Another programmer poll ?

[Rebel]

It's obvious classic old style (off-line) programmer tournaments (CSVN and ICGA to name a few) where programmers gather somewhere in the world to compete and interact socially is under pressure due to the popular alternatives created on the Internet.

I would like to measure the interest among chess programmers with as main theme if they ever plan to attend such an old style (off-line) programmer tournament where programmers gather to play.

What are their wishes, what are the obstacles for them, what would convince them to enter.

For that purpose I plan to create a(nother) poll page (as the CCC software is insufficient), do a(nother) programmer mailing (I think I have most of the programmers email, est. about 80-90) with a call to participate in the poll.

The gathered data will be important for tournament organizers like the CSVN and the ICGA.

But first I think it's crucial the poll should contain the right questions, hence this thread. How does the poll should look like?

So I would say, shoot!

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

First shot

[1] I never plan to attend an ICGA | CSVN tournament because it's too expensive, money and|or timewise.

[2] Do you ever plan to attend an ICGA tournament under the current conditions? [yes, no, maybe]

[3] Do you ever plan to attend an ICGA tournament if conditions are changed to your satisfaction? [yes, no, maybe]

[4] If the ICGA would allow (just) one derivative of a freeware open source would that change your mind about your participation? [yes, no, undecided]

[ Put your comments and other wishes here ]

-----------

Same questions for the CSVN tournament.

A fifth question. I like to know what programmers think especially the new generation.

[5] What if the ICGA | CSVN would allow more than one derivative of a freeware open source ** PROVIDED ** they 1) have added a considerable amount of elo (say 50-100 elo minimum) ** and ** 2) obviously have a different playing style measured by the current (or improved) similarity tester. Are you in favor to modernize tournaments in this way? [yes, no, undecided]

[ Put your comments here ]

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

[5] Don't ever bother me again with such questions.

[Bloodbane]

1 and 2 are basically the same question. Otherwise I think the list is good.

[hgm]

It is definitely a good idea to make a survey on what could be achieved. But you should be careful that it won't become an exercise in futility, like polling the general population whether they ever plan to participate in the Olympics, and then ask them all kind of questions what form of doping usage would be acceptable to them. It is clear that most people have no business being in the Olympics, they very well know that, and that they don't want to participate even if they could should not be seen as a loss.

WCCC and Leiden both have in common that one of the attractions to go there is face-to-face contact with other programmers. But unlike Leiden, the WCCC has a second purpose, namely to determine a Champion. From the disappointing attendance to on-site as well as on-line tournaments, it seems clear that it would not be possible either way to get a representative field of top-engines together if you require active participation, even if it is as trivial asbregistering, or being on-line for a part of the duration of the tourney. The only way seems to be simply recruiting participants.

So the major question should IMO be how 'active' participants would feel about it if some of the top contenders would be programs entered by the organizers to make for a representative tournament, operated by a person not in any way connected to the development team. I realize this would raise a lot of problems. E.g. it would be difficult to justify asking $500 entry fee from a commercial program if the programmer knows he will be entered for free if he doesn't show up at all.

In addition, I think the results of a poll would only be of any use if all interviewed would answer the questions on the assumption that their programs would be, say, top-10 or top-20. While other programmers of course must always remain welcome in WCCC events, it is understandable that people who have zero chance of reaching a decent place in the ranking will not consider the expense of attending worth it. And we should also not care much about that. So more useful questions would be:

Would you attend if your program was top-3 of the World, and had a real change of winning?

No-answers to that question should worry us, and deserve further analysis of the underlying motivation.

[Evert]

I would like to measure the interest among chess programmers with as main theme if they ever plan to attend such an old style (off-line) programmer tournament where programmers gather to play.

What are their wishes, what are the obstacles for them, what would convince them to enter.

Don't know. These things seem like fun, but: chess programming is a hobby (and not my only hobby either), I have a job, a social life and a family. Within those constraints it's difficult to fit in a multi-day off-line event.

Perhaps that is a difference in the present field and the field some 15-20 years ago: there are relatively more people in chess programming who do it in their spare time, but have relatively strong programs.
I don't know if that's true (I wasn't there 20 years ago), but it's worth keeping in mind.

For that purpose I plan to create a(nother) poll page (as the CCC software is insufficient), do a(nother) programmer mailing (I think I have most of the programmers email, est. about 80-90) with a call to participate in the poll.

I don't think a poll is necessarily going to be useful, unless you take extreme care that you're not biasing it towards what you think the reasons might be.

Anyway, comments:

[2] Do you ever plan to attend an ICGA tournament under the current conditions? [yes, no, maybe]

This is too vague: what have you learned when the answer is "no"? "Current conditions" is not explicit enough.

[4] If the ICGA would allow (just) one derivative of a freeware open source would that change your mind about your participation? [yes, no, undecided]

This question betrays a bias that you think this is a problem for many. I'm not so sure it is.
For me personally, it's a non-issue, since it doesn't apply to my program (ok, in principle someone could make a derivative of it, but realistically I don't see that happen any time soon).

[bob]

It's obvious classic old style (off-line) programmer tournaments (CSVN and ICGA to name a few) where programmers gather somewhere in the world to compete and interact socially is under pressure due to the popular alternatives created on the Internet.

I would like to measure the interest among chess programmers with as main theme if they ever plan to attend such an old style (off-line) programmer tournament where programmers gather to play.

What are their wishes, what are the obstacles for them, what would convince them to enter.

For that purpose I plan to create a(nother) poll page (as the CCC software is insufficient), do a(nother) programmer mailing (I think I have most of the programmers email, est. about 80-90) with a call to participate in the poll.

The gathered data will be important for tournament organizers like the CSVN and the ICGA.

But first I think it's crucial the poll should contain the right questions, hence this thread. How does the poll should look like?

So I would say, shoot!

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

First shot

[1] I never plan to attend an ICGA | CSVN tournament because it's too expensive, money and|or timewise.

[2] Do you ever plan to attend an ICGA tournament under the current conditions? [yes, no, maybe]

[3] Do you ever plan to attend an ICGA tournament if conditions are changed to your satisfaction? [yes, no, maybe]

[4] If the ICGA would allow (just) one derivative of a freeware open source would that change your mind about your participation? [yes, no, undecided]

[ Put your comments and other wishes here ]

-----------

Same questions for the CSVN tournament.

A fifth question. I like to know what programmers think especially the new generation.

[5] What if the ICGA | CSVN would allow more than one derivative of a freeware open source ** PROVIDED ** they 1) have added a considerable amount of elo (say 50-100 elo minimum) ** and ** 2) obviously have a different playing style measured by the current (or improved) similarity tester. Are you in favor to modernize tournaments in this way? [yes, no, undecided]

[ Put your comments here ]

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

[5] Don't ever bother me again with such questions.

There is an ambiguity that needs clarification. "if the ICGA allows just one derivative"... What does that mean? The original PLUS a derivative? Just one from any particular family? and how is the "one" chosen?

Second you have another ambiguity when you mention "different playing style." That is not something that is objectively answerable. We've already seen that a tiny change to stockfish can produce "a different playing style."

I'd hate to see the vagueness in the current rules replaced by something that is even more (a) vague; (b) confusing; (b) subjective; (d) unenforceable.

[michiguel]

It's obvious classic old style (off-line) programmer tournaments (CSVN and ICGA to name a few) where programmers gather somewhere in the world to compete and interact socially is under pressure due to the popular alternatives created on the Internet.

I would like to measure the interest among chess programmers with as main theme if they ever plan to attend such an old style (off-line) programmer tournament where programmers gather to play.

What are their wishes, what are the obstacles for them, what would convince them to enter.

For that purpose I plan to create a(nother) poll page (as the CCC software is insufficient), do a(nother) programmer mailing (I think I have most of the programmers email, est. about 80-90) with a call to participate in the poll.

The gathered data will be important for tournament organizers like the CSVN and the ICGA.

But first I think it's crucial the poll should contain the right questions, hence this thread. How does the poll should look like?

So I would say, shoot!

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

First shot

[1] I never plan to attend an ICGA | CSVN tournament because it's too expensive, money and|or timewise.

[2] Do you ever plan to attend an ICGA tournament under the current conditions? [yes, no, maybe]

[3] Do you ever plan to attend an ICGA tournament if conditions are changed to your satisfaction? [yes, no, maybe]

[4] If the ICGA would allow (just) one derivative of a freeware open source would that change your mind about your participation? [yes, no, undecided]

[ Put your comments and other wishes here ]

-----------

Same questions for the CSVN tournament.

A fifth question. I like to know what programmers think especially the new generation.

[5] What if the ICGA | CSVN would allow more than one derivative of a freeware open source ** PROVIDED ** they 1) have added a considerable amount of elo (say 50-100 elo minimum) ** and ** 2) obviously have a different playing style measured by the current (or improved) similarity tester. Are you in favor to modernize tournaments in this way? [yes, no, undecided]

[ Put your comments here ]

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

[5] Don't ever bother me again with such questions.

There is an ambiguity that needs clarification. "if the ICGA allows just one derivative"... What does that mean? The original PLUS a derivative? Just one from any particular family? and how is the "one" chosen?

Second you have another ambiguity when you mention "different playing style." That is not something that is objectively answerable.

That is what the similarity test + cluster analysis + bootstrap analysis measures precisely. Playing similar or different moves is what makes a style, and that is what is measured, almost by definition. Of course a small change can change styles, that is not really new. This is not necessarily related to clones or not.

Miguel

We've already seen that a tiny change to stockfish can produce "a different playing style."

I'd hate to see the vagueness in the current rules replaced by something that is even more (a) vague; (b) confusing; (b) subjective; (d) unenforceable.

[Ferdy]

It's obvious classic old style (off-line) programmer tournaments (CSVN and ICGA to name a few) where programmers gather somewhere in the world to compete and interact socially is under pressure due to the popular alternatives created on the Internet.

I would like to measure the interest among chess programmers with as main theme if they ever plan to attend such an old style (off-line) programmer tournament where programmers gather to play.

What are their wishes, what are the obstacles for them, what would convince them to enter.

For that purpose I plan to create a(nother) poll page (as the CCC software is insufficient), do a(nother) programmer mailing (I think I have most of the programmers email, est. about 80-90) with a call to participate in the poll.

The gathered data will be important for tournament organizers like the CSVN and the ICGA.

But first I think it's crucial the poll should contain the right questions, hence this thread. How does the poll should look like?

So I would say, shoot!

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

First shot

[1] I never plan to attend an ICGA | CSVN tournament because it's too expensive, money and|or timewise.

[2] Do you ever plan to attend an ICGA tournament under the current conditions? [yes, no, maybe]

[3] Do you ever plan to attend an ICGA tournament if conditions are changed to your satisfaction? [yes, no, maybe]

[4] If the ICGA would allow (just) one derivative of a freeware open source would that change your mind about your participation? [yes, no, undecided]

[ Put your comments and other wishes here ]

-----------

Same questions for the CSVN tournament.

A fifth question. I like to know what programmers think especially the new generation.

[5] What if the ICGA | CSVN would allow more than one derivative of a freeware open source ** PROVIDED ** they 1) have added a considerable amount of elo (say 50-100 elo minimum) ** and ** 2) obviously have a different playing style measured by the current (or improved) similarity tester. Are you in favor to modernize tournaments in this way? [yes, no, undecided]

[ Put your comments here ]

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

[5] Don't ever bother me again with such questions.

Separate icga from csvn.

Some Q's below with categories.

x. Expenses
* Would you like to enter if a sponsor will shoulder all your expenses and compensation for time spent?
* Would you like to enter if a sponsor will shoulder all your expenses except your compensation?

x. Derivation
* I don't participate if there is derivative entered.
* I don't care about derivatives.
* Would you like to participate if a derivative (>= 500 than the base) will be entered?
Example, derived engine "AA", yr 2014, rating 2700, and base engine "A", yr 2006, rating 2100.
Derivation started in yr 2006.
"A higher margin is better for the author to possibly prove that he has spent time and other resources
to achieve that strength."
* Would you like to participate if there is a derivative (>= 500 than the base) that is allowed to enter
but is stronger than your engine?
* Would you like to participate if there is a derivative (>= 500 than the base) that is allowed to enter
but is weaker than your engine?
* Would you like to participate when all the allowed derivatives (>= 500 than the base)
started from a base engines that are weaker than your engine at a time when the
derivation started? Example, derived Engine "A", yr 2005, rating 2200 vs your engine "X", yr 2005, rating 2400.
* Would you like to participate if the derivative (>= 500 than the base) entered passes thru a voting
system from original authors. Only those derivative that got majority votes will be accepted.

x. Prize
* I will participate if there are financial rewards to top 3 finishers.

x. Social
* I will participate if I know all the participating authors.
* I will participate to meet the computer chess programmer legends.
* I will participate to meet the computer chess gamers community.
* I will participate to meet the young guns of this generation.
* I will participate to meet the people responsible for the creation of this organization.

x. Personal
* I like to participate if there is a moment where I have health, time and money and in a place that I like.
* I like to participate if there is a moment where I have health, time and money.
* I will participate to be recognized.

x. Support and advancement
* I will participate to support my organization.
* I will participate to present my ideas on computer chess.
* I will participate to talk to my fellows in improving my organization.

x. Business
* I will participate to promote my business products

[bob]

It's obvious classic old style (off-line) programmer tournaments (CSVN and ICGA to name a few) where programmers gather somewhere in the world to compete and interact socially is under pressure due to the popular alternatives created on the Internet.

I would like to measure the interest among chess programmers with as main theme if they ever plan to attend such an old style (off-line) programmer tournament where programmers gather to play.

What are their wishes, what are the obstacles for them, what would convince them to enter.

For that purpose I plan to create a(nother) poll page (as the CCC software is insufficient), do a(nother) programmer mailing (I think I have most of the programmers email, est. about 80-90) with a call to participate in the poll.

The gathered data will be important for tournament organizers like the CSVN and the ICGA.

But first I think it's crucial the poll should contain the right questions, hence this thread. How does the poll should look like?

So I would say, shoot!

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

First shot

[1] I never plan to attend an ICGA | CSVN tournament because it's too expensive, money and|or timewise.

[2] Do you ever plan to attend an ICGA tournament under the current conditions? [yes, no, maybe]

[3] Do you ever plan to attend an ICGA tournament if conditions are changed to your satisfaction? [yes, no, maybe]

[4] If the ICGA would allow (just) one derivative of a freeware open source would that change your mind about your participation? [yes, no, undecided]

[ Put your comments and other wishes here ]

-----------

Same questions for the CSVN tournament.

A fifth question. I like to know what programmers think especially the new generation.

[5] What if the ICGA | CSVN would allow more than one derivative of a freeware open source ** PROVIDED ** they 1) have added a considerable amount of elo (say 50-100 elo minimum) ** and ** 2) obviously have a different playing style measured by the current (or improved) similarity tester. Are you in favor to modernize tournaments in this way? [yes, no, undecided]

[ Put your comments here ]

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

[5] Don't ever bother me again with such questions.

There is an ambiguity that needs clarification. "if the ICGA allows just one derivative"... What does that mean? The original PLUS a derivative? Just one from any particular family? and how is the "one" chosen?

Second you have another ambiguity when you mention "different playing style." That is not something that is objectively answerable.

That is what the similarity test + cluster analysis + bootstrap analysis measures precisely. Playing similar or different moves is what makes a style, and that is what is measured, almost by definition. Of course a small change can change styles, that is not really new. This is not necessarily related to clones or not.

Miguel

I think it is more complicated than that. Over the years, there have been quite a few different programs that have user-selectable playing styles, i.e. "tactical", "cautious", "aggressive", or sometimes even named after specific players.

I still think the term "playing style" is ambiguous. What is the specific threshold for simtest? I'm only interested in rules that are crystal clear, since there has been SO MUCH whining about the existing rule 2. What's the point of having a new rule 2 that still has ambiguous or subjective parts that are just as difficult to interpret as what we already have. If this is not about clones/derivatives, I don't see the point, as that is what rule 2 and this discussion is all about.

We've already seen that a tiny change to stockfish can produce "a different playing style."

I'd hate to see the vagueness in the current rules replaced by something that is even more (a) vague; (b) confusing; (b) subjective; (d) unenforceable.

[Rebel]

Okay, I put something together after having considered all your input, thanks so much. I would like to receive some feedback before I officially launch it, this poll better be good.

I have chosen to keep things as simple as possible as it will be hard enough to display the results of the few (I think) crucial questions in a satisfying way.

http://www.top-5000.nl/poll.htm

[Ferdy]

Okay, I put something together after having considered all your input, thanks so much. I would like to receive some feedback before I officially launch it, this poll better be good.

I have chosen to keep things as simple as possible as it will be hard enough to display the results of the few (I think) crucial questions in a satisfying way.

http://www.top-5000.nl/poll.htm

In item 1, probably better to make it a checkbox, to catch more than 1 reason.
In item 4, 100 is too small, with massive resources at present make it a minimum of 500 rating points, after all this is a world championships.