< Earlier Kibitzing · PAGE 5 OF 11 ·
|Jul-19-11|| ||crawfb5: Thanks, but I mentioned game collections mostly because of the earlier discussions of event and site normalization and date sorting. For example, if one wished to begin with a normalization of an important historical tournament like Game Collection: Hastings 1895, then <Benzol> already has the games selected and organized by round, with the full date indicated. Adding the full date as part of normalization would have the additional advantage of slowly improving the sort.|
If CG does start a normalization project at some point, then might as well begin with the low-hanging fruit where someone has already done some of the spadework.
|Jul-20-11|| ||bartonlaos: Madam Librarian:
Maybe this site already does this and I haven't noticed, but there are some tournaments that are organized in such a way as to split their references on Chessgames.com to multiple threads.
I would like to request that such finished tournaments be reunited in a different fashion, or else to have links provided to each part so that the games are kept as one unit. That way the user doesn't have to hunt through the database for the rest of the games in the tournament. Here's the example:
US Championship (Group A) (2011)
US Championship (Group B) (2011)
US Championship (Knock-Out) (2011)
|Jul-20-11|| ||CG Librarian: <crawfb5> OK, I understand. <chessgames.com> is of course in charge of the project (which is still in the planning stage), and I assume most of the tedious normalization will be automated, but I'm sure I will have a role in it when the time comes. Looking forward to it.|
|Jul-20-11|| ||CG Librarian: I have a few general questions:
1. What resources should I know about for research? I'm currently using FIDE's and ICCF's player lookups, Olimpbase, and sometimes just Google. I know there are other databases but they have errors just like we do, and I'd prefer something more reliable.
2. Are Mohd, Muhd, Md ever actual names or just abbreviations for Mohammed? Besides English and Spanish ones, are there other name abbreviations I should know about?
|Jul-20-11|| ||Phony Benoni: <CG Librarian> For earlier players, Jeremy Gaige's <Chess Personalia> is generally considered a standard and reliable work. Unfortunately, it's quite out of date (last revised in 1987), so it will be no help at all with anything in the last 25 years.|
Also, as far as I know it is only available in print. If it's not available to you, a number of us have copies (you've probably seen it mentioned in a few correction slips) and surely someone will be willing to help with individual questions.
|Jul-21-11|| ||Stonehenge: <Are Mohd, Muhd, Md ever actual names or just abbreviations for Mohammed?>|
Yes, they are abbreviations for Mohammed / Muhammad / Muhammed etc.
|Jul-21-11|| ||Benzol: Don't know if "The Oxford Companion To Chess" by Hooper & Whyld is available on-line but it's a great tome for anyone interested in Chess history.|
|Jul-24-11|| ||Viewer Deluxe: < <CG Librarian>… (there are sometimes two ways to write a move when a piece is pinned, and Chess Viewer Deluxe even displays them identically, though the PGN is actually different).>|
The reason CVD displays such moves identically is because the viewer implements strictly the PGN standard’s recommendations -- namely, the disambiguation and normalization rules. Whenever there’s a need to disambiguate, the move’s original file is included (if different from the destination) and then (only if necessary) the origin’s rank. There’re many more (probably little known) transformations done by CVD. For example, if “+” or “#” indicators are included in the PGN text after moves where they aren’t correct, CVD will remove them for clarity. The opposite is also handled correctly. New “+” and “#” indicators will be inserted in places where they’re missing in the original PGN text (but needed). That’s very natural and helpful and most of us will consider it obvious. I find it strange that many other chess viewers don’t bother to include such features.
|Jul-25-11|| ||chessgames.com: This isn't a high priority situation but some people like to solve mysteries like this one, so I'll share it with you. Take look at B Heberla vs P Plakalovic, 2003, but be warned that the moves become absurd on or around move #19.|
Here's what I figured out about this game: If you try to correct the game by playing 19.Qe3 instead of 19.Qd3 then the moves all seem to make sense all the way up to 24.Qxg6+ which presumably should be 24.hxg6. Moreover, it explains the mysterious sacrifice of the bishop on c4: White believed that prying the kingside open was worth the bishop.
The problem with this theory is that if you correct the game by making it read 19.Qe3 and 24.hxg6, then the result of 1/2-1/2 becomes a mystery. After 24.hxg6 Black is defenseless and might as well resign: e.g. 24...hxg6 25.Rh8+! mates.
Put all this together, and I have a hunch that this was a really nice chess game, except the score has become damaged. Maybe somebody brighter than myself can figure out what actually transpired based on the clues we have. Or perhaps the correct form appears in some database, book, or bulletin.
|Jul-25-11|| ||SwitchingQuylthulg: <chessgames.com> I'm willing to bet you're right and the game continued 19.Qe3 Qb6 20.b4 Nxb4 21.axb4 Qxb4 22.h4 b5 23.h5 bxc4 24.hxg6 1-0... the final result at least is confirmed here:|
<GKSz Polfa Grodzisk Mazowiecki (POL) v. SK Glasinac (BIH)
Acs, P. 2606 HUN v. SCG 2467 Jeremic, V. ˝-˝
Cyborowski, L. 2550 POL v. SCG 2434 Scekic, M. 1-0
Miton, K. 2546 POL v. SCG 2345 Benderac, S. 1-0
Grabarczyk, M. 2508 POL v. SCG 2191 Batinic, P. 0-1
Heberla, B. 2420 POL v. SCG 2168 Plakalovic, P. <<<1-0>>>
Adamski, J. 2335 POL v. BIH 0 Kosonic, S. ˝-˝>
|Jul-25-11|| ||crawfb5: <Switch> looks to be correct. This site also has the result as 1-0.|
Every database I checked had the same incorrect score. Checking the PGN at the tournament site <Switch> provided confirmed the error started there (19. Qd3, 24. Qxg6+, and 1/2-1/2).
|Aug-01-11|| ||chessgames.com: <SwitchingQuylthulg: <chessgames.com> I'm willing to bet you're right and the game continued 19.Qe3 Qb6 20.b4 Nxb4 21.axb4 Qxb4 22.h4 b5 23.h5 bxc4 24.hxg6 1-0... the final result at least is confirmed here: ...> Brilliant research, we now have great confidence in that score. B Heberla vs P Plakalovic, 2003 shows the corrected form now.|
|Aug-02-11|| ||twinlark: <CG Librarian>
<1. What resources should I know about for research? I'm currently using FIDE's and ICCF's player lookups, Olimpbase, and sometimes just Google. I know there are other databases but they have errors just like we do, and I'd prefer something more reliable.>
Wikipedia can be useful, especially if it's properly sourced as these sources can in turn be very useful and can also provide leads to other useful sources.
A World Championships Index was compiled Weeks and Marks at http://www.mark-weeks.com/chess/wcc... This is useful for all the championship events, Candidates and Interzonals, but not so good for the zonals if you ever need to drill down that far.
http://www.ugra-chess.com/ provides info about the 2010 Olympiads as it hasn't been updated to Olimpbase yet.
Information about zonals: http://www3.sympatico.ca/g.giffen/z...
http://chess-results.com/ is reliable and widely used by official tournaments to display results, but only covers the last decade or so. You can search by player, year, tournament or country.
TWIC is also useful and extensive but needs a fair bit of work to mine the information on an ad hoc basis.
http://al20102007.narod.ru/ (Rusbase 1920-1994) is excellent for archived information about events in the Soviet Republics, and you'll find stuff there you won 't find anywhere else.
Let me know what sort of other info you need as there are many other more specialised sites, like a site about notable Armenian players: http://chessarmenia.blogspot.com/
|Aug-02-11|| ||twinlark: <ceegee-el>
P.S. You mentioned you look at the FIDE player lookups. Hopefully that means you also use the link to all the FIDE registered tournaments played by that player under classical time controls since 2001.
These are not always complete as until 2006 or 2007 or thereabouts, only games against rated players were reported, whereas FIDE now provides links for tournaments played after 2006 to results tables that include all players, including those with zero ratings. Even these have to be treated with caution, as sometimes some rounds are omitted for no obvious reason. I reckon it's generally worth getting collaborative information if there's any ambiguity, say from chess results.com.
|Aug-03-11|| ||CG Librarian: <Phony Benoni> <Benzol> Thanks for the recommendations. Chess Personalia does keep popping up, so CG is going to invest in a copy.|
<Stonehenge> Thanks. I almost assumed that Wim was an abbreviation for William, so I know it's best to be careful. ;)
<twinlark> Wow, thank you, that list is very helpful.
About the FIDE site, do you mean the "View Games of a Player online" link? I haven't been in the habit of looking at it because I didn't think it was complete. But like you said, even if it's not, it still can be useful.
I have another research question. Where do you get full player birthdates? On FIDE all I find is the year.
|Aug-04-11|| ||twinlark: <CGL>
<About the FIDE site, do you mean the "View Games of a Player online" link?>
No, it has interesting information and some PGNs but it's not very useful. I mean the "Click to view All previous periods" link. It's the entry to a treasure trove of information.
The FIDE database isn't especially user friendly, but it has an immense amount of information and has improved quite materially over the last few years. It's the single most valuable source I use for constructing bios, in conjunction with other sources.
<I have another research question. Where do you get full player birthdates? On FIDE all I find is the year.>
That is actually one of the hardest bits of data to find...you have to prospect around for them and find them in odd places, like their websites, or in player bios on official tournament sites, or national federation websites (eg: http://armenianchessplayers.blogspo...), or just use good old fashioned elbow grease on google.Speaking of which this may help: http://www.google.com.au/search?q=c.... And then there's stuff like this that matches dates to players rather than vice versa: http://maskeret.com/cgi-bin/today.cgi Also this one has been useful on occasions: http://www.echesspedia.com/
It's the most frustrating part of writing bios apart from trying to make sense of team and league reports unless you're into detective work for this sort of information.
|Aug-04-11|| ||whiteshark: For some reason all Kosintseva vs Kosintseva games mashup twice
http://www.chessgames.com/perl/ches... Worth to adjust?|
|Aug-04-11|| ||Phony Benoni: I've got a little dilemma which may have bothered others as well.|
I'm planning to submit games from the 2nd National Chess Federation of the U.S.A. Congress, held at Bradley Beach in 1928. (I had to do something; there are no games available from the Western Championship/US Open for that year.)
Now, I think the PGN should start like this:
[Event: "2nd NCF Congress"]
[Site: "Bradley Beach, NJ"]
Now, this event is generally known (when known at all) as "Bradley Beach 1928". I didn't know it was the 2nd NCF Congress until recently, and I doubt that more tnan five or ten thousand of our members know that either.
However, all the search results screens display "Event" information rather than "Site". I'm afraid people will be looking for "Bradley Beach", and simply pass over "2nd NCF Congress".
Does this strike others as a problem? Or am I being qualmy about nothing?
|Aug-04-11|| ||SwitchingQuylthulg: <whiteshark> That's just a bug :)|
|Aug-04-11|| ||twinlark: <Phony Benoni>
Couldn't you describe the event as "2nd NCF Congress (Bradley Beach)" or something like that?
|Aug-04-11|| ||Phony Benoni: <tinwlark> That is an alternative, though the clumsiness and repetition bother me. It's the same conflict I often have at work: Do it "right" and depend on the programmers to make the display of information useful, or tweak the data to "fix" the display myself?|
Perhaps it's nothing to worry about.
|Aug-08-11|| ||chessgames.com: Somebody suggested writing a utility which examined women's events and then set the gender flag to 'woman' in the appropriate player pages.|
We really liked this idea, so we made a utility that allows admins to do this on demand. It's called "womanizer". (Just a bad pun, we didn't mean anything by that.)
So we started off running it on a few women's olympiads. We found many mistakes where obvious women were labelled as men; the biographer just must have forgotten to hit the "F" flag. Great!
We also ran into odd things like this: Sochi 22nd Women's R (1966) -- clearly games like Polugaevsky vs Matulovic, 1966 should not label anybody as a female. Yet women are in the roster. Not sure what that is, but with that "R" at the end it's clear something there is mislabeled. Anyhow, it's finding oddities, which is good.
So in short, it will help straighten out some database issues and also uncover bad data that can be found easily. For now however it's still experimental; we don't want to let it run unsupervised otherwise we'll wake up to find that Milan Matulovic turned into a woman.
|Aug-09-11|| ||chessgames.com: <Phony Benoni: <tinwlark> That is an alternative, though the clumsiness and repetition bother me.> Me too, although in casual discussion we describe tournaments as "Hasting 1905" and "Zurich 1953" all the time. Nevertheless, I'd still prefer to see the site in the Site-tag and not repeated in the event tag for good measure.|
<It's the same conflict I often have at work: Do it "right" and depend on the programmers to make the display of information useful, or tweak the data to "fix" the display myself?> I say do it right and "hope that the programmers" address the problem.
For starters it's not the end of the world if somebody looking for Bradley Beach doesn't find it immediately because they don't recognize "2nd NCF Congress". At least we weren't wrong, we just weren't helpful. Perhaps also one day we might be able to squeeze the Site-tag into the game listings, so it looks something like <2nd NCF Congress, Bradley Beach> which would be just peachy. Or perhaps we'll solve the problem some other way.
One thing's for sure, the best solution is not stuffing data into fields that weren't intended for it. Then I fear we'll one day be looking at something like <2nd NCF Congress (Bradley Beach), Bradley Beach> which should make anybody cringe.
|Aug-10-11|| ||crawfb5: There are scattered games in the database with something like <New Orleans-Wch> in the event (and sometimes site) field. It's a heterogeneous group, from various events and various years, from some entry glitch in the distant past.|
I think I can solve a few from 1933. In May a qualifying tournament was held in New York to pick two players to join Marshall and Kashdan on the US team for the Folkstone Olympiad.
The crosstable from the June 1933 <Chess Review> looks something like this:
Denker vs Fine, 1933 is certainly from that event. I submitted a correction slip on that one because there was a minor problem with the game score.
Levenstein vs Fine, 1933 was also from that event.
Chess Review printed three other games (Dake-Willman, Willman-Hassialis, Beckhardt-Simonson) that we do not have in the database. I'd wager any of the 1933 "New Orleans" games between any of the players in the crosstable above are from this qualifying tournament.
|Aug-10-11|| ||Phony Benoni: I suspect that the <New Orleans-Wch> games took place in various tournaments and matches in New York. For instance, there are several from Duras in 1913 where the opponents and results match his in the Rice Chess Club tournament.|
It would take significant checking to verify this, but that's a place to start.
< Earlier Kibitzing · PAGE 5 OF 11 ·
Bobby Fischer Tribute Shirt