Normal Topic ChessPub Live Event Coverage (Read 95 times)
Nernstian59
Senior Member
****
Offline


I Love ChessPublishing!

Posts: 297
Joined: 12/15/21
Re: ChessPub Live Event Coverage
Reply #1 - yesterday at 21:43:01
Post Tools
Tony - This coverage is an interesting new feature which I don't recall seeing in any descriptions of the site's new capabilities. At first blush, the content presented seems impressive, but then the realization sets in that the AI is just automating the process we users go through when we search for a position in the accumulated Chess Publishing updates. 

A suggestion: Include the opening or Informant code in the titles of these items in the live event coverage. There are 16 posts for Round 4 of the 2026 British Championship, and this information would help those of us who are primarily interested in seeing new developments in "our" openings to zero in on the pertinent sections. For example in this Round 4 coverage, there's a section titled "Pert's early king move hands Sooraj the advantage", which doesn't indicate that a Caro-Kann is being discussed. In fact, the ensuing text doesn't reveal the opening either. The diagram is somewhat helpful, but given that the game proceeded (after (1.e4 c6) with the less common 2.d3 d5 3.Nd2 e5, it wasn't immediately obvious to this C-K novice that the diagram position arose from a Caro-Kann.

A bit more about the section on the Sooraj-Pert game: the opening phrase "Richard G Pert met Sooraj M R with the rare 7...Kf8" sounds a bit awkward, at least to me. This phrase is used with some frequency in the live event coverage, but isn't as natural or informative as "Pert played the rare 7...Kf8 against Soorai in a Caro-Kann that opened with 2.d3 d5 3.Nd2 e5".

Much worse for this section is the fact that Pert didn't even play 7...Kf8. According to the game score on lichess, which is given via a link after the text for Sooraj-Pert, Black actually played the normal 7...0-0. The text in this section goes on to say "Sooraj replied 8.Nxd4", which is also incorrect according to the game score. Sooraj's actual 8th move was 8.exd5. He played Nxd4 on his 9th move. It appears that 7...Kf8 is just an aside or a suggestion made by the engine or whatever is providing the automated annotation for the game. Is there a way to check the AI's output for such obvious errors before it's posted?
  
Back to top
 
IP Logged
 
GMTonyKosten
YaBB Administrator
*****
Offline


Mr Dynamic?

Posts: 3200
Location: Clermont-Ferrand
Joined: 12/19/02
Gender: Male
ChessPub Live Event Coverage
08/03/26 at 08:59:19
Post Tools
I've just noticed the Live Event Coverage page on ChessPublishing, at https://www.chesspublishing.com/events which is generated by AI, apparently. It also appears on a Twitter feed. Personally, I find a lot of the tactical moves suggested could do with some explanation and a bit of analysis.  Undecided
  
Back to top
IP Logged
 
Bookmarks: del.icio.us Digg Facebook Google Google+ Linked in reddit StumbleUpon Twitter Yahoo