Comments/Ratings for a Single Item

One of the improvements that this website could have is to make possible that any contributor could delete any of his/her posts.

Yes, indeed, it would.  I hate staring at posts removed but still indexed.  It irks me.  And as to the website being dead, I think it just about is.  Perhaps a design contest might intrigue ex-varianteers?

If you are interested in volunteering, please contact David Howe. 

I agree the site needs more conversation. If you go over the comments section you'll see there is a slow churn of members, or members designing and/or commenting. Look closer, and you'll see a core group several years ago that talked with each other, ran contests and group designs, and essentially had an acknowledged leader in Ralph Betza. A number of others have served as lesser activators for the site. But the main thing is that several people participated regularly or semi-regularly and there were a handful of interesting things, discussions, recent designs, contests... going on at more or less the same time. So a Ralph Betza is not required to have a good site* but group participation is. 

Who are the members of the group? Who will participate and follow through? 

A design contest is always fun - who will judge it? Based on my experience, it turns out you need several judges. You could have everyone in the contest judge all the entries except their own, which would relieve a lot of stress and pressure on the erstwhile judges and allow people to participate more fully. 

You can also run a set time tournament, where everyone starts with say 3 - 5 months of grace time, and nothing else. No time is ever added. This guarantees the tournament will end in 6 - 10 months. All games are played at once. Each player is assigned black in half the games at start, and white in the other half. Once the games start, all players must win, lose, draw, or time out in every game. It works, and prevents tournaments from going on for years. It works for small numbers. However, it is highly unlikely we would get so many participants that it would break the format.

Discuss this here, see who will commit to what, and we shall see what happens, if anything. I would not mind seeing this site a little livelier. Enjoy!

*Thank goodness, because Ralph is quite a hard act to follow, being a master in both chess and variants as well as a good writer, someone who could explain variants and analyze them for the rest of us.

Further to Joe Joyce's comments on my problems signing in, wondering if this is connected  to linking my Noble Wings rules page to the preset - though that looks ok, a solid block of text, with no punctuation - rules buried in the middle, now shows under my game page.
Problems signing in & sending first started soon after starting playing Joe's updated (tally board) Command & Manouvere..

What you say, Joe, is true; there is a core group of persons talking through time slowly, calmly on a wide range of topics. My concern, my worry [almost desperation] comes from to see the very, very, very... small number of players that at present are playing in Game Courier. What happens? Why the inventors do not play their own games?

Instead of contact to David Howe, I would rather that he or Hans update the information about how one can help to run this site. Thanks for the suggestion.

I've added a random selection option to our query script.  From the search form on the Topic/Main Index page there is a checkbox for this option.  (I don't claim this is very uniformly random.  Let me know if you have any problems.)

Should we also add a random search link to the menu?  Random game page, random Java applet, random something else?

Maybe it might help to have one submenu for "Random", to make the list of possible choices for random of what kind you want.

Also, I have added the random item checkbox to the Advanced Search menu.

Sorry about the outage today folks.  Everything should be fine now, but if you run into any problems please email us.

Moves at a Glance?

For the pages on my XBoard website that describe the various Chess variants it supports, I have developed a nice interactive gimmick to show the move of the various piece types, which I call 'Moves at a Glance'. It is basically a move diagram that shows the participating piece types positioned close to the center of a chess board (so they have room to move in all directions). When the user clicks on one of the pieces, the squares it can move to are highlighted with dots of various colors, according to a legend besides the board. So it is basically equivalent to a move diagram for a large number of piece types, collapsed into one, the user selecting the piece for which he wants to see the moves. But in case of special moves such as e.p. capture or castling it would also be able to show the side effects, such as moving the Rook and King when you click on one of the Rooks in the diagram positioned to castle.

For an example, see for instance the description of Chess with Different Armies.

I wondered if it would be possible to implement something similar on chessvariants.org, and make it available to contributors through the template they use for posting their own game descriptions.

The way it works on my XBoard website is that the diagram is just an html table that contains a piece image in each cell, and defines 'onmouseDown' and 'onmouseUp' events as routines in a simple JavaScript program associated with the page, which then apply and remove the board markers in the table. For me each cell calls the same event routine, with the cell coordinates as parameters. So that the routine must be aware of what stands where, and is thus specific for each rule page, or shared by just a few rule pages for variants that share many pieces.

But for general use this could of course be altered, and the contents of the cell, or more accurately, the desired pattern of board markers, could be passed as an argument to the 'onmouse' event handlers. For instance as a string with the Betza description of the piece in that square. (That seems simpler than a FEN to specify which squares should be marked in which color by hand, although the JavaScript 'engine' could be made to accept both formats.)

The poster of a new variant would then only have to supply a list of (squareCoords, pieceImage, BetzaString) combinations, from which the posting pre-processor would then construct the html table (for a 10x10 board, say, although this could also be made user configurable) with the provided info in each cell. The piece image could just be the name of an alfaerie gif file present in a standard directory on this site.

Would it be a good idea to have such a feature here? If so, would it be feasible?

I like the interactive diagram.  For use on this site, I would be a little worried about compatibility; for instance, your script doesn't run on my tablet, so I'm left with just the static arrangement of pieces.

> for instance, your script doesn't run on my tablet

Oh, I wasn't aware of that. But I just tried it on my wife's Galaxy Tab S, and indeed something strange happens there. It seems to ignore short taps on the squares, and only triggers on long presses. But then it doesn't only show the moves, but in addition a system popup appears that asks me what I want to do with the image.

I am sure this must be fixable, however. JavaScript is so standard on the web that there must be ways to make the page work on both PCs and tablets.

The following 'moves at a glance' should also work on touch-screens now:

http://hgm.nubati.net/rules/CWDAx.html

At least, it did work on my wife's Android tablet now. Could you please test if it works on yours too? The problem apparently was that a 'touchStart' event does not also generate a 'mouseDown' event (which the script was using) when the touch point is not stationary, and it is in practice almost impossible to touch without moving a little. I now have the script also react to the touchStart itself. 

That also gave me the opportunity to make it work with a touch screen slightly different from with a mouse: when touching you obstruct your own view, so it was pretty annoying that the move markers would disappear as soon as you end the touch (as they do on mouseUp), as you would never have a good view of them. So I made it 'sticky' in that case: first touch of a piece shows its moves, second touch on that same piece then clears them (so you can see all pieces again), touch on another piece also clears any displayed moves and shows those of the new piece.

Yes, it works on my tablet now (also Android).

I'm not very familiar with how to make this available to contributors for inclusion on their own pages.  We had (and still do, but they now are very seldom used) the scripts ffen2diag and such;  can we just upload your script to a convenient place on our site and have contributors include from there?  (I'll be out of town for about a week and probably won't get around to this until I return; maybe just email me about implementation details?)

After I replied to the forwarded email, I found your discussion here about moves at a glance. It's something different than what I had imagined. I had imagined something showing the moves made in a game rather than move diagrams. I have sometimes thought of having Game Courier display legal moves, but the main stumbling block was that legal movement is defined server-side, and JavaScript is client-side. But I have now thought of a way it could be done. I could write a GAME Code subroutine that scans the board for pieces, calculates a list of legal moves for each piece, and writes the lists as a 2D JavaScript array. It could be used in the post-game code in place of a stalemate subroutine and indicate whether there is stalemate by returning the number of legal moves.

Does this page have a Facebook page?

No, we don't have a facebook page.  (Should we?  What purpose would it serve?)

It would require a moderator, so I understand why not. Promoting the site is the best reason I can think of, I've seen others try to steer people here on chess.com but there is also alot of trolls on chess.com

The Janggi hyperlink on the homepage is broken and returns a 404 error.

The Janggi hyperlink on the 404 error page does work, however.

Thanks. That's now fixed.