Your network blocks the Lichess assets!

lichess.org
Donate

My Problem with the Encyclopaedia of Chess Openings

Many people have criticized the ECO system, including Thibault, but you are not providing an alternative solution.

How should we cluster openings? Based on the moves themselves? The transpositions would kill that idea. Use different plans? How would that work? Even at plan level you can start by attacking one flank and then switching. Defenses can morph into attacks, gambits are sometimes stupid and sometimes the best engine continuation, mirrored positions make sense sometimes - so Black or White are less relevant, pawn structures?

Even the naming of the openings doesn't make sense to me. Why are some positions "defense", so named after a Black structure and plan, and some positions are White? Didn't both players get to that position, one with an opening and one with a defense? I would have a name for the White structure and a name for the Black one. And some named openings have multiple, sometimes very different, structures, but follow a similar plan.

I guess if we define a proper and universally accepted position distance metric, then we can use clustering algorithms. But is there such a metric. Or can it even be found?

I never cared for the ECO system, since, as you mentioned, it's useless. But finding a useful system is far more difficult that it might appear at first glance, as well.

Many people have criticized the ECO system, including Thibault, but you are not providing an alternative solution. How should we cluster openings? Based on the moves themselves? The transpositions would kill that idea. Use different plans? How would that work? Even at plan level you can start by attacking one flank and then switching. Defenses can morph into attacks, gambits are sometimes stupid and sometimes the best engine continuation, mirrored positions make sense sometimes - so Black or White are less relevant, pawn structures? Even the naming of the openings doesn't make sense to me. Why are some positions "defense", so named after a Black structure and plan, and some positions are White? Didn't both players get to that position, one with an opening and one with a defense? I would have a name for the White structure and a name for the Black one. And some named openings have multiple, sometimes very different, structures, but follow a similar plan. I guess if we define a proper and universally accepted position distance metric, then we can use clustering algorithms. But is there such a metric. Or can it even be found? I never cared for the ECO system, since, as you mentioned, it's useless. But finding a useful system is far more difficult that it might appear at first glance, as well.

@broiledlemming

The answer, seemingly, is nobody. Nobody seems to use the ECO for any of its intended purposes, and nobody ever seems to talk about it. If I am mistaken I'd love to have a conversation in the forums, but I just can't see why this is still such a prevalent piece of reference material in the chess community despite its seeming lack of popularity.

No one uses ECO simply because we have computers nowadays that can store inputted moves and allow exploration of opening lines. ECO was for when computers were not commonplace and we had to use books instead.

Also is ECO a 'prevalent piece of reference material'? People use computers these days, it ain't 1980. ECO isn't needed because we have computers.

@broiledlemming >The answer, seemingly, is nobody. Nobody seems to use the ECO for any of its intended purposes, and nobody ever seems to talk about it. If I am mistaken I'd love to have a conversation in the forums, but I just can't see why this is still such a prevalent piece of reference material in the chess community despite its seeming lack of popularity. No one uses ECO simply because we have computers nowadays that can store inputted moves and allow exploration of opening lines. ECO was for when computers were not commonplace and we had to use books instead. Also is ECO a 'prevalent piece of reference material'? People use computers these days, it ain't 1980. ECO isn't needed because we have computers.

Many people have criticized the ECO system, including Thibault, but you are not providing an alternative solution.
@TotalNoob69 said in #2:

A broader question here is, why do we even need opening categorization? I would love for somebody to enlighten me, as I'm probably being a little ignorant here, but I am failing to see any benefit to an opening categorization system. What purpose would such codes serve, even if they were well-made?

> Many people have criticized the ECO system, including Thibault, but you are not providing an alternative solution. @TotalNoob69 said in #2: A broader question here is, why do we even need opening categorization? I would love for somebody to enlighten me, as I'm probably being a little ignorant here, but I am failing to see any benefit to an opening categorization system. What purpose would such codes serve, even if they were well-made?

@myphoneisdying said in #4:

Many people have criticized the ECO system, including Thibault, but you are not providing an alternative solution.
@TotalNoob69 said in #2:

A broader question here is, why do we even need opening categorization? I would love for somebody to enlighten me, as I'm probably being a little ignorant here, but I am failing to see any benefit to an opening categorization system. What purpose would such codes serve, even if they were well-made?

The reason opening codes existed in the past is because people needed an index to explore lines. Like if you wanna find a place on a physical map, you need a code that directs you to that place (if you don't know where a place is located, a index may say that Country A is located at B5, where B5 can be found on a grid on the map).

If you didn't have categorization then it would be a random mess of lines and you wouldn't be able to explore an opening by looking in the index for where the opening line is.

But we don't need it anymore because we have computers. Just like we don't need physical maps with indexes because we have the internet where we can explore a digital map and type in a place. We can simply play moves on a chess board and it will gives us the opening theory continuations.

@myphoneisdying said in #4: > > Many people have criticized the ECO system, including Thibault, but you are not providing an alternative solution. > @TotalNoob69 said in #2: > > A broader question here is, why do we even need opening categorization? I would love for somebody to enlighten me, as I'm probably being a little ignorant here, but I am failing to see any benefit to an opening categorization system. What purpose would such codes serve, even if they were well-made? The reason opening codes existed in the past is because people needed an index to explore lines. Like if you wanna find a place on a physical map, you need a code that directs you to that place (if you don't know where a place is located, a index may say that Country A is located at B5, where B5 can be found on a grid on the map). If you didn't have categorization then it would be a random mess of lines and you wouldn't be able to explore an opening by looking in the index for where the opening line is. But we don't need it anymore because we have computers. Just like we don't need physical maps with indexes because we have the internet where we can explore a digital map and type in a place. We can simply play moves on a chess board and it will gives us the opening theory continuations.

@myphoneisdying said in #4:

Many people have criticized the ECO system, including Thibault, but you are not providing an alternative solution.
@TotalNoob69 said in #2:

A broader question here is, why do we even need opening categorization? I would love for somebody to enlighten me, as I'm probably being a little ignorant here, but I am failing to see any benefit to an opening categorization system. What purpose would such codes serve, even if they were well-made?

If you ask me, anything large enough should have a way of splitting it into more manageable chunks, but that's the software developer in me talking. There is no way to systematically learn chess if you cannot categorize its elements, and that includes openings.

From a more advanced standpoint (that I am not sure I can do justice to), if we would cluster openings into thematic categories, then they would be not only easier to learn, but teaching the right chess ideas by showing common motifs rather than rote memorization of moves. But I am a noob, so what do I know?

@myphoneisdying said in #4: > > Many people have criticized the ECO system, including Thibault, but you are not providing an alternative solution. > @TotalNoob69 said in #2: > > A broader question here is, why do we even need opening categorization? I would love for somebody to enlighten me, as I'm probably being a little ignorant here, but I am failing to see any benefit to an opening categorization system. What purpose would such codes serve, even if they were well-made? If you ask me, anything large enough should have a way of splitting it into more manageable chunks, but that's the software developer in me talking. There is no way to systematically learn chess if you cannot categorize its elements, and that includes openings. From a more advanced standpoint (that I am not sure I can do justice to), if we would cluster openings into thematic categories, then they would be not only easier to learn, but teaching the right chess ideas by showing common motifs rather than rote memorization of moves. But I am a noob, so what do I know?

Why only 5 letters? That is very weird to me, when you have all of the alphabet at your disposal. Just a random thought.

Why only 5 letters? That is very weird to me, when you have all of the alphabet at your disposal. Just a random thought.

@RuyLopez1000 said in #3:

@broiledlemming

The answer, seemingly, is nobody. Nobody seems to use the ECO for any of its intended purposes, and nobody ever seems to talk about it. If I am mistaken I'd love to have a conversation in the forums, but I just can't see why this is still such a prevalent piece of reference material in the chess community despite its seeming lack of popularity.

No one uses ECO simply because we have computers nowadays that can store inputted moves and allow exploration of opening lines. ECO was for when computers were not commonplace and we had to use books instead.

Also is ECO a 'prevalent piece of reference material'? People use computers these days, it ain't 1980. ECO isn't needed because we have computers.

I agree as far as actual use by humans goes, I'm more so talking here about how it seems to be implemented into more or less every piece of computer chess software I've ever interacted with. Both Lichess and chess.com implement these in their opening explorers and analysis boards, as does Scid vs PC and other local software that I've seen and used. I guess my point was more so the fact that computers, as you mentioned, have kind of made the ECO obsolete, and I can't wrap my head around why support for it is so prevalent across software today as a result.

@RuyLopez1000 said in #3: > @broiledlemming > > >The answer, seemingly, is nobody. Nobody seems to use the ECO for any of its intended purposes, and nobody ever seems to talk about it. If I am mistaken I'd love to have a conversation in the forums, but I just can't see why this is still such a prevalent piece of reference material in the chess community despite its seeming lack of popularity. > > No one uses ECO simply because we have computers nowadays that can store inputted moves and allow exploration of opening lines. ECO was for when computers were not commonplace and we had to use books instead. > > Also is ECO a 'prevalent piece of reference material'? People use computers these days, it ain't 1980. ECO isn't needed because we have computers. I agree as far as actual use by humans goes, I'm more so talking here about how it seems to be implemented into more or less every piece of computer chess software I've ever interacted with. Both Lichess and chess.com implement these in their opening explorers and analysis boards, as does Scid vs PC and other local software that I've seen and used. I guess my point was more so the fact that computers, as you mentioned, have kind of made the ECO obsolete, and I can't wrap my head around why support for it is so prevalent across software today as a result.

@TotalNoob69 said in #6:

Many people have criticized the ECO system, including Thibault, but you are not providing an alternative solution.
@TotalNoob69 said in #2:

A broader question here is, why do we even need opening categorization? I would love for somebody to enlighten me, as I'm probably being a little ignorant here, but I am failing to see any benefit to an opening categorization system. What purpose would such codes serve, even if they were well-made?

If you ask me, anything large enough should have a way of splitting it into more manageable chunks, but that's the software developer in me talking. There is no way to systematically learn chess if you cannot categorize its elements, and that includes openings.

From a more advanced standpoint (that I am not sure I can do justice to), if we would cluster openings into thematic categories, then they would be not only easier to learn, but teaching the right chess ideas by showing common motifs rather than rote memorization of moves. But I am a noob, so what do I know?

This is the exact kind of concept I've been toying with. I don't think these sorts of things will ever be perfect by any means, but the idea of use as a pedagogical tool has had me interested for a while. More of just a personal tinkering project, but I think some sort of pseudo-systematic way to classify position types could be of use for people trying to understand how openings and variations relate to each other, rather than specific lines or positions.

@TotalNoob69 said in #6: > > > Many people have criticized the ECO system, including Thibault, but you are not providing an alternative solution. > > @TotalNoob69 said in #2: > > > > A broader question here is, why do we even need opening categorization? I would love for somebody to enlighten me, as I'm probably being a little ignorant here, but I am failing to see any benefit to an opening categorization system. What purpose would such codes serve, even if they were well-made? > > If you ask me, anything large enough should have a way of splitting it into more manageable chunks, but that's the software developer in me talking. There is no way to systematically learn chess if you cannot categorize its elements, and that includes openings. > > From a more advanced standpoint (that I am not sure I can do justice to), if we would cluster openings into thematic categories, then they would be not only easier to learn, but teaching the right chess ideas by showing common motifs rather than rote memorization of moves. But I am a noob, so what do I know? This is the exact kind of concept I've been toying with. I don't think these sorts of things will ever be perfect by any means, but the idea of use as a pedagogical tool has had me interested for a while. More of just a personal tinkering project, but I think some sort of pseudo-systematic way to classify position types could be of use for people trying to understand how openings and variations relate to each other, rather than specific lines or positions.

@Jisu101 said in #7:

Why only 5 letters? That is very weird to me, when you have all of the alphabet at your disposal. Just a random thought.

It's very simple. They published a book in 5 volumes, and each volume gets a letter.

This is also why you get the apparent misclassification of some openings that the blog post laments about, they were trying to keep the size of all volumes roughly equal.

Of course, these limitations make the system much worse nowadays, when nobody is opening a physical ECO volume to look things up.

@myphoneisdying said in #4:

A broader question here is, why do we even need opening categorization? I would love for somebody to enlighten me, as I'm probably being a little ignorant here, but I am failing to see any benefit to an opening categorization system. What purpose would such codes serve, even if they were well-made?

Something a good classification system could be interesting for would be analyzing player performance by opening family, comparing the frequency at which a player or groups of players play various opening families, etc.

Although you don't specifically need codes for this, you need some systematic classification, and why not use codes?

Lichess insights can already show some stats by opening family/variation, but this could be greatly improved.

@Jisu101 said in #7: > Why only 5 letters? That is very weird to me, when you have all of the alphabet at your disposal. Just a random thought. It's very simple. They published a book in 5 volumes, and each volume gets a letter. This is also why you get the apparent misclassification of some openings that the blog post laments about, they were trying to keep the size of all volumes roughly equal. Of course, these limitations make the system much worse nowadays, when nobody is opening a physical ECO volume to look things up. @myphoneisdying said in #4: > A broader question here is, why do we even need opening categorization? I would love for somebody to enlighten me, as I'm probably being a little ignorant here, but I am failing to see any benefit to an opening categorization system. What purpose would such codes serve, even if they were well-made? Something a good classification system could be interesting for would be analyzing player performance by opening family, comparing the frequency at which a player or groups of players play various opening families, etc. Although you don't specifically need codes for this, you need some systematic classification, and why not use codes? Lichess insights can already show some stats by opening family/variation, but this could be greatly improved.