• By using a multi-site Site tag in every game users are aware of all the locations the tournament was played in when viewing the PGN of any single game from the tournament.

• The actual location of a specific game can be specified in the PGN via a GameSite tag.

• The location can also be specified with a PGN comment at the start of the game (which would be more viewable than a GameSite tag).

• <CG> can be flexible as to which location it displays on the game page, and use either the Site or GameSite tag, or both.

E.g. I think <Tab> and <Missy> would be happy if the GameSite tag were used, if present. I don't really deeply care about the last point, ie. which tag <CG> uses on the game page, though I have an opinion of course.

My concern is interoperability, which requires normalization.

But the real answer is that the <Date> tag isn't used for normalization, canonically.

The idea is pragmatic - the chess database software has to have a sliver of hope in order to group tournament games together from a giant soup pot full of games from who knows where/when. Ask yourself how you'd do it, sorting them out by hand.

You'd want the games to all have to have the same Site tag.

Really, using multi-site tag is a side benefit for people, the software just wants a consistent tag, and would be happy to use <[Site "ABC123xyz"]>, provided it was unique.

As for the date, I'm a big promoter of the bracket date concept - which might deserve its own PGN tag - though I haven't used it as yet. I do have a routine, pRD( P ), which automatically generates a Round/Date dictionary for a given tournament. It's quite useful, I agree.

You'll probably find a few of other years of the Tata Steel tournaments here on <CG> only use

<[Site "Wijk ann Zee NED"]>

in their PGN, ignoring the different locations of a couple of the rounds.

I'm fairly sure that's what <TWIC> and <365> do as well.

To me, that's a preferable option as compared to denormalizing the tournament - plus it avoids the concern <MissS> seems to express in his/her last post.

But you don't get to decide, that's the point. The algorithms have already been written, and are used daily across the entire world inside <ChessBase> and <SCID>.

We could use use a TournamentID tag in the PGN, if it would that simple. Why involve both Event/EventDate?

I've talked about this very issue with you before of course, similar to the discussion with <ckr> here:

https://zanchess.wordpress.com/2019...

<z> Aha ok thanks for the examples.

It seems to me that a [game site] tag would be a good tool.

Well, that's arguable.

<Why involve both Event/EventDate?>

<Steamed Colleagues>

Argh I just uploaded this pgn and I notice it has malformed syntax in the source tag: [Source "'Krustamīkla. Sahs. Brid‍s', December 5, 1936, p. 12 ] is missing the final " mark.

[Event "Helsinki 50-year Jubilee"]
[Site "Helsinki"]
[Date "1936.11.12"]
[EventDate "1936.11.10"]
[Round "3"]
[White "Stoltz, Gosta"]
[Black "Petrov, Vladimir"]
[Result "1-0"]
[Source "'Krustamīkla. Sahs. Brid‍s', December 5, 1936, p. 12 ]

1. e4 c5 2. Nf3 d6 3. d4 cxd4 4. Nxd4 Nf6 5. Nc3 e6 6. g3 d5 7. exd5 Nxd5 8. Bb5+ Bd7 9. Nxd5 exd5 10. Qe2+ Qe7 11. Bxd7+ Nxd7 12. Nb5 Qxe2+ 13. Kxe2 Rc8 14. Re1 Bc5 15. Kf1+ Kf8 16. Bf4 h6 17. Rad1 Nf6 18. c3 Rd8 19. Be5 a6 20. Bxf6 gxf6 21. Nd4 Kg7 22. Nf5+ Kg6 23. Ne7+ Bxe7 24. Rxe7 b5 25. Rd4 h5 26. a4 bxa4 27. Rxa4 Rb8 28. Re2 Rb6 29. Rd2 Rd8 30. Ke1 Rdb8 31. Ra2 Rb5 32. Kd1 a5 33. Kc1 Re8 34. Ra4 Re1+ 35. Kc2 Re4 36. Rad4 Rxd4 37. Rxd4 Kf5 38. Kb1 Kg6 39. Ka2 f5 40. Ka3 Kf6 41. h3 Kg5 42. b4 axb4+ 43. cxb4 Kf6 44. Ka4 Ke5 45. Kxb5 Kxd4 46. Kc6 Kc4 47. b5 d4 48. b6 d3 49. b7 d2 50. b8=Q d1=Q 51. Qb5+ Kc3 52. Qa5+ Kb2 53. Qxf5 Qd8 54. h4 Qe8+ 55. Kd6 Kc1 56. Qg5+ Kc2 57. Qe7 Qg8 58. Qe2+ Kc1 59. Ke7 Qg6 60. Qc4+ Kd1 61. Qxf7 Qg4 62. Qd5+ Kc2 63. Kf6 1-0

This pgn will be rejected automatically, is that correct?

I uploaded several other games from this event at the same time, with correct pgns. So I figure I would wait to see what happens when those other correct pgns get published.

Well, I would probably have stood corrected, except I did qualify it at less than a certainty.

Maybe I should have started with "Once upon a time..."

<I've put Chess on Tour site to all games 2014-2019.>

Probably not using "Chess on Tour" as the Site tag, but the actual cities!

Still <Tab>, I do wonder, did you do all those corrections one game at a time, or does <CG> allow corrections on a Round-by-Round basis?

Well, that's arguable. <<>>>

Of course, I misphrased it, and should have written:

<"But <we> don't get to decide, that's the point." <<>>>

But I didn't want to be accused of being a Socialist!

Too bad the bidding has ended, as this might be definitive:

https://new.uschess.org/news/blindf...

Then again, maybe not:

https://pbs.twimg.com/media/DfE6H_J...

Glad to see this player's name has been "modernized" too:

Kateryna Alexandrovna Lagno

Thanks, I was wondering... you are dangerously dedicated, that's fer sure!

The rule should be to write material in the intros so that no matter what year they get read they have the same meaning.