Talk:Robert Haralick
From Wikipedia, the free encyclopedia
 |
This article is within the scope of WikiProject Biography. For more information, visit the project page. | ||
| Start | This article has been rated as start-Class on the project's quality scale. [FAQ] | ||
[edit] Small disclaimer
If I might set an example of full disclosure in the scholarly tradition: ironically enough, I have a past connection with R. M. Haralick, but nonetheless I believe my additions satisfy WP:NPOV. At any rate I think I have provided sufficient verification for the connection with bible codes.---CH 10:53, 1 June 2006 (UTC)
[edit] Writing style
The prescribed style for biographies is to not use titles like "Dr." or "Professor" except maybe at the very start. See Wikipedia:Manual of Style (biographies). --McKay 05:56, 16 March 2007 (UTC)
[edit] Constraint satisfaction/"Forward checking" algorithm
One aspect not mentioned on the Haralick page is the "Forward Checking" algorithm that Dr. Haralick and I created. (J.J. McGregor simultaneously created and published a very similar algorithm, we are often given co-credit.)
This algorithm is now mentioned in computer science books in the topic of "constraint satisfaction". A Google Scholar search Will show the article introducing the algorithm as just about the most often cited single article in the subject. J.J. McGregor's work was very similar, but ours was much more accessible and prominently published. (This article was also for some time Haralick's most cited single article. OH--forward checking has its own redirect, which is certainly significant enough for mention here!)
The actual history (documented nowhere I know of) is that Dr. Haralick invented a version we called (I believe) "partial lookahead", and I made another improvement to strip out additional consistency checks by realizing that another set of checks could be removed. I coded the algorithm, and Dr. Haralick was very excited in some testing late the next night. We published a comparison of algorithms named "Increasing Tree Search efficiency for Constraint Satisfaction Problems" in 1979, and longer version in 1980 which introduced the algorithm.
It is now available as a part of suite in 9.1 version of SAS/OR (CLP manual). It is widely taught because of its simplicity, and for 10 years as the most efficient known algorithm. Research that came after produced methods that allow constraint satisfaction to be used in puzzle solving applications like genome sequencing--I do not know the degree to which our earlier work provided any foundation, but it is certainly cited. Here is a tutorial to understand the basic area: http://4c.ucc.ie/web/outreach/tutorial.html .
Also I can confirm the involvement of Dr. Haralick in "bible codes". A friend has been a vocal and well respected opponent, noting these codes can occur strictly as statistical anomaly, e.g. could be found anywhere if looked for. I personally have had my hands on decks of cards (many drawers full), back in the days one used those, containing coded copy of the bible, in mid-late 70's, while we were researching various pattern recognition topics. I personally take exception to the language "..deceptively depicted as encodings in books like Moby Dick and War and Peace from those real encodings that occur in the Torah text" which appears in a sentence along with normal pattern recognition results. This is a cliam made by Dr. Harallck and others, widely disputed by others. The author of this biography should not take upon him/her self to analyze the accuracy of such a claim. I believe it should appear in a separate paragraph, and should be accompanied with disclaimer language since it is so controversial, and should in no case be presented as though the WIKI authors are representing Torah codes as "real".
Sorry don't know how to log in at the moment, and don't want to promote myself so won't attempt to add to this article myself, will leave decisions on the issue to others. Thanks 64.136.209.228 20:40, 30 June 2007 (UTC)Gordon Elliott
