I hope the site comes up with a solution, as until then, it looks like the 50 move rule applies, and will someone stop that elephant from trumpeting and doing all sorts of other things in that room.
I just realized even the 50 move rule may not suffice:
Pawns can move and still no way to win.
The 50 move rule suffices. Eventually. In that position it may be 299 moves later if both players try to prolong it.
So you and all those who think this thread is not important will not come around here to discuss not important things and waste your time, instead you will go somewhere else to discuss the important things of mankind, then Auf Wiedersehen
Ya sure about that? Here's an example where even a capture does not make the position winnable. Could theoretically be prolonged a multitude of 50 every time white moves a pawn 1 square, if white was the one trying to win on time.
Yes, and eventually there are no pawn moves left and the game is drawn after 50 move moves.
I agree. A virtual arbiter should be available upon request for the dead positions and other considerations as well. This has to be ruled like in tennis when the players request the hawk. If the outcome has no grounds there should be some penalty for the player. New rules would be necessary to make the online game really fair and a arbiter avatar for certain situations is at the time missing, otherwise where is the fair play?. Chess.com should make these tools available specially when there are people paying for a service which encourages fair play.
Full marks to ArtemKozirev for persistence and concern for fair play. The Virtual Arbiter would face the same difficulties as those expressed by Elroch.
Is there any chance this measure will be implemented?
If so, it will be a notable first for Chess.com
I think we're forgetting something here: This a free site. Premium members are welcome to pay for our virtual arbiter.
"How many things would you attempt if you knew you could not fail?" Robert Frost.
One day, long time ago the first chess programs started to be available, nothing that a 2000 elo player could not defeat, later the programs improved and challenged and defeated the best players, now the engines are virtually unbeatable to a human. Today there is no chess arbiter program or engine except some ideas in this thread. The need is there .... later there will be good engines until the avatar can efficiently do the task in an unmatchable way. Rome was not built in one day. All ideas count, how this will be implemented is another story. "Imagination is better than knowledge" Albert Einstein and "The only thing that does not exist is the one which has not yet been created" Jorge L. Borges. An idea even if in a dream, already exists.
I think we're forgetting something here: This a free site. Premium members are welcome to pay for our virtual arbiter.
I was premium for a year, but then canceled it once I did enough tactics puzzles. I didn't see any virtual arbiter.
I meant the feature currently being proposed on this thread.
Again, I don't think having more JavaScript scripts running on our browsers is a good idea. For instance, some devices are have their hands full with chess sets and their smooth transitions. No too long ago, I was in the middle of a game and it froze for a good minute before I could play again.
So, seeing how we're discussing solutions anyway, I suggest the virtual arbiter should run on the servers. However, that's exactly the issue. This site is a free site for most members, and the virtual arbiter would drain server resources which actually costs a lot of money. So, premium members will have to foot the bill... for all of us.
Can someone explain to me why the would have to be a 'virtual' arbiter?
Why can't the arbiter be human?
Though, I agree with everyone, @ArtemKozirev included. Understanding dead positions is progress for the world of chess. If we all understood them well enough then maybe we could quickly agree to a draw rather postponing the inevitable.
Doesn't have to be virtual, One human could do the job. How many dud calls he would get is anyone's guess.
Full marks to ArtemKozirev for persistence and concern for fair play. The Virtual Arbiter would face the same difficulties as those expressed by Elroch.
Is there any chance this measure will be implemented?
If so, it will be a notable first for Chess.com
I am shocked Elroch's engine
... the latest Stockfish, the strongest conventional chess engine.
didn't recognize it to be a draw, remember these positions are already drawn, not "drawish", the game has ended if you apply the rules of chess. It just doesn't fall into stalemate or insufficient material, the site just has to program a way to instantly recognize the 6th way of drawing like this.
So easy to say, but not at all straightforward to do. All engines "know" is what moves are legal and a rough way of evaluating positions without analysis.
Similar to a tablebase, every dead position could be computed and registered into the system for instant recognition.
lol. You have a convenient planet for the database?
I mean if they have tablebases for every possible position with 7 pieces, the #of dead positions is far smaller, so they could take that approach.
They surely aren't, and even generating a 7 piece tablebase (5 pieces and pawns + 2 kings) required a large computational resources - a supercomputer with 78,660 cores - and needs 140 Tb to store.
Generating an 8-piece tablebase would presently require one of the 10 fastest supercomputers in the world, and these cost about $100 million and have better things to do with their time. 
What if white moves a pawn every 49 moves?
As I said, white runs out of pawn moves.
How do you figure there are less possible positions with 7 pieces than dead positions?
Just a rough judgement. Needs more thought to be reliable, but very difficult to calculate. A starting point is to consider the number of pawn structures where no capture, promotion or mate is ever going to be possible. There are clearly a very large number of these. Then you can add dead pieces to some of them. There are some very exotic examples involving multiple pieces blocking each other, but of course the squares for these pieces are not numerous in each case.
(And it's actually "fewer").
They could consider placing a button for players to press to check for dead position. That way it will not take much computer/server resourses and it will be similar to over the board tournaments where you will call the arbiter when such a position happens. It will probably not be able to solve 100% of the cases since we know from real tournaments there has been cases where such posistion has given rise to much debate if it was truely dead or not. (Misuse of the botton should ofcourse give a penalty).
The other solution is to do nothing and let everyone play after the curent rules and once in a while take a bad beating. As long as the rules are know I dont see a problem with that approach and I personally prefer clear rules compared to an unknown computer algorith is able to do its job or not.
The answer is quite simple.
Chess has rules and laws.
There are FIVE ways to draw a game:
Stalemate
By Agreement
50/75 move rule
Insufficient material
Threefold repetition
You claim there are 6 - Dead Position and that this site and others should recognize it.
WRONG.
Get the rules and laws of chess changed and then maybe we can talk, but for now, your DEAD POSITION will only be a draw through the 50/75 move rule or threefold repetition (eventually).
Even top engines can't understand positions like this. I had a wonderful blitz game where I trapped an opponent's bishop and analysing it afterwards the engine got it wrong with enormous amounts of computing time. It thought the 100% trapped bishop (like in the first diagram) was worth 3 pawns.