|
< Earlier Kibitzing · PAGE 62 OF 65 ·
Later Kibitzing> |
Feb-08-25
 | | Stonehenge: Bacrot vs Lautier, 2003 (kibitz #2) |
|
Feb-08-25
| | Stonehenge: <Duplicate FIDE IDs> here Player Directory (Beta) needs to be cleared out. |
|
| Feb-08-25 | | stone free or die: Like the sortable column upgrade. |
|
Feb-08-25
 | | jnpope: <Stonehenge: But there are no aliases visible to us poor editors. I want to be able to zap those aliases.> I've also come across this "problem".
I think it's a (flawed) best match algorithm of Daniel's and there really is no alias to zap. My solution to defeating the algorithm would be to create a PID for <D Antic> and move the games under that PID. Then all future games for "D. Antic" should go under that PID as that is now the "best match" and they shouldn't end up in Dragan Antic. However, if the intention is that <D Antic> should be an alias for Dejan Antic, then merge the newly created player <D Antic> into that PID and that will become an alias for Dejan. |
|
Feb-10-25
 | | stevemcd87: <Stonehenge: <invisible aliases>> To identify a player, the gameloader checks for the following in order: CG's pid(e.g WhiteChessgamesPid, WhiteChessgamesid, ChessgamesWhiteID), FIDE number (e.g WhiteFideId), Long name (e.g Dragan Antic), short name (e.g D Antic), and lastly alias. Perhaps we can have the gameloader first check for aliases before short name so that we can assign the alias to whom ever is still active, rather than assigning the game to the first short name it matches. Let me know if you'd like to give that a shot. |
|
Feb-11-25
| | Stonehenge: Both Dejan and Dragan don't have D. Antic as an alias. Therefore, games uploaded as D. Antic should make a new player ID: D. Antic. Then it's upon the editor to take futher action if necessary. |
|
Feb-13-25
| | Stonehenge: From a correction slip:
Abraham Moses Sussman Photo from The Boston Herald, 6 February 1903 https://ibb.co/N2v7tzpV Maybe you can upload it? |
|
Feb-16-25
 | | MissScarlett: Abraham Moses Sussman |
|
| Feb-17-25 | | stone free or die: I'd like a page number too, but I guess you can't have everything. gg. |
|
Feb-18-25
| | stevemcd87: <Stonehenge: Both Dejan and Dragan don't have D. Antic as an alias. Therefore, games uploaded as D. Antic should make a new player ID: D. Antic.
Then it's upon the editor to take futher action if necessary.> They both have the short name <D Antic> so the script designates the game to the first player that matches the short name in the database. Are you suggesting to remove the logic to check for short names in the gameloader script? Leaving it up to the editor to merge <all> short named players to it's "correct" player. We may want to discuss that idea with the rest of the editors as it may make uploading games more time consuming then before. I was suggesting to check for aliases before short name so the editors have the choice to whom the short name should belong to. Let me know if I need to elaborate further, I know it may be confusing |
|
Feb-18-25
| | Stonehenge: <Let me know if I need to elaborate further> OK, I'll ask a stupid question. Do we need to check for short names? Say there are ten different John Smiths, Jack Smiths, James Smiths etc. in the database. All with the short name J Smith. A game gets uploaded with the name J Smith. So it goes to the first player with the short name J Smith. This is rather ridiculous. Why not make a new pid? Of course when there is already a long (sic) name J Smith in the db it will go there. That cannot be helped. Unless the date is way out of range for that player (maybe that's something to look into too). |
|
Feb-18-25
| | stevemcd87: <OK, I'll ask a stupid question. Do we need to check for short names?> I suppose with the more common short names, it can be a nuisance but may be helpful for non-common short names. If nobody objects to the removal of short name checks, I will remove it. I suppose we can always add it back if we'd like. |
|
| Feb-18-25 | | stone free or die: You know, there should be a way for a submitter to signal, in a clear cut manner, that a new player needs to be generated. I would encourage biographers to always use the <ChessgamesWhiteID> (or BlackID) with the <CG> player id's for uniquely identifying the player associated with the game. Perhaps a negative ID could be used to signal that this is a new player that needs special handling? (This might need a little more thought on the details, but I think the general outline is valid.) |
|
Feb-18-25
| | stevemcd87: Perhaps a <find & replace> feature in the editor page to be able to update their short name to their long name may be helpful as well. <Unless the date is way out of range for that player (maybe that's something to look into too).> There is code written for this as well though it seems it doesn't work all the time. |
|
Feb-18-25
| | Stonehenge: <There is code written for this as well though it seems it doesn't work all the time.> Could it be that it only works when DOB and DOD are known? Say there's one John Smith in the db (born 1900-died 1990), then a new game played by another John Smith, before 1900 or after 1990, will create a new pid. |
|
Mar-03-25
| | Stonehenge: stevemcd87 chessforum (kibitz #1531) Doesn't work anymore after FIDE has changed the url lookup of players. |
|
Mar-03-25
| | stevemcd87: <Stone>
We've removed the short name logic from the gameloader.
<Could it be that it only works when DOB and DOD are known?> Yes. Are there players where that logic doesn't work? <Doesn't work anymore after FIDE has changed the url lookup of players.> Working on it now. |
|
Mar-13-25
| | Stonehenge: <Are there players where that logic doesn't work?> See V Petrovic vs Z Majeric, 1982. Vojislav Petrovic became a new pid but it could only have been Vojislav Petrovic since Vojislav Petrovic died in 1944. |
|
Mar-26-25
| | MissScarlett: The <[White/BlackTeam]> PGN tags are being relegated during upload to the bottom of the PGNs: https://www.chessgames.com/nodejs/g... Can we leave them as submitted? |
|
Apr-03-25
 | | Tabanus: <Steve> the Niemeyer Juniors (1966) should be <1967>, just in case it will not fix itself. The 13 games CG had from before were all dated 1966 instead of 1967. |
|
Apr-05-25
| | Tabanus: Niemeyer Juniors (1967), fixed. |
|
Aug-16-25
| | Stonehenge: Player Directory (Beta), Duplicate FIDE IDs needs to be cleared out again. Folke Ekstrom (kibitz #29) this is a known bug, right? |
|
| Aug-16-25 | | stone free or die: One word of caution - though rare(?), duplicate <FIDE ID>'s can be legitimate, as FIDE recycles player id's upon their death. E.g. I know for a fact that Bent Larsen's FIDE id was recycled not long after his death. |
|
Aug-17-25
| | stevemcd87: I don't understand the bug. I see the win against Euwe ( F Ekstrom vs Euwe, 1946 ) listed on Folke Ekstrom 's page. |
|
| Aug-17-25 | | stone free or die: I was just responding to the mention of duplicate FIDE id's. But looking into this a little, I'm also a bit confused. Was the FIDE id reference meant to be <CG> id? Also the "but doesn't appear in the list of Ekstrom games" statement can refer to the list of games on the player's page, but also via the tournament page. Where was it missing? |
|
 |
|
|
< Earlier Kibitzing · PAGE 62 OF 65 ·
Later Kibitzing> |
Kibitzing
