By the way, for the same reason a version of Scrambled Eggs with the Lines of Action piece set would be helpful !

It is great to have this game here, but would a version of the board with numbers instead of coloured spots be possible ? The board reads very bad for partially colour-blind people like me.

pauloaguia: That is a nice idea, I like it. Such poll games use to be very popular, and I know that at least the french-speaking community has already organized one here, using a fellowship discussion board.
It unfortunately seems to be a lot of work to program, however it would probably generate a lot of activity in return, when people will argue about what the best move is or was ; a discussion board attached to the game would probably be called for.
One must realize that one never manages to get all players voting, so that such games are playable only with a fast time control (one day, and if you were away you just missed one vote), or with a fixed percentage of votes needed for the move to be triggered.

Under a player's profile, I would like to see his former user names. I think that it is a bit too easy to change one's identity anyway, but at least this way the information would still be accessible.

Fencer: emmett's request sound good, but I would prefer a complete filter builder, even if we have to wait longer for it. Otherwise we will end with a lot of separate requests, like :
1) Filtering by a range of time control
2) Filtering by vacation / no vacation
3) Filtering by week-end days / no week-end days
4) Filtering out Fischer time controls
5) Filtering out games without autopass
6) Filtering by match type
7) Filtering all but random colour games

Well, basically, every create game option could lead to a filter. Personally, I would probably use at least 1, 2, 3 and 5.

Fencer: Great !

Fencer: Since you seem to be online, any interest in the idea of a feature requests tracker ? It got buried deep in this board, but the thread is here : http://brainking.com/en/Board?bc=3&plla=709992

joshi tm: Fair enough, but don't forget to multiply that split second by the number of unfinished groups in all running tournaments !

AbigailII: I meant an algorithm including the SB calculation. More convoluted, but definitely possible : a trivial solution is to try all possible results, but that does not sound efficient.

joshi tm: Actually it is making an excellent case of why a feature requests tracker would be better than this board !

joshi tm: I also second that, even though as CryingLoser said, this would not be enough when one player has not finished any game before all other games in the pool have finished.
But it may be a bit tedious to write the algorithm which checks whether the winner of a section is certain. If someone volunteered to write the Java code for that, it might speed up things :-)

tonyh: Thank you ! Another tough thing with feature requests as they are is that sometimes they get buried so fast that you don't even know whether Fencer is aware of them :-)

One tough thing with feature requests is that most of the time we don't know what Fencer thinks about them and if he intends to implement them some time. And even when he says that he likes the request, it is very possible that after the request gets buried deep in the present board, everyone will forget about it.

It would be terrific if features requests could be entered, commented upon and followed in a system almost identical to the bug tracker. It would certainly be very easy to program, nothing more is needed than using the bug tracker page with a parameter, which would offer a different set of states, like Accepted and Refused by Fencer.

On the plus side :
- Knowing that they would get a feedback on their requests, users would certainly be encouraged to propose more feature requests.
- Fencer could use the system as a reminder of what he intends to do.
- Fencer could use it also to enter and track the improvements he thinks about on his own.
- People who look at the size of the feature request list would certainly be more understanding about why their own request isn't fulfilled asap.
- The transparency over feature requests would make a tremendous impression on the new users. I would be surprised if any similar site would offer anything close to that. Actually I don't even think that they are close to the current state of Brainking's transparency :-)

On the minus side :
- Knowing that they would get a feedback on their requests, users would certainly be encouraged to propose more feature requests.
- Fencer could see the system as adding pressure on him, like "hey, you accepted this feature request 6 months ago, and it is still not here". I recommend to make very clear that no deadline of any kind is implied on an accepted feature request, and that Fencer would keep any right to change his mind, rejecting a previously accepted request, or accepting a previously rejected one.

mctrivia : 3)
pauloaguia: I can't agree more with all you said.
I just don't even understand the argument about rolling the dices. Do people really believe that they are physically rolling the dices when clicking on "Roll dice" ? We all know that it is just some random number generator kicking off.

pauloaguia: Ah yes you certainly need the spaces before and after the link.

pauloaguia: Ah well thought I knew how to do it. How did you do ? :-)

AbigailII: Except that you can allow your opponents to use auto-pass without being forced to use it yourself.

mctrivia: I find your propositions excellent, maybe the best possible policy. I fully second them !

Fencer: So I was wrong when thinking that this agreement would be a temporary solution ; but I don't understand why it could not be that in the same game one of the opponents is using autopass and the other is not.
Apart from that, autopass is a wonderful feature :-)

grenv: I can't believe anyone could reasonably object to an opponent using autopass any more than they could object to their opponent moving quickly.

If doubling is an option, at least the options should be "double" and "pass" which would at least skip the step of rolling.

I agree 100% on both issues. But pauloaguia is probably right that the rule that both players must agree is only intended for the test phase, which sounds reasonable.

mctrivia: Oh yes you are right, I missed that there was no submit button until the move was finished. Then I also second the motion to move it at the top.

Fencer: When on a small screen, it could happen that even with the smallest board size one has to scroll anyway, but I don't see how having the submit button on top would help ; then one would have to scroll in order to see the entire board.
What about a bloody Javascript keyboard shortcut to submit the move ?

Thad: And colours reversed. Good idea !

Harassed: 1)

Fencer: Yes not only to autopass, but also to automove ? Great news, many autothanks !!

WellyWales: Unlike the others I think that it is an excellent proposal. It is very different from controlling one's game number by oneself. It would be much easier if the system could reject automatically challenges, or kick the person out of a tournament when it starts (who can know at subscrition time how many games he will have at the time when a tournament will start?). I think this is the way it works for pawns, so very little more to implement.

Actually, I would have become a paying member much sooner if I could have set a limit on my number of games.

It would be very helpful to be able to do a text search through all of one's private conversations. That includes in- and out- mailbox messages, but also all messages exchanged during games.
Sometimes I remember having had a nice conversation with someone about some particular subject, but cannot remember who it was, or I remember the old handle but the user now has a new one.

Of course it would also be helpful to search across all discussion boards (including fellowship boards), but that is a different search. Usually one remembers very well if somethings comes from a public board or a private conversation.

BIG BAD WOLF: A nice puzzle is : How many games are played in total in such an elimination tournament of n players (without counting the games vs "ghost players" of course) ?

Fencer: Very cool, thematic tournaments and .mat backgammon format will be two great new features

grenv: I agree, it would be even better if the computer decided randomly who played before the start of the game, without the unnecessary pass move we have one time out of two.
But Fencer may think that it would be a kind of auto-pass, in which case this feature is doomed.

Matarilevich: I see nothing about that in the game rules, but for sure no game starts with a double. In a over-the-board game, when both opponents roll the same die at the beginning of the game, they throw the dices again.
pgt's idea sounds good.

pabloca:
(1) can certainly be supported by the game engine, it works like that in Ambiguous Chess !
(2) I agree with that too. It would be fairer for newcomers who get white-challenged by rating-hungry people.

MTC: I agree 100%.

pauloaguia: I would have the same suggestion about the "private game" attribute (although in fact I would even prefer if private games didn't exists) which can easily get unnoticed. And all this would be consistent with the fact that the "no vacation" attribute is already marked in red : simply marking in red all non-standard game attributes.

Fencer: Cool !

The notifications about finished games, tournaments, etc. come in our inbox as well as the personal messages. They often say something that we already know and are not something that I want to keep, so that I am deleting those messages very regularly, in order to keep the view of the personal messages sent to me. This is a repetitive and boring task.
Those notifications are useful, but it would be much better if they came in a separate, new mailbox, called something like "Notifications". Then either we could leave them pile in there, either have a "Purge" button emptying this mailbox, either they could be scheduled for deletion after they are more than one month old, for instance. In the latter case, we could still choose to keep any notification we want, by archiving it.

King Reza: grenv is right, Ambiguous Chess (Atomic Chess too) is a perfect information game, you get no extra information after playing your move, you have to wait for your opponent's move to get this information... exactly like in standard chess ! For all these games there is no special interest in selecting "move and stay here".

plaintiger: Yes, I like your request, it seems logical for all games where you may get an info after you moved, so also for Dark Chess (are there others?).

*BOB*on*Bush*: I just want to make clear that I never meant to imply in any way that you or other people playing private games were probably cheating, and I am sorry if I could be interpreted in this way. Actually I would rather guess that the cheating on BK is rare. Nevertheless, it could easily happen.

*BOB*on*Bush*: Agreed, the problem is different with games of incomplete information like Battleboats. Nevertheless, in those games the best strategy is certainly not to play each game in the same way. It is rather the most boring strategy.

The problems I have with private games :
1) They are often used by the strongest players, so that nobody can learn from seeing their games. I have nothing personal against those who do that, after all it is allowed by the system and gives them an advantage. But in my opinion BK is scoring an autogoal by allowing that.

2) They can be used to hide cheating, either cheating by use of playing computers, either by pre-arranged games designed to manipulate the rating system. The second option would not be valid if private games were always unrated as have been proposed, but the first one remains.
Of course Fencer can still see the private games, but when cheating occurs it is always the users who spot it in the first place.

Well, if private games were limited to unrated games of incomplete information (Battleboats, Dark Chess, ....) it would already be a good start.

hexkid: I think that grenv does not like the "feature" of setting a game private, and I fully agree with him.
Actually, I am hereby making a feature request for the abolition of private games

Hrqls: Your version would be perfect with me. In fact, the same type of filter like the one for posting games in the waiting room could be used.

Fencer: Thank you !

LionsLair: I support that. Generally, it seems to be good design to have all information having a significant influence on the game's future appear in the display. It was the same sort of upgrade when the check counter has been added to Three-check Chess.

Fencer: Too bad about Gigamic because they have other great two-players games
But here is a great list of abstract games, always with mention of the copyright when there is one. There I found all the most interesting ones I knew and many others : http://www.di.fc.ul.pt/~jpn/gv/index.htm .
Tycho: I vote against Abalone, because though it is a very nice game, it is all too easy to build a unbreakable wall and draw in this way. Even the authors acknowledged that, and had to invent contrived etiquette rules to force the players into taking otherwise unnecessary risks.
The other ones are great suggestions, I may add some very simple and quite well-known games (all on the website I mentioned) : Pogo (good start in stacking games), Twixt, Hex (good start in connexion games), Isola.

Summertop: I don't know whether this is the biggest problem, but there are also different letter values in different languages. In English it would be strange to use K=10, W=10 like we do in French.

Mr. Shumway: This would also be nice !

volant: I think it is impossible to do now and it would be a nice feature. I see so many times someone posting something like 6 waiting games with each color in some unbalanced game, and some time later there are still 5 black games available but only 1 white game - which mean that the contenders take the white challenges but ignore the black ones.
To get around that, you can offer two-games match, but those are then counted (statistics- and rating-wise) as only one game.

joshi tm: Yes, I would also like this feature ! Of course nobody is forced to start more games that he would like, but it is much easier to decline a challenge when you simply can't start a new game (ideally, people could not even challenge you when you are at the maximum). The game limitation, even if it was only 20 games, was one thing I found cool about the pawn status.
Now the question is whether BK would like to help curing BK addiction !
CryingLoser: I like your idea of forcing the above game limitation on players who time out, the problem being of course to determine a simple and fair algorithm for it.

mctrivia: Of course ! I thought all agreed that the most meaningful rating is the one at the end of the game, not at the beginning.