Using any formal classification system to classify essays assumes that the
person that wants to find an essay already fully understands that formal
classification scheme. I doubt that the majority of people who are looking
for essays would know the ACM Computing Classification System. This isn't
to say that the ACMCCS is a bad system, as it seems quite comprehensive.
It's just that it would take me a while just to come up with the formal
category that I thought that the essay I was looking for, might fall under.

It would be like trying to find essays about whales, where the essays were
categorized by the formal taxonomic structure of all life on earth. I
wouldn't have a clue, and it would take me some effort to just find the
category "Cetacea Balaenidae" in the taxonomy of life, let alone finding
the specific essay I wanted about whales. It would be more useful and
efficient for the average person to classify the whale essays under "big
fish", and perhaps "large water-dwelling mammals" as well.

To make categorization easy, the categories should be intuitive. For
information discovery, intuitive categories are much more important than
being concise or formal. A person searching for an essay shouldn't have to
learn a whole categorization scheme just to be able to find a particular
essay.

J's penchant for renaming certain programming concepts as English
parts-of-speech makes keyword searching for specific concepts even more
hit-and-miss than in many other programming languages.  J users would
probably classify verb-building essays under the category "verbs". Essays
on building verbs would then likely be completely missed by a novice
looking for ways to define basic J programs.

An intuitive classification scheme would also put those same essays under
at least one or more other categories - something like "programs" as well
as perhaps "subroutines". Verbs may or may not meet all the formal criteria
for a subroutine, but I would wager to say that more people coming to J
from other languages would find the program-building essays by looking at
the "subroutine" category,rather  than the "verb" category. In information
discovery, intuitiveness is more important than exact formalism.

Different people will have different ideas about how a specific essay
should be classified, so any many-essays-to-one-single-category scheme will
cause a certain set of users to completely overlook essays that were
classified in a category that they didn't think of, or know about. What one
needs instead, is a one-essay-to-many-categories classification scheme
where an essay can be classified in many different ways, depending on the
worldview and vocabulary of the user.

Thus comes the idea of categorizing by collaboration. In this way, you get
the collective vision of many users, both novice and expert, classifying
each essay in the ways and the words that a wide range of users might
understand.

Skip

On Fri, Jul 20, 2012 at 7:31 AM, Brian Schott <[email protected]>wrote:

> http://en.wikipedia.org/wiki/Category:Classification_systems
> http://en.wikipedia.org/wiki/ACM_Computing_Classification_System
>
> Would the ACM system, or part of it, be a starting place?
>
>
>
> --
> (B=)
> ----------------------------------------------------------------------
> For information about J forums see http://www.jsoftware.com/forums.htm
>



-- 
Skip Cave
Cave Consulting LLC
Phone: 214-460-4861
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

Reply via email to