Author: Peter Davison
Date: 08:56:15 09/07/00
Go up one level in this thread
On September 07, 2000 at 11:07:09, pavel wrote: >On September 07, 2000 at 10:43:16, Robert Hyatt wrote: > >>On September 06, 2000 at 18:50:05, Uri Blass wrote: >> >>>On September 06, 2000 at 18:32:35, Uri Blass wrote: >>> >>>>On September 06, 2000 at 17:29:41, Steven Schwartz wrote: >>>> >>>>>Thanks to José de Jesús García Ruvalcaba we have a new >>>>>poll question. It appears to be a hot topic here now. >>>>> >>>>>You are invited to vote at the CCC polling site: >>>>>http://www.icdchess.com/ccc/poll/index.shtml >>>> >>>> >>>>I voted for paying more for the beta testers but I think that it is dependent in >>>>the job that they do. >>>> >>>>If the testers do the job that I read that the chessmaster8000 beta testers >>>>did(only playing 30 games against different personalities at fast time control) >>>>then I think that getting the new version is enough(I think that in this case >>>>they also get the real new version) >>>> >>>>If the testers help the programmers by giving ideas and not only play some hours >>>>against the program then it is not enough. >>>> >>>>I could abstain but I felt that I had to vote for the beta testers after I read >>>>the idea that the programmers do the beta testers a favour. >>>> >>>>The beta testers can test the commercial programs and I do not see the big >>>>advantage of getting it some monthes before the release of a new version when >>>>beta testers do not get even it because they get only the beta version that they >>>>test that is usually slightly weaker. >>>> >>>>In my case the fact that I was a beta tester of Junior4.x did not save me from >>>>buying Junior5 in order to test later Junior5.x >>>> >>>>Uri >>> >>>I can add that I do not blame other people for this fact and it is my fault that >>>I tried to give ideas when I was not asked for it but I understand that other >>>people see the job of the beta tester as giving more than games so I had to vote >>>for the beta testers. >>> >>>If there was a clear definition of the beta tester as someone that has a job to >>>work not more than 10-15 hours and give only games then I could vote in a >>>different way. >>> >>>Uri >> >> >>I personally believe that there is great misconception in the term "beta >>testing". The correct details of what is to be done is normally that you >>receive the beta version, and you use it as though it were a production >>version, but with specific error-reporting requirements. That is how most >>beta testing is done. In rarer cases, beta testing is used to look for more >>specific things. IE "will this program install on many different configurations >>of hardware and operating system versions?" which might be hard to answer in a >>lab setting. >> >>"playing a few games at short time controls" is _not_ "beta testing". Unless >>someone has coined a new beta testing description that is different from the >>SE books I use. >> >>With what I have read here so far, _most_ are not doing beta testing, and >>getting a free copy of a program is probably quite fair. Some (Uri for >>example) go much deeper into things looking for obvious (or not-so-obvious) >>flaws that need attention. Spending that much time is definitely worth paying >>for. >> >>It would seem that the current beta testing approach isn't working very well, >>based on the later bugs and bugfix versions released. The testers aren't >>getting much for their efforts. The programmers aren't getting a lot of effort >>from the testers. Seems 'equitable' in a way. :) > >I agree fully with this, >beta-testing is not about playing a bunch of games between computer programs. > >If that so then the guys from the chessbase and rebel can do it themselves. >there can be another reason for beta-testers (IMO)-----> >for example Christian Koch is a beta-tester for "Gandalf" so he is playing a lot >of games with gandalf against other strong chess playing programs. and then >posting them in public forums, as a result "promoting" the "gandalf" engine in a >way to public. This has nothing to do with finding bugs for the programmer. > > >Pavel Very well pointed out. It is fascinating to watch the loyalty of "beta-testers" to particular programs. Almost as if the "beta-tester" has been bought. Then the way that these "beta-testers" argue and fight over which program is best - not only that, but argue and fight that other opponent programs are bugged or weak. Not so much that programmer fights are an extension of the chess game by other means, but more that beta-tester fights are an extension of programmer fights by other means. Extended zero-sum games, by proxy. Group dynamic. All very well, until cunning programmers and software companies and academics realise that the proxy fighting can be won and lost, that this has implications for their position and revenues, and, importantly, that by _cultivating_ "beta-testers" with free samples, praise, maybe a little financial support, they can acquire _numbers_ of supporters who then go on to make their PR for them. Note that it is not even remotely relevant whether the "beta-tester" provides any advice or feedback to the programmer/company. All that matters is that the PR service has been bought. "Beta-testers" are pawns. Those fighting this battle are as follows: ChessMaster publisher. Who has collected up many willing helpers over the last few months and uses a paid employee to monitor and post to this newsgroup. Bob Hyatt. Who collects support with every download. Rebel company. Which has a policy of giving free samples away to 'key' babblers on the newsgroups. ChessBase. Which works in mysterious ways (to me). Victory, as ever, goes not to quality, not to 'right', but to quantity. Which program/programmer gets the maximum amount of noise made about it. The noise is a function of the manipulative cultivation of the babblers. All pawns.
This page took 0.01 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.