "primary" page for the glossary
entries, but I cannot convince myself that it is necessary.
Well, I am sure there are more questions to answer, but these are the ones
that seem to be most pressing.
"Be more concerned with your character than with your reputation. Your
character is what you really are while your reputation is merely what others
think you are." -- John Wooden
Be sure to visit the Linux Tutorial: http://www.linux-tutorial.info