<1. Errors processed first-in-first-out, no exceptions.> OK, this was an overstatement--what I really meant to say is that it would follow a first-in first-out format as a rule rather than the random method we use now.

<It's very frustating to keep submitting corrections when nothing ever seems to change.> We sympathize with you, but believe me: even if the corrections haven't been implemented, they are truly stored faithfully. We have a list of correction slips that would stretch to the moon and back, but we don't lose a single one of them--until of course we actually make the correction. So even if you don't see the fruits of your efforts immediately they are not being lost.

Here's a tip: when it comes to sort of obvious corrections, like say a "Susana Ferreira" and a "Su Ferreira" who are obviously the same, it's not that critical that you tell us. We hire people to look for things like that all the time, and we correct a scores of things like that daily. I see people make submissions like that, and I know their hearts are in the right place, but it's really duplicated efforts. We would have come across that correction anyhow. In fact, often when we process the correction slips we find a large number of them about player records that don't even exist anymore--admins have already merged them where they need to go.

Thanks DCP23

DCP23: <MacNilsus: Thanks. And do they have the games here? Couldn't find them...> <siamesedream: <MacNilsus> You can find women games only on official site:>

<Actually, CG.com has the women's games, they are all here but the link is top-secret.>

Here it is:
17th European Team Championship (Women) (2009)

for example:

1. e4 e5 2. Nf3 Nc6 3. Bb5 a6 4. Ba4 Nf6 5. O-O Be7 6. Re1 b5 7. Bb3 O-O 8.a4 scores 37.9% for White but if want to see the % score since say 2000 I need to do a search for the games and compute it

<exhibition games, odds games, etc. are excluded from this statistic.>

<chessgames.com>, what does "etc." mean? are blitz games or rapid games (when marked as such) excluded from the stats, or does "etc." refer to some less intuitive category of games?

The way the software actually works is that it looks for keywords which included "BLIND" and "ANALYSIS" and "EXHIBITION" and "ODDS" and so forth. If it sees one of those keywords, it doesn't count.

to be entirely sure, if a game is marked "BLITZ" or "RAPID", it will <not> be counted in the stats?

the reason i ask, is if the above is true, then based on tests i did with a number of known rapid and blitz events, including internet games with non-classical time controls, the <main cause of error> in the player stats is a substantial number of blitz and rapid games being marked as "CLASSICAL" which i assume is the default category when new games are imported.

the missing classification of games is one of the major beafs i have with chessbase' databases (which i still use a lot), both their mega database and the online one, and i think it would be a huge "selling point" for chessgames.com if one could wash the db to create a mostly accurate classification of games of major players. a great help in this process would be to be able to either have "game type"/classification as a search parameter - or alternatively, simply to include a "game type" column in the game listings. of course, the latter would expose all the wrongly classified games in the db - but i would assume that the desire to correct these errors weighs more heavily than the downside of exposing the current inaccuracies.

(another, related issue with the game listings, and something i've never really understood, is why games can't be listed sorted by year+tournament title. is this a limitation of the db backend or something else? if "on-the-fly" sorting is too costly, one could do batch-oritented db-reorderings now and then, and if the order is based on some current index or such, i would be surprised if the db engine wouldn't allow you to have several indexes to speed up search. technical issues aside, as a user i find it quite inconvenient that games from the same event aren't listed together for a given player - but i assume this issue must have been raised on several previous occasions...)

returning to the game classifications: anchored to tournament tags, i could supply a long list of games that should be reclassified to either blitz or rapid, and it would be much more efficient for all of us if that could be done as a db update for games from those tournaments instead of going through the web interface for each single game in question. the validation of the correction would then be to check if the <events> in question were rapid/blitz or classical - instead of fiddling around with individual games. i could even supply "documentation" for most of the events in question.

I wanted to let you know that the pairings for the first round of the FIDE World Cup are up: http://www.fide.com/component/conte...

If you search for only Platz (search "Platz") it will take a wild guess at which one you want, and give you Dr. Joseph Platz. If he was the Platz you were looking for you'd say "gee, that was easy!" but if not, then it can be frustrating. However, suppose if you try

search "H Platz"

then you get "Hans Platz". However that's all the "EZ Search" behavior--if you are searching for a player somewhat obscure like this, you should really use the Advanced search instead.

Go down a few inches on the homepage for the Advanced Search. Now if you enter "Platz" as the "First player" and you'll get a list of all the Platzes in our database.

When it comes to non-standard, but very long time controls, we're not exactly sure where we stand. If both players invest several hours into a game we are inclined to call it "Classical", even if it doesn't fit the format that FIDE happens to endorse of late.

< Will Chessgames cover the games live?? >

The answer is, "Yes--if at all possible." We don't have a reporter in Moscow, but unless the entire internet is blacked out (which is unlikely) then yes, we'll have live coverage.

I presume cg.com will be getting the moves from elsewhere than the http://russiachess.org site then?

While internet isn't a Single Point of Failure architecture, chess organizers seem to have ignored <scaling> and <high availability> issues. The reliability of these chess relays and broadcasts, has been worse, than what people experienced 20+ years ago... with a 1200 baud modem.

Very good point <zarg> and I very much agree. That’s one of the reasons I started my TFD-Proxy project. Replacing the client part of the broadcasting software created a more reliable overall solution. Yes, there isn’t much that we can do if the official server is completely silent but that’s not so often the case. It turned out that a lot of grey areas can be handled by a carefully designed proxy software.

However, what I would like see the most, is a relay/replicating/mirroring standard document. DGT may provide their own broadcast SW, but they should not have anything to do with the server relay architecture.

<chessgames.com: What the official sites do as a matter of policy is completely up to them.>

I thought the main issue to address would really be non-technical, namely how to handle the organizer sponsor issue at the relay site.

After the broadcast of the Kasparov-Karpov match broke down all over the internet, <frogbert> and me had a short technical discussion over at his player page, about the "brilliant" idea of providing live video feeds from a <single> relay/broadcast server.. search for "zarg P2P".

From the outside, what looks to be needed is a technical work group to come up with a public document specifying a relay/replicating/miroring architecture, host-to-host among the organizers and the chess sites servers. It's IMO the responsibility of the technical people at ICC, playchess, chessgames, FICS etc to make this happen, it's not DGT who should put down requirements or provide lock-in. The organizers might lack the technical resources.

This sound as rather bad design. Relay-to-servers should have higher priority than broadcast-to-users.

The sponsors deserve a robust and reliable infrastructure. I read most of your discussion with Mr. Runde about P2P infrastructure for chess broadcasts. You could even approach military calibre communication using techniques like you describe. But for now, just having a single backup server restricted to sites like Chessgames, Monroi, etc. would be a fine step forward.