I think <CG> could open this up to known contributors to promote and facilitate such valuable contributions.

You know <Tab>, <SCID> was just trying to be helpful.

I'd urge a measure of caution - <SCID> is like Trump, if you attack him you can expect a counter-attack with 3 x force.

Anyways, <SCID> decided to give it another go, and looked at the Swiss xtab sorted by Rating (something SCID does know about).

Seems there's a readily apparent cut-off, with a clear preponderance of games for players with a rating > 2400.

It would be interesting if someone were to inquire of <Avdiu Nysret> the exact situation, and criterion used. (Maybe it's just a question of number of DGT boards available?)

* * * * *

BTW- I agree with this:

< I was hoping that they might eventually present all the games of the 23 qualifiers (of which some has only 8 not 11 games), and that CG would take them in later.>

There is always the possibility that some of the missing games were actually forfeits. Or perhaps they are missing because the opponent was low-rated?

(Which would typically show up as missing games from the early rounds in a Swiss).

Years ago I was doing some fiddling around to see how binary-object uploads work in HTML and I made a hidden feature intended to let people upload a graphic to a game collection. The Game Collection: "Aron Nimzowitsch: A Reappraisal" by Keene used this feature as a proof-of-concept.

Even though binary uploads exists all over the net it takes a little savvy to do it right. People will upload 1 terabyte disk images to see if it crashes the server, if you let them.

Yes, you will find exception to that rule both in new games and older ones, but especially in the new games. Here's why:

When there is an event that includes both strong and weak players alike, we try to get all the scores possible. Then, when the event is over, we may run a program which is designed to prune out the "bad games."

Why, you may wonder, don't we simply prune out the bad games as they are being input into the system, rather than collect them only to later give them the axe?

There are actually several good reasons. First is the unlikely but interesting possibility that somebody who would get the axe with our normal criteria actually wins the event, or at least seriously outperforms.

Then there is the issue that in the early stage of the tournament, if you don't have all the games the leaderboard will betray that fact. Later it's likely that no 1500 players are on the leaderboard so it won't matter, but at first it is obvious when you're pruning.

Moreover, there are cases where the first round shows a name with no rating indicated and later we see that they are rated 1380 (or 2380). If we use the prune algorithm at the end of the event, we can do so using all of the information available, whereas if we do it round by round, the first rounds may be operating ignorant of facts later to be revealed.

Finally there is simply the fact that if you or a friend play in such an event, we're happy to let you bask in your victory on our pages. The trick is to do something with the game: kibitz to it or put it in a game collection. Once you do that it will be immune from deletion regardless of how low the ratings are.

FISHY GAME 1818445

T Magnusson(1411) vs Olafur Orn Olafsson(1149) (A46, 37 moves, 1-0) 2016 [Reykjavik Open]

(a) neither player is rated>2100
(b) total number of games played by T Magnusson is 35 total number of games played by Olafur Orn Olafsson is 14 (c) no kibitzing on this game
(d) no kibitzing on T Magnusson
(e) no kibitzing on Olafur Orn Olafsson
(f) Nobody has this game in a collection.
(g) tournament is an 'Open'
(h) neither player has a biography

<The Game Collection: "Aron Nimzowitsch: A Reappraisal" by Keene used this feature as a proof-of-concept.

Even though binary uploads exists all over the net it takes a little savvy to do it right. People will upload 1 terabyte disk images to see if it crashes the server, if you let them.>

Well, the Nimzowitsch page looks quite good.

The upload feature's greatest challenge isn't preventing obvious hack attacks, at least I wouldn't think it would be.

Certainly a limit on the number of uploads/day and the max size would have to be built in. I would also assume that only "editors + vetted" contributors would be granted permission. I assume the only worry would be if some hacked logged into one of these accounts. This must not happen too often, else you'd already have problems.

I think the great worry would be innocent uploads of copyrighted material. Requiring a source tag would help there. Not a great concern, but probably the biggest headache.

Of course, besides DOS-attacks, one must also worry about inappropriate images being uploaded. A good audit trail, plus allowing other editors to back out a change might be useful there.

Adding an automatic notice of an upload in the Bistro might be helpful.

<
I want to return to this idea sometime in 2016 but next on the agenda is the aforementioned [Source "_____"] tags. >

Yes, this should already be implemented. Really, I think we can live with almost any format a biographer wants to use - so long as it isn't stripped out it would be useful.

I would have been more inclined to use a pre-stage system, allowing immediate publication of live games, and early adoption of tournaments - with the caveat that all games/results/moves are provisional.

Then, after proper vetting, and maybe a wait for tournaments to stabilize their results/games, a promotion into the main db.

That would involve merging comment streams, I suppose. Some complexity seems unavoidable, either way.

Perhaps we can cajole <CG> into accepting this modest programming challenge.

E.g.

http://chess-results.com/tnr206558....

could be used to produce a valid leader-board for the tournament.

It actually could be used to produce a complete and valid stub version of the tournament, which could then be overlaid with whatever games are deemed worthy of inclusion to <CG>.

It's a little embarrassing to confess this after all these years, but better late than never. I have never been able to figure out how to navigate their site.

Take your 17th European Individual Chess Championship as typical example. Somehow you arrived at this wonderful page:

http://chess-results.com/tnr206558....

Now suppose you didn't spoon-feed me that link. I would then have go to the homepage http://chess-results.com/ and see no such link. They have a "Tournament search" link that takes me to a form, but I must not be using it right. They have a list of tournaments but for the most part they seem to be extremely minor ones.

My main complaint about them is that their setup is not really good at showing shared results, so, for example, they will give a "first" and "second" place even in the (admittedly rare) case when the actual tournament doesn't set tiebreak rules and recognizes shared first place. But other than that, they are very good and reliable.

Official sites will often link to them from their own "pairings and results" section; if they don't, you can check where the tournament is held, and then narrow the search by clicking the relevant nationality code in the "Federation selection" on their main page.

But, truth in advertising and credit where credit is due...

I really was following <Tab>'s links for that one tournament.

I have to admit that I too find the site a little hard to navigate. But I haven't much practice with it too either, as I'm generally not in such a rush that I can't wait for TWIC (who really provides much needed clean-up of the PGN in many cases).

Let's do <AM>'s first:

https://www.google.com/search?q=17t...

The first link gives a list of participants:

http://www.chess-results.com/tnr206...

There's a button on the page -

<"Show tournament details">, which gives:

http://www.chess-results.com/tnr206...

Just one more key step:

Find the <LISTS> list, and select

<Final Ranking crosstable after 11 Rounds,>

http://www.chess-results.com/tnr206...

First, let's assume you know the location, and that the tournament is very recent.

Start at the homepage:

http://www.chess-results.com/Defaul...

Then select <Tournament Search> to get here:

http://www.chess-results.com/Turnie...

(or just start here with a bookmark).

We use the location <"Gjakova">, which is very unique (unlike, say, "London") to do a search. Since it's so unique we only get the one tournament after we hit "Search": ...

I can't show this step, as the url didn't update during the search.

The page doesn't display how many hits it found, or provide good feedback (again, imo). However, if you scroll down the results of the search should be there - unfortunately, it's off my screen if I don't scroll down - so it's easy to think the search failed (again, due to poor feedback).

Anyways, the tournament <17th European Individual Chess Championship> shows up, with this link:

http://www.chess-results.com/tnr206...

We at a familiar place from the <AM> search, a list of the participants.

Again, the Swiss xtab is a <LISTS> link:

<Final Ranking crosstable after 11 Rounds>

http://www.chess-results.com/tnr206...

I suppose one would have better confidence knowing, a priori, that the tournament had 11 rounds.

If I designed the web site, I would offsite the round number and tournament format a little better.

One could start this style search using the tournament name. Here, using "European" could be used. Again, do the search and scroll down. The link we're interested in is 17 entries down.

It looks like if you scroll horizonally, Rd, n, and db key entries are available. There doesn't seem to be any indication of average rating, or tournament category - which might be helpful when filtering.

1) First wave - the scores for all games played with DGT boards, available immediately.

2) Second wave - the scores from games recorded on paper, available only after transcription.

I just prefer working from inside the site (I like the challenge, so sue me...) ;p

<Tab: <<Official sites will often link to them> Rather, organizers are <using> Chess-Results (directly, by feeding data to it) as their results page.>>

Yes, and more importantly, as their pairings tool as well- that's why it's so reliable (at least <up to> the last round) - the organizers absolutely need it to be correct to get their next round pairings right. :)

So also note that - for the same reason, the last round may be the most neglected of all. The organizers no longer need it, so it's just a question of how much they care if the final standings are correct online or not.

Another detail: suppose the organizers link to their tournament, but with the version of the page in their own language. There's a language selector at the top row of the page, but if you click ENG on that, it will send you to the English version... of the main page, so don't. :s

Yet the solution is very easy; the page url always contains the language string.

Example: (http://chess-results.com/tnr224593....)

.../tnr224593.aspx?lan=20

Now all you need to know is that English is lan=1, just fix the url manually, and you're right where you want to be.

<zanz: <We at a familiar place from the <AM> search, a list of the participants.>>

Yes, that's the tournament's "main page" according to Chess-Results organization. Everything else is just one click away. :)

<I suppose one would have better confidence knowing, a priori, that the tournament had 11 rounds. >

You do! In ongoing tournaments, you will see that the last round results/pairings format is almost always given as, for example, "Rd.2/9".

There is one other more serious difficulty, btw - the Russian organizers often enter their details (tournament name, player names) in cyrillic script, so a Latin alphabet search won't find it. :\

And even if you do find the tournament, all the player names are in Russian... In these cases, I take the page url, and run it through Google Translate.

Notable Chess Games

At first glance it isn't much different from the "most kibitzed games" list that you find in our statistics page, but this is a bit different. It's the "notable game" algorithm, slightly tweaked, applied not just against the entire database but decade by decade.

So for example, think of some
Notable Chess Games of the 1910s then click and see if it has the one you had in mind. The gold coin game? Lasker-Thomas? Sure enough, they're all there.

I decided it just didn't make sense to break up the 19th century into decades because there was so little recorded chess in the earlier half. And so we have simply Notable Chess Games of the 19th Century.

Greco, Philidor, Ruy Lopez, etc. are found in Notable Chess Games prior to 1800.

Of great interest of course is the most recent list, Notable Chess Games of the 2010s.

We're using special "secret sauce" triggers now to try to detect which are the most notable games. One new trigger is when a game gets linked to from another game, like if somebody says "This reminds me of ____". When you start to see the same game in posts like that, it's bound to be a very well known game.

Others secret triggers we don't want to disclose in fear that people might try to game the system, but they are pretty well thought out. The proof is in the results.