< Earlier Kibitzing · PAGE 293 OF 293 ·
|Aug-19-14|| ||rookhouse: <Tabanus> Again, I NEVER said I did not agree with citing sources. I literally have hundreds of sourced clippings on Manchester 1890.|
I just asked a simple, logical question that was based on the fact that the majority of the game collections on CG do not have any cited sources.
I have over 300 historical chess books in my possession, over 50 bookmarked websites that I use for research, 8 different state library archives at my disposal, and I have spent two entire days at the John G. White Collection in Cleveland gathering historical information on chess.
Sourced material is NOT a problem for me. Some of the users on here (not you) really need to think about the decaffeinated version of their coffee ..... seriously.
|Aug-19-14|| ||Tabanus: <rookhouse> Then you are more than welcome here! We have all the technical facilities, but quite few biographers.|
Probably we should start one day working on the already voted in "Events with Missing Information", but ... there are thousand other events.
|Aug-19-14|| ||OhioChessFan: <This fine collection awaiting one vote for promotion: Game Collection: Petrosian - Portisch Candidates Quarterfinal '74>|
I like "disconcerting" but "off-putting" with a link to Botvinnink would be more interesting.
|Aug-19-14|| ||rookhouse: Thanks <Tabanus>! I agree, there are several out there that would take minimal effort to complete.|
Example: <suenteus po 147> has a nice intro and game collection for New York 1948-49. He references my old website for his source and the games do not have dates.
I have the original tournament book, so he could site that as the source and I could go through and input the game dates for him. It would then be ready for nomination.
|Aug-19-14|| ||rookhouse: Speaking of <suenteus po 147>, his forum appears to be down and he has not posted in nearly a month. |
I was going to reach out to him about New York 1948. Anyone been in contact with him?
|Aug-19-14|| ||zanzibar: <RE: Tromso Open Olympiad (2014)>|
I'm still working on this one tournament. I've found 97 players where the <CG> player matched up is lacking a FIDE id.
Here's a sample of eight of them, where N_games > 11 (and so the player was pre-existing, and wasn't added to <CG> from this tournament):
< 53910 / $ 0 Joao Simoes () -- -- Y 2275 / 28
88656 / $ 0 Jorge Sanchez (1982.00.00) PA -- Y 2207 / 35
127224 / $ 0 Karl Johan Ribbegren (1955.00.00) MC -- N 2211 / 46
119003 / $ 0 Miguel Jose Maconi Mucongoma (1981.00.00) MZ -- Y 2194 / 18
99443 / $ 0 Moakofi Notha (1984.00.00) BW -- N 2152 / 20
127343 / $ 0 Moulaye Brahim Hemam () -- -- Y 2070 / 15
103406 / $ 0 Ricardo Medina Viana (1962.00.00) HN -- Y 2063 / 25
127245 / $ 0 Suradj Hanoeman (1987.08.17) SR -- Y 2030 / 14>
(CG id / $ FIDE id Name dob cnty fed bio? rating / N_games)
There are 89 players which have presumably been added to <CG> from Tromso Open, but without a FIDE id. Which is unfortunate, especially given the fact that the original FIDE PGN had the FIDE id for all players.
Of course it used a tag <CG> wasn't expecting and so was missed for the first couple of rounds. Too late, for the "damage" was already done.
<The question still remains - how is <CG> going to clean up all this?>
Here's a link to all the data showing the right FIDE card data for each missed <CG> entry:
Here's a sample:
<FIDE vs CG - Tromso 2014 Open Olympiad (CG players w/o FIDE id)
6600875 Fuentes, Adrian (1998) PAN 2105 / 0
150994 Adrian Fuentes () -- -- N 150994 / $0 (2105/9)
7700750 Winter Atwell, Adrian (1985) TTO CM 2110 / 0
150743 Adrian Winter Atwell () -- -- N 150743 / $0 (2110/9)>
( FIDE id Name (dob) fed titles rating / N_games (last rating period) ... CG info )
<Can CG use this info to update its database?>
|Aug-19-14|| ||zanzibar: More on <Tromso Open olm 2014>|
Here is a merge list, i.e. basically the same player's games split over two <CG> players.
< 80944 / $ 3100090 Orlando Santana (1960.12.23) PR -- N 2255 / 13
151002 / $ 0 Orlando Santana Otero () -- -- N 2125 / 1
* * *
17258 / $ 3100260 Alejandro Montalvo (1968.00.00) PR -- Y 2299 / 50
150729 / $ 0 Alejandro Montalvo Rosario () -- -- N 2296 / 2
And these ones, where we again have a duplicate FIDE id in the data set:
< 53107 / $ 3300145 Johnny Cueto Chajtur (1972.00.00) BO -- Y 2372 / 73
150831 / $ 3300145 Jonny Cueto (1972.00.00) BO -- Y 2283 / 2
* * *
141859 / $ 3902048 Andres Guerrero (1990.00.00) VE -- N 2323 / 15
150770 / $ 3902048 Andres Guerrero Vargas (1990.00.00) VE -- Y 2318 / 2
* * *
21969 / $ 4000510 Elvira Sakhatova () KZ -- Y 2330 / 19
55540 / $ 4000510 Elvira Berend (1965.09.19) LU -- Y 2353 / 80
* * *
103873 / $ 7800592 Muhammad Younus Younus (1968.07.15) PK -- N 0 / 12
150794 / $ 7800592 Younus Muhammad (1968.00.00) PK -- N 0 / 2
* * *
127350 / $ 8500444 Olamide Ajibowo (1977.00.00) NG -- Y 2201 / 17
150728 / $ 8500444 Olamide Patrick Ajibowo (1977.00.00) NG -- Y 2168 / 2
* * *
59773 / $ 19900058 Gueye Gorgui (1966.00.00) SN -- N 0 / 13
150826 / $ 19900058 Elhadji Gorgui Gueye (1966.00.00) SN -- N 0 / 2
One reason all these seems so laborious, is that the original FIDE set was relatively clean - if the FIDE ids were used.
But the <CG> data entry really should flag new user entries which have the same FIDE id. If the FIDE id was added after the fact, that should be flagged as well.
I am really worried about how the actually movelists match up. That's the next phase.
|Aug-19-14|| ||zanzibar: I guess there are 28 duplicated player entries that were found here.|
|Aug-19-14|| ||chessgames.com: Zanzibar, I read your blog and took note of this comment <I like the programming part, at least it appears so since I just did it. But I’m certainly not intending on submitting 100 correction slips, not at my pay scale – but my blessings to whoever does!>|
Please--NOBODY do that. It isn't going to make this easier at all, it just adds an extra layer of work.
It seems like you've already done the heavy lifting. When your program outputs something like this
< 11502339 Nazih Ziyadah (0000) PLE 0 / 0
150747 Ziyadah Nazih () -- -- N 150747 / $0 (0/4)>
That means that we have a player #150747 (Ziyadah Nazih) who does not have a FIDE number, but you've figured out that the FIDE number is 11502339? Does it also mean that the proper name should be "Nazih Ziyadah" and we have it flipped?
If that's the case, things are not as bleak as you make it sound. Software could be written to run through your software's output and modify our player file, quite easily. It would be one-time-use software but sure easier than 100 correction slips.
In cases where there are duplicate players that need merging, well, we'll cross that bridge next. And there, correction slips are warranted, if there's no better way.
All of this needs to be set against the backdrop of who these players are. It's not like we ended up with two Aronians; we have an unrated Palestinian player without all of his data filled in. Not ideal, but not the end of the world. At least one of his games doesn't meet our data quality standards and is slated for deletion soon. (We can't do that while the event is in progress because the games would keep coming back to us.) A few of these players from smaller nations may lose all of their games, and vanish altogether, thus solving the problem a different way.
Of course some of these players are very strong indeed, so don't think I'm trying to dismiss them all. But there's a tendency to find low-rated/unrated players in your list, due to the fact that the strongest players are the ones who would have been more easily identified.
So the very first step needs to be to screen the Olympiad for quality, and then we see what we've got leftover. Then your program's output could help fill in some FIDE numbers (etc) for these players.
|Aug-19-14|| ||chessgames.com: Before I drop the hammer on a number of recent Olympiad games, let me just give an example to assuage any fears.|
Mdluli Meluleki vs D Ntagasigumwami, 2014
Here we have a player rated 1817 from Burundi beating an unrated player from Swaziland. On the face of it, it doesn't meet our quality guidelines.
If you find that this game is actually a gem in this rough, quick: drop it in a game collection or kibitz about it. That will spare it. We'd never delete a game that has already sparked interest. (So that 3-move checkmate stays.)
|Aug-19-14|| ||zanzibar: <chessgames> you mean that you don't intend to keep all the Olympiad games?|
That's certainly wasn't the goal I was aiming for.
Yes, my program did most of the work, with me supervising of course.
I believe all the information needed to update CG is on the blog posts - and indeed could easily be done automatically if you don't want to double-check it.
* * * *
OK - to explain the data - you have the FIDE card matched to the CG player. The raw FIDE name (and other data), as found on the card, is what I list.
Unless there is a compelling reason <CG> should use the FIDE name, following the comma convention to form the presentation (or display) name. If there isn't a comma in the name, then the FIDE name = CG name.
|Aug-19-14|| ||HSOL: Surely an Olympiad game is more worthy of being still in the collection (due to the competition it was played in) than loads of simul games, blindfold games, prison games (Bloodgood or whatever his name is against nobodies (in the chess world)), 22xx players against computers, games from U8 and U10 championships, user games against 1500 level competition uploaded by the user.|
|Aug-19-14|| ||chessgames.com: <you mean that you don't intend to keep all the Olympiad games?
That's certainly wasn't the goal I was aiming for.> |
I hate to break it to you, but that's what we do every Olympiad, and for giant opens, and all sorts of tournaments. Your efforts have not triggered this policy. On the other hand, your data on your blog is great, and will be used.
<HSOL> Bloodgood was actually a strong expert, and some of his games have entertainment value, maybe even theoretical value for Grob enthusiasts. But have you seen that Meluleki vs D Ntagasigumwami game? It's childish. It's just not our goal to archive every game we can find.
Agreed, we already do have some very poor games in the database already, but our goal is to eliminate them. That's why the collection forms have a "poor quality" checkbox. We don't stick to this dogmatically; this week's Monday puzzle was between players under 1600, but those are the exceptions.
|Aug-19-14|| ||zanzibar: <chessgames> Let me guess, without looking at it, that Meluleki vs D Ntagasigumwami is the fool's mate game. |
It has to be... but deleting it wipes out one of the more commented upon games in tournament.
OK - I realize my role isn't to decide, but I'm of the opinion that <CG> should keep all of 'em (agreeing with <HSOL> here). I won't be miffed getting overruled on this - so long as I can get this post off.
Firstly, this is still a new concept to me. I never really thought about <CG> having a "quality"(?!) filter beyond what one would intuitively expect. I just assumed all big modern FIDE events were uploaded completely (surely disk space isn't an issue anymore - given the low cost of Tb drives these days).
E.g. all Fischer games, no matter how poor the opponent, etc. And there are some pretty piss-poor games in those old historical tournaments us biographers love to catalog. Which brings me to the question...
How does <CG> decide what games have merit then?
- Do you run a blunder check program?
- Take a vote using your star panel of judges? (Howard Stern, Paula Adbul, Simon, etc.?)
- Phone-in viewer vote?
- Do it statistically according to <CG> user views/downloads?
- Or more prosaically, e.g. number of moves or average player rating?
Inquiring minds want to know!
Seriously though, you already automatically uploaded all the Olympiad games, so it seems more work than what it's worth to do the weeding.
If it were up to me I'd keep all the games, even the stub games, in order to get the complete crosstable. I like the Olympiads window into the world too - and who knows, maybe some hotshot young player falls in love with <CG> after finding his local heroes games here, and goes on to become WC.
Plus I bet some of the low quality games have some nice tactical puzzles in them (sometimes far beyond the ratings of the players at the board, if <CT> can be an indication).
|Aug-19-14|| ||rookhouse: <chessgames> Sorry, but I have to agree with <zanzibar> and <HSOL> to a certain degree.|
Olympiad games seem much more valuable to the historic content of the site, as opposed to a lot of the published games on here that are a complete joke.
I am still trying to figure out for the life of me why a 1400 player (and well known con artist) like Ed Trice has his own player page and 5 presumably fictitious games published on here?
If you are indeed trying to eliminate past errors in judgement, it seems as if that would be an excellent place to start.
I also agree with <zanzibar>'s reasonable request to have some type of published criteria on what is acceptable for submission. I know this could be more difficult than it sounds, but what is the alternative?
|Aug-19-14|| ||crawfb5: The general guidelines for game submission are here:|
PGN Upload Utility
|Aug-19-14|| ||rookhouse: Thanks <crawfb5>, but the current guidelines are not specific enough for the purposes of the Olympiad debate. Plus they were obviously not adhered to in regard to my previous example of Trice (and I suspect many, many others).|
|Aug-19-14|| ||WCC Editing Project: |
I'm sorry to have been so strident. Clearly I didn't understand you properly, which is my fault, not yours.
I hope you can forgive my rudeness.
I'm thrilled that you join us with such great resources and the willingness to create new historical tournament pages.
With regard to the fine work by <suenteus po>, which was usually done in concert with <Phony Benoni> and <Benzol>, sometimes he's gone for some time but he usually comes back.
You are a cg.com editor right?
With regard to improving an existing tournament, after letting folks in the <Bistro> know what you plan to do, if you don't get a timely response you can just go ahead and improve the event. I have gone through a dozen or more of <suenteus po's> USSR Championships collections just to drop a source for his information- Cafferty and Taimonov's book on the championships. If you are willing to do more work on these collections and other old collections in terms of adding dates and sources and so forth, that would be most excellent.
<Cg> I agree with colleagues that all of the Olympiad games be added, because it is a major historical event. As such, completeness for the historical record should trump the necessity to add in some sub-par games.
|Aug-19-14|| ||zanzibar: For the record, I played over
Mdluli Meluleki vs D Ntagasigumwami, 2014
and yes, there is little reason to keep this game, other than for completeness.
Well, there is this:
After 32.Nf1?! Black has a M-6:
click for larger view
Two moves do it, each of which allow a maneuver against what critical square?
And how does the mate work if White doesn't open this flight square?
|Aug-19-14|| ||chessgames.com: We appreciate the notion of keeping the games for the sake of historical completeness, and arguably even the lowest tier of Olympiad games meet that qualification. After all, what if you were from Swaziland, wouldn't you be hurt to see that Chessgames deemed your nation's team not worthy of inclusion? So we're not champing at the bit to delete games, but be aware that it happens, and I believe for good reason.|
I don't have any opinion on the authenticity of Trice's games, although as quality goes, some of them are undoubtedly vanity games. That doesn't particularly bother us; there are only about two hundred vanity games in the database total. The number of poor quality games is probably in the thousands if not 10K. In any case, feel free to submit a correction slip if you think they really don't belong.
|Aug-20-14|| ||Shams: I agree, the Trice games should go. He already took up enough oxygen around here for one lifetime.|
|Aug-20-14|| ||rookhouse: <WCC> Thank you for the kind words, but I do fully understand and applaud your efforts for historical validity. When I first got into chess history many years ago I was reckless and naive to the necessity of proper sourcing. I had the benefit of receiving some "tough love" from a couple of excellent historians in Neil Brennan and Taylor Kingston. I additionally have had the honor of receiving guidance and exchanging ideas with the great John Hilbert.|
I have tons of historical data "gathering dust" on my both my hard drive and on my bookshelves due to life changes (kids, job, etc.) and do not have nearly the free time I used to.
That being said, I would be happy to help out on some of this stuff as my time permits and will let you know of any tournaments and/or matches to which I can contribute.
<CG> I hope the users on here appreciate the fact that you are as flexible and understanding of a site owner as you seem to be. Plus the fact that you have such a high tolerance for taking crap from many of us (especially me). I think your change of heart on the Olympiad debate was definitely a wise one.
<Shams: I agree, the Trice games should go. He already took up enough oxygen around here for one lifetime.>
That my friend, is about as perfect of a statement that I have ever read. Well said!!
|Aug-20-14|| ||zanzibar: Great news at least somewhere in the world...
Wordpress has restored the old editor!
It's known as "Classic Mode", which reminds me of "Classic Coke" for some reason.
(Of course it's not 100% restored - the old easy way of adding a photograph/image is still gone in lieu of <Add Media>.)
|Aug-20-14|| ||Karpova: I wrote a new intro for Ostend (Championship) (1907).|
But I don't want that anything gets lost, so here are the parts I didn't keep:
Tournament Book written by Dr. Tarrasch.
1907 fanden in Ostende zur selben Zeit zwei große Schachereignisse statt: das von Dr. Tarrasch auf den Namen "Champion-Turnier" getaufte Großmeisterturnier sowie das III. internationale Schachturnier.
Am Champion-Turnier beteiligten sich sechs Großmeister, die einander viermal begegneten und die folgendermaßen abschnitten:
[At Ostend in 1907, there were two big chess events at the same time: the 3rd International Tournament, and the Grandmaster Tournament in which Dr. Tarrasch won the title "Tournament Champion". Six grandmasters participated in the Champion Tournament, each meeting the other four times.]
Other Great Tournaments / Chess Events in 1907:
- Karlsbad (1907)
- Lasker - Marshall World Championship Match (1907)
|Aug-20-14|| ||zanzibar: I did a little more on name matching between <CG> names and <FIDE> names for <Tromso Open olm (2014)>.|
It's really not so important, more of a refinement after the data is put into order - if you want the <CG> / <FIDE> as correct as possible. (And in review potential <CG> mistakes (or <FIDE> mistakes) can be uncovered.)
You can find the long-winded version of the long-winded comparison here:
There are some discrepancies which deserve a followup... as always.
But not these duplicates: <
119013 Herman Hou-Meng Ho | Herman Hou-Meng Ho 16300190
150759 Herman Ho Hou-Meng | Herman Hou-Meng Ho 16300190
151081 Salim Al Amri | Salim Mohammed Salim Al Amri 22000089
150991 Al Amri Salim | Salim Mohammed Salim Al Amri 22000089 >
Which it seems <CG> has already merged together. Yeah!
There are also some trivia matches that the program doesn't filter out just yet (though it does a reasonable job of filtering a working set of 859 name-pairs to just 94 needing examination).
< Earlier Kibitzing · PAGE 293 OF 293 ·
A free online guide presented by Chessgames.com