Author: José Carlos
Date: 05:56:18 03/13/01
Go up one level in this thread
On March 13, 2001 at 05:38:09, John Hatcher wrote: >On March 13, 2001 at 05:12:43, Jouni Uski wrote: > >>I think I have posted this previously, but it's good time to repost. It makes >>no sense to made mini update every two weeks. There should be law to prevent >>update, if real engine strenght is not raised by 30 rating points :-) >>Of course SERIOUS bugs can be corrected. I really respect Junior6, which is >>from 1999 and still almost unbeaten in SSDF matches! >> >>Jouni > > >I couldn't agree less. Your "30 rating points" criterion is arbitrary and >excessive. A programmer should feel free to update his program whenever he >wishes. Why should he have a buggy version in use by thousands of people when >he has found fixes for the bugs? > >Personally, I appreciate a programmer who fixes and improves his program >regularly. What is the down side? If the programmer has found bugs why should >he wait for self-appointed, unsolicited, "testers" to tell him about them? How, >exactly, are you inconvenienced by his updates? You can't caculate the strength of a program if, before you have finished testing, you must throw all away and start again. I prefer a time-depending update creterium. For example, update every 2 months at most. If you release a version with a seriuos bug, of course you can fix it. But if every version you release has serious bugs... you have a problem as a programmer. As a professional programmer on bussines field, that I am, I know it's much better to wait and test, even if you're already delayed, that giving the customer a buggy program. José C. >Regards, >John
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.