Greetings from Wikibooks! (and problems with transpositions) • page 1/1 • Lichess Feedback • lichess.org

Hazyshade edited

Hi lichess! I'm Hazyshade on lichess, but Chi Sigma on Wikibooks. I noticed recently that your opening explorer is pulling in content from the Chess Opening Theory wikibook.

It was me who spun the Chess Opening Theory wikibook off from the Chess wikibook in 2005. My contributions to the project have been on and off since then, but about 11% of the pages in Chess Opening Theory were created by me. We passed 1,600 pages this week!

I had to make certain decisions with the Wikibook for ease of navigation. One of them was around redirect pages for converged lines. It's best to illustrate with an example:

The wikibook has a page called

Chess Opening Theory/1. e4/1...e5/2. Nf3/2...Nc6/3. Bc4
and a page called
Chess Opening Theory/1. e4/1...e5/2. Nf3/2...Nc6/3. Bc4/3...Nf6

It also has a page called

Chess Opening Theory/1. e4/1...e5/2. Bc4/2...Nc6/3. Nf3 which redirects to the above move order.

But it doesn't have any redirect page called

Chess Opening Theory/1. e4/1...e5/2. Bc4/2...Nc6/3. Nf3/3...Nf6.

Because the two lines are considered to have converged upon White's third move, all child positions are titled using the 2. Nf3 move order. A reader wishing to find the "missing" position 1. e4/1...e5/2. Bc4/2...Nc6/3. Nf3/3...Nf6 can rewind the game one move at a time to find the redirect landing page, and then navigate forwards from there.

Although it might make sense for every position using the 2. Bc4 order to redirect to the corresponding position using the 2. Nf3 order, it would be impossible to make this a general policy. The average position 10 moves into the Hippopotamus Defence would need several thousand redirects pointing to it, representing the several thousand possible move orders by which it can be reached, and no-one would agree on which move order is the True Order.

The upshot of this is that some pages don't exist and never will exist however complete the book is.

Can the script be reconfigured to identify positions by their FEN, rather than by the moves?

If getting the FEN from the page diagrams is too difficult, I have a table with all the pages that exist and their corresponding FEN.

Best wishes,
Chi Sigma

Hi lichess! I'm Hazyshade on lichess, but Chi Sigma on Wikibooks. I noticed recently that your opening explorer is pulling in content from the Chess Opening Theory wikibook. It was me who spun the Chess Opening Theory wikibook off from the Chess wikibook in 2005. My contributions to the project have been on and off since then, but about 11% of the pages in Chess Opening Theory were created by me. We passed 1,600 pages this week! I had to make certain decisions with the Wikibook for ease of navigation. One of them was around redirect pages for converged lines. It's best to illustrate with an example: The wikibook has a page called * Chess Opening Theory/1. e4/1...e5/2. Nf3/2...Nc6/3. Bc4 and a page called * Chess Opening Theory/1. e4/1...e5/2. Nf3/2...Nc6/3. Bc4/3...Nf6 It also has a page called * Chess Opening Theory/1. e4/1...e5/2. Bc4/2...Nc6/3. Nf3 which redirects to the above move order. But it doesn't have any redirect page called * Chess Opening Theory/1. e4/1...e5/2. Bc4/2...Nc6/3. Nf3/3...Nf6. Because the two lines are considered to have converged upon White's third move, all child positions are titled using the 2. Nf3 move order. A reader wishing to find the "missing" position 1. e4/1...e5/2. Bc4/2...Nc6/3. Nf3/3...Nf6 can rewind the game one move at a time to find the redirect landing page, and then navigate forwards from there. Although it might make sense for every position using the 2. Bc4 order to redirect to the corresponding position using the 2. Nf3 order, it would be impossible to make this a general policy. The average position 10 moves into the Hippopotamus Defence would need several thousand redirects pointing to it, representing the several thousand possible move orders by which it can be reached, and no-one would agree on which move order is the True Order. The upshot of this is that some pages don't exist and never will exist however complete the book is. Can the script be reconfigured to identify positions by their FEN, rather than by the moves? If getting the FEN from the page diagrams is too difficult, I have a table with all the pages that exist and their corresponding FEN. Best wishes, Chi Sigma

benwerner

Maybe might make sense to create a GitHub issue (https://github.com/lichess-org/lila/issues) or join the Lichess Discord (https://discord.gg/lichess)? Those are generally better places for dev discussions.

Would it be possible to add redirects for all the FENs? If you have a table with all the FEN<->page correlations, it should be possible to create them automatically. Currently, the opening explorer just directly asks wikibooks whether a page titled with the given move order exists so not sure there's a nice way to do the translation from FEN to the right page on that side. Maybe might make sense to create a GitHub issue (https://github.com/lichess-org/lila/issues) or join the Lichess Discord (https://discord.gg/lichess)? Those are generally better places for dev discussions.