I expect it would be easy with db access. 
It’s possible but difficult: this tournament has 10038 games. With just one query I get the full list of games with result.
I can retrieve players list 100 at a time, so it’s another 23 queries (2233 players).
But if I wanted to check every single game I should be doing 10038 queries.
If we use 3 seconds to send a query, get a response, process it and store in the db and wait a little before sending a new query (to avoid being kicked off), it would take 8 hours to get all games.
To save some queries, I could check only those games that weren’t finished in the previous check. That means that I should also read my db to get the list of previously unfinished games in order to make the queries.
Now it’s way simpler: I just do the same request every time and store it as it is in my db and the whole process takes just few minutes.
4 Likes
Thanks for considering my musings on a new data visualization. If there was a stored procedure on OGS’s back end that did the work and just spit out the info through a new api that would save a lot of work elsewhere. Doubt creating the api HistogramOfStillRunningGamesInTournament(TournamentNum) is on the top of the OGS improvements list
1 Like
New picture on tuesday. Perhaps monday wasn’t the best choice! 
Changes since last update:
- 95 games completed
- 8 players disqualified
- one resigned
- timuzhti made it to the top of the leaderboard
- 57 players completed all of their games, 3 of them with 9 victories
- 5 groups completed all of their games
5 Likes
Programmers stuff:
yesterday I tweaked my tools to get players list from termination-api instead of V1 API, since V1 doesn’t update correctly player’s rankings.
Maybe there’s a bug also in termination-api, I described it here: https://forums.online-go.com/t/what-happened-to-tournaments-players-ranking/34054/25
We’ll see.
I just wanted to mention that I couldn’t find in the termination-api something like the “details” and “rounds” pages that I download from V1:
https://online-go.com/api/v1/tournaments/59567
https://online-go.com/api/v1/tournaments/59567/rounds
There are some useful informations such as the current round, the starting date of the tournament and the full list of games with outcome.
If someone knows more about the termination-api and how to get these data from there, please tell me. Otherwise I’ll just continue with the V1 source.
Martin and RanRan’s flying knife game has developed into an unusual variation with the atari (40), which usually extends.
This variation was actually played iin the professional game Liang–Nie '90 Liang Weitang (7p) vs. Nie Weiping (9p) | Waltheri's go pattern search although some contextual stones were present.
Some early midgame discussion happening in [GoFio] ManuelGA 3k – banane 1k. What looks to have been a fairly peaceful opening is now spicing up a little~
Chat regular FinrodFelagund 2d is down to his last game, having just played this nice connection tesuji against imakyutpie 1d.
MooToYouToo seems to have taken an advantage against Animiral, although the game is still continuing.
Baklan – le_4tc has kicked off violently.
2 Likes
Yes, finally famous enough to be noticed in this thread!
6 Likes
You were always famous enough to be noticed in our hearts
3 Likes
New picture on (late) monday.
Changes since last update:
- 54 games completed
- 7 players disqualified
- one resigned
- 38 players completed all of their games, 4 of them with 9 victories
- 8 players have all their games still in progress (they were 11 last tuesday)
- 5 groups completed all of their games
- only two groups have less than 50% games completed
Because of the bug in (where I read) rankings, I have some weird chart to share.
They don’t show an actual bug, since I am the only fool that reads those numbers. 
Anyway, here is some fun stuff:
People becoming suddenly shodan and back.
An otherwise fairly stable chart had some shakes.
I wonder what will happen now: will I change my tools, looking for some better source? Or will the bumps revert to normality?
6 Likes
Aren’t the bumps a normal effect of the recent ranking updates? Anyway congrats on 666th post, @S_Alexander did you think your baby tournament would reach that far as to reach the number of the beast in the forums?
3 Likes
This is only the beginning!
3 Likes
It is!
The yellow curve moving to the right is that.
The spike at 1d is just a glitch in the source I read.
That could also be viewed as a consequence of the recent ranking update because something happened under the surface and the variables I was reading before are no longer updated or reliable.
I don’t know if they will ever be fixed, so I don’t actually know what I should do now. All my charts involving rankings are someway broken.
1 Like
I say wait a bit and see if it stays unreliable or if it seems alright. It could be some under-the-hood changes, a glitch, or perhaps an eventual consequence of the ranking updates as the new points get redistributed?
I’m not sure how much of what I just said is possible but it’s never a good idea to throw your work down the drain if it might not be 100% broken ^^
1 Like
I had some issues in the last days and my toy seemed broken. Out of memory errors. Database slowing down. And I don’t even know why: pieces that worked for months now are experiencing new problems. I managed somehow to download data and take my picture, but still don’t know how to put things together.
Whatever!
Changes since last update:
- 107 games completed (wow!)
- 13 players disqualified
- 3 resigned
- 58 players completed all of their games, one of them with 9 victories
- one less player with all of his games still in progress , now there are 7
- 3 groups completed all of their games
- only one group has less than 50% games completed
6 Likes
In my search of old records in the 1880s, I stumbled upon the term 郵便碁 (play via mail), and one of them is supposed to be the first-ever play by mail game in history played between June 1883 till Nov 1884. I feel it could be called the oldest long correspondence game.
It was not finished when this was published and still on-going according to the description, and sadly most likely never was finished because the white player 村瀨秀甫 (Murase Shūho, or more widely known later as Honinbo Shuho) died the next year in 1885.
And there was a second play by mail game played by the same black player 三戶與彰 at the same time between June 1883 till July 1885. And it was finished (white resigned)
It is interesting to know that players have been dabbling with simultaneous correspondence games for more than a century already.
8 Likes
In our western culture it’s typical the tale of people playing chess by mail, especially people living in isolated places such as lighthouses. 
It’s the first time I hear about mail-go. It’s quite obvious that it happened, but I never heard before.
1 Like
There was even a 日本郵便碁愛好会 (Japanese play by mail Go club) existed since the 1960s. But didn’t know that it went so far back in time in the 19th century.
I even read there were players charge to play teaching play by mail games as well, even in the 19th century.
また、明治期に長野県内に滞在して囲碁を指導した鈴木善之助 （すすきぜんのすけ）、１８２５（文政8年）？～１８９９年（明治３２年）との棋譜も収められています。郵便代は全部で８円くらいかかり、高段者への謝礼は１０円、一局に数年かかる上、費用もかさむため、それほど流行しなかったそうです。
One play-by-mail game charged for 8 to 10 円 per game (the value of 1 円 at Meiji era was pretty high, probably equal to hundreds of US dollar today)
Seems like correspondence games rarely changed throughout the ages.
6 Likes