Posted by Bill McCuistion (198.64.206.92) on October 09, 1999 at 15:05:05:
In Reply to: good logic but........... posted by gary g on October 09, 1999 at 13:37:09:
gary g,
My intention is not to redesign CHMB. It seems to be working very well for its intended purpose. This board seems to serve two purposes for the CH community: (1) work on the soft stuff and (2) work on the hard stuff. I'll let you decide which is the hard and which is the soft.
Sharing experience, strength and hope with fellow CH family is certainly a valuable and worthwhile endeavor. DJ should be satisfied in his efforts on this front.
Finding information in such a wide ranging forum as this is another matter. I don't believe that I proposed a categorization structure at all, much less stipulated any level of complexity. And, you are correct, ideally there should be a structured scheme if classification is to be used. And, there are several ways in which such a classification scheme could be implemented.
The usual way is for the content author to assign
the category at the time the topic is presented. This presumes that the author knows the scheme and can and will assign the appropriate category to the subject matter item.
There are other approaches. For example, content can be classified and categorized after the fact. I have worked with and designed such systems. In essence, the approach we used involved technology used by the CIA to track terrorist activity from information gathered from wire-taps, newspapers, the internet, etc. This technology has been commercialized and is in the market today. Knowledge trees were constructed to associate concepts to categories. This avoids the problem of keyword-based categorization approaches. Content is passed-through the indexing engine and ranked in terms of relevance to various branches of the knowledge tree. Concept-based searches of the knowledge-base then produced surprisingly accurate and precise results, surpassing what even experienced subject matter experts would achieve when they associate the classifications with the content manually.
Simplistic approaches tend to yield simplistic results. Sometimes simplistic results are all that is required. The CH problem is complex. No pun, but trying to assimulate the body of knowledge around CH will make your head hurt. Not being able to retrieve what you are seeking is frustrating, costly and leads to over-simplifications and wrong conclusions. If we only knew what we knew!
Now, I am not even making a suggestion here. Well, actually I might be, related to the top-level topic: Sometimes it is simply too hard to find what you are looking for because of all the chaos. I'm using the term chaos carefully here.
How did you find this site? Did you use a search engine? DJ explains in his intro that he took great pains in learning how to put the right entries in the right locations, to draw those who need this site to this site. I am not suggesting that CHMB be changed. That is not necessary. I would like for the content to be more accessable. I have a tool which can index the site and make it more searchable. This index can be tied to a CH knowledge-tree and made to be concept searchable. This requires no change to CHMB. The tool places a load on the host-computer and the network, as you can imagine. I would coordinate with DJ before anything like this were attempted. Redesign it, no. Supplement it, perhaps.
I have reviewed my past postings. I have made numerous postings. I don't notice any which suggested any type of redesign to the CHMB, ch.com, or to the way the site is operated. I have however, attempted to deal with my CH, help others do the same, float ideas, stay out of the politics, add some humor, and be a good citizen.
Now, I have a business to tend to. I have invested way more time on this than I had intended to. I believe it was a good investment. My bout seems to be winding-down. And, I'm about out of ideas on the CH topic, anyway. The ideas now belong to the group. If they got under your skin, get over it. They were all made in the spirit of community and recovery.