Author: Will Singleton
Date: 20:55:32 10/29/99
Go up one level in this thread
On October 29, 1999 at 20:51:59, Dann Corbit wrote: >On October 29, 1999 at 20:39:00, Georg v. Zimmermann wrote: >>>The spirit of discovery is wonderful, but to have each and every chess >>>programmer reinvent the wheel is a billion times worse than having an >>>explanation of the wheel and an explanation of the engine so that someone can >>>come up with something better. If you don't know what is already there, the >>>probability that you will come up with something better is vastly reduced. >> >>hmmm, >>I can't think of a better example: >>Right now what is happening is people who start programming do read and learn >>from code that the only way to get into the air and fly is to use gas leighter >>than air, and so we have dozens of balloons, one flying faster than the other >>... >>but doesn't this prevent anyone from inventing a plane? >In order to make the plane, you must understand a glider. In order to make a >jet plane, you must understand a piston plane. In order to make a rocket plane, >you must understand a jet plane. Now, it is possible that some super-genius >will go right from stone wheel to X-15 but I doubt it very much. > >>Maybe we are still missing something fundamental, and I want to keep hoping that >>I'll discover it- and yes, I think maybe the chances are better when you write >>your _very_ own program. >How will you know if you made an airplane and everyone else is still using >baloons if they won't tell you what they are using or how it works? In other >words, if you don't tell your discoveries to others then we don't know if we >invented something new or not. > >An illustration: >When I was in the Air Force, I was trained in electronics. My first assignement >was Patrick Air Force base in Florida. I was told, "Forget all about that >electronics stuff, you are going to write computer programs." >They gave me a book on Fortran syntax and a book on PL/1 syntax. The first job >I had was to read some records off of a 9 track tape and sort them, creating a >report. After learning what "tape" "record" "JCL" and "sort" meant, I went off >(with PL/1 book in hand) and wrote code to do it. The code worked, but Captain >Holmes was puzzled. "This looks like a bubble sort." he said, "That's not very >efficient." > >"What's a bubble sort?" I responded. You see, I had never seen any kind of sort >before and I just figured out a way to put the records in order. We changed the >routine to make a call to the system sort and it ran more than ten times faster. > If I had enough sense to look into sort a bit harder, that foolish reinvention >would have been avoided. When I first figured it out, I felt kind of proud of >myself and thought I had come up with something novel. But there were already >much better solutions available. > >All of this is not to say that revolutionary ideas do not occur. But (I >believe) they almost never occur in a vacuum. Dann, I understand your point, and it is well taken. But, as I see it, you ignore the difference between publishing source code and publishing ideas. You illustrate the elementary example of the novice programmer and bubble-sort. Notwithstanding the recent crafty bubble-sort thread, you oversimplify; your example doesn't address the issue. That is, the ideas and methods of sorting can be explained through pseudo-code and code-fragments. The actual implementation is left to the programmer. To publish a complete system, working in all respects, does not add more to the general pool of knowledge than does publishing an article explaining the issues. In fact, it does less, and furthermore, has the obvious drawbacks. This seems clear to me, much like the concept that people should work for a living rather than receiving welfare. Do you not agree? Will
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.