ARCHIVED POSTS
< Earlier Kibitzing · PAGE 729 OF 1118 ·
Later Kibitzing> |
| Jul-31-14 | | crawfb5: <zanzibar: I now realize what I really want the most - the ability to edit a forum post.> As you may have discovered, a straight copy and paste usually kills any hyperlinks. I usually open another tab in my browser, do a copy and paste, and then I have the original post still open to click on any hyperlinks to get the URLs. I post the new version and look it over. When I'm happy, I then delete the old one. It's not an ideal solution, but it's an imperfect universe. |
|
| Jul-31-14 | | zanzibar: <crawfb5> Yes, I've used that technique (and made the same lament - ha!). There's another technique I use - which involves going back in the history to find the edit page. Then I have the raw input - which I re-edit (and can even do a preview). Once happy with that I cut-and-paste, return to the forum (else it complains about duplicate submission) and post a new item. But then I still have to delete my old post.
It's a bit of a rigmarole, but as you said about the universe... (Plus I don't have to do all the links, and you know me, sometimes I have a *lot* of 'em!). Thanks. |
|
Aug-01-14
 | | chessgames.com: I use the back-page technique myself. It's a labor to say the least. I suppose that teaches us all to be very careful when proofreading our posts. What <zanzibar> is asking for would have to work like this: We keep a table called RecentKibitz, that holds the raw kibitzing text of the past hour. The "delete" link will have a new link under it that reads "edit". Pushing it will bring you into a preview page with the raw text available again for your editing. I've long believed that if you want to edit something it should be reposted, reinserting itself into the chronological stream, otherwise imagine the shenanigans some people would get away with. Insult somebody then quickly turn it into a compliment. Pretend to have predicted all of Carlsen's moves before he played them. There's no end to the tomfoolery. However, I think there should be a very short grace-period, perhaps 2-3 minutes. We all know that feeling you get when you proofread something ten times, but only after you click "POST" do you see a glaring mistake. It could be done with a bit of work; it's a matter of software engineering. |
|
Aug-01-14
 | | OhioChessFan: I also use the back page technique, on that <very> rare occasion I make a mistake of some kind. You do have to get back to the forum page itself before pasting in your copy or you'll get a mean warning that says you can't do what you're doing. If you do the edit thingy, I'd agree with a very short window of opportunity, and not at all on live games. |
|
Aug-01-14
 | | SwitchingQuylthulg: My instinct says the "edit" button, if introduced, should work by deleting the original and treating the edit as a completely new post - essentially what we're doing now, just automated. In that case there would (at least in theory) be no downside to introducing an edit button, as the only effect would be to cut down the number of steps required. The practical problems would be that 1) most people wouldn't expect that kind of behavior from an edit button, and 2) sometimes one makes two or more posts to the same page in quick succession and then finds a silly problem with the first post, necessitating more than one post deletion. In a case like that it would make more sense to have the edited post keep its original reply ID and chronological position. |
|
Aug-01-14
 | | Annie K.: Most boards that allow editing add an automatic 'edited' stamp to the end of every post that had been edited - this serves well to limit the "believability" status of retroactively edited text. On the whole, though, I would prefer the option to just get the raw text (particularly links and FEN) from posts to repost with corrections as a new post, rather than enabling back-editing. This is a chess site, after all - we don't take moves back on the board (I should hope) :p - and likewise should not expect to have a right to change what we have already posted. The ability to delete and repost is enough. :) |
|
Aug-01-14
| | chessgames.com: <My instinct says the "edit" button, if introduced, should work by deleting the original and treating the edit as a completely new post - essentially what we're doing now, just automated.> Exactly. And then there's no need for Annie's suggestion to mark edited posts with an asterisk etc. |
|
Aug-01-14
| | chessgames.com: Boy, this social media experiment has reminded me why our discussion forms are egalitarian, lack "like/dislike" buttons, and are only presented in chronological order. Case in point: we post a daily puzzle on Facebook. Somebody on FB falls for a red herring and leaves an incorrect answer in the comments. Then others come, say "yeah, that looks about right", and give them "like". Soon the incorrect answer floats to the top, where it gathers even more likes (because few people read through the entire list of comments). Somebody may reply and correct them, but it's too late: once it has gained momentum it will be locked at the top. It strongly implies, "this is the right answer", even though it's nowhere close. Chess is all about the pursuit of truth, and truth is not a popularity contest. |
|
| Aug-01-14 | | zanzibar: <chessgames> Yes, such hijinks might be possible, though I think most people operating at that level actually prefer to admire their insults. You could do similar things with "delete" too, but more clumsily. Here's how <ChessTempo> does it. It allows a post to be edited (and deleted) until another subsequent post "locks" in in place. That always struck me as a good compromise. |
|
Aug-01-14
| | Annie K.: <cg: <Boy, this social media experiment has reminded me why our discussion forms are egalitarian, lack "like/dislike" buttons, and are only presented in chronological order.>> Did I tell you lately that I love you? ;D |
|
Aug-01-14
| | chessgames.com: <Annie> Aww, you're so sweet. <zanzibar> I can't imagine what would happen if a new post was made while somebody was editing a previous one. |
|
Aug-01-14
| | Annie K.: Thanks. ;)
Exactly what I was thinking - sounds like a really clever idea, and it's one I haven't seen in action yet - except for the point that people could be putting a lot of work into rewriting a post they found a problem with, only to find out when they tried to repost it that it's too late. Then again, in that case they could just copy their improved post to a new post, with explanation. It's not a bad idea, really. |
|
| Aug-01-14 | | zanzibar: <chessgames> As Annie says, you just post a new version (using cut-and-paste, or maybe just doing a new post altogether). [Update - just can back and could have re-edited]
I also remember now that <CT> pushes a big fat <WARNING> message to you if you're editing a post while a new post is received. (A header label is updated asynchronously at the top of the edit page.) So you know right away. |
|
| Aug-01-14 | | zanzibar: <Dubai Rapid WC 2014> I have lots of issues/questions on this one. Let's start here: What about clean up these duplicate games on <CG>?: Gabriel Sargissian -- Mohamad Al-Modiahki (E63) 15
Gabriel Sargissian -- Mohamad Al-Modiahki (E63) 47
Ivan Salgado Lopez -- Tahir Vakhidov (C42) 17
Ivan Salgado Lopez -- Tahir Vakhidov (C42) 30
Das Debashis -- Ivan Sokolov (E21) 14
Das Debashis -- Ivan Sokolov (E21) 28
It would seem the earlier versions the the games (<= 17 moves) were truncated, and the games with more moves are the correct ones to keep. My question is how <CG> allows duplicates with identical game headers into it's database. Aren't they flagged at submission time? |
|
| Aug-02-14 | | MarkFinan: Still can't see the games in desktop view. #Justsaying ✌ |
|
| Aug-02-14 | | thegoodanarchist: <chessgames.com>: There is an error on the AA-ME rematch page Euwe - Alekhine World Championship Rematch (1937). Your summary states that <Euwe won the first game with the <<black>> pieces...>. But the first game link goes to
Euwe vs Alekhine, 1937, where clearly Euwe has the White pieces. |
|
Aug-02-14
| | chessgames.com: <What about clean up these duplicate games on <CG>? ...> You are right, they need to be merged. Later today. <It would seem the earlier versions the the games (less than or equal to 17 moves) were truncated, and the games with more moves are the correct ones to keep.> I'm sure you are right. <My question is how <CG> allows duplicates with identical game headers into it's database. Aren't they flagged at submission time?> No, it's not a general rule. There have been many matches and tournaments in which this kind of logic would fail us. Consider the short draw Kasparov vs Kramnik, 2000. (Some say that Kasparov was mentally defeated in round 13, and just gave up hope tying to make progress against the Berlin Wall.) It wouldn't be impossible that in the same match we see a longer game of the same variation. It was be a bit hasty to brand the shorter one as a dupe. So it's not a bad idea but I don't see it as very safe. It will return on rare occasion a false positive. |
|
Aug-02-14
| | chessgames.com: <thegoodanarchist> Yes that is an error, even the score table below shows Euwe had white. |
|
| Aug-02-14 | | zanzibar: <<My question is how <CG> allows duplicates with identical game headers into it's database. Aren't they flagged at submission time?> No, it's not a general rule. There have been many matches and tournaments in which this kind of logic would fail us. Consider the short draw Kasparov vs Kramnik, 2000. (Some say that Kasparov was mentally defeated in round 13, and just gave up hope tying to make progress against the Berlin Wall.) It wouldn't be impossible that in the same match we see a longer game of the same variation. It was be a bit hasty to brand the shorter one as a dupe.> I'm wondering if we're talking about two different kinds of duplicates here. There's the kind which is looking at the moves for the game, disregarding the rest of the header info. I think that's what you're referring to - where lots of short games might get tagged if just matching moves. I'm thinking of a duplicate where the PGN header tags match and might, or might not, have the move list match. For example, if <CG>'s software had the "concept" of a tournament - it could flag duplicate games much more restrictively. E.g. for a single RR only one player-pairing would be allowed. |
|
Aug-02-14
 | | WCC Editing Project:
My dear good <thegoodanarchist> Here is a preview of the new WCC intro to the <Euwe-Alekhine 1937 Rematch> which you or anyone else can look at: Game Collection: WCC: Euwe-Alekhine Rematch 1937. We are gradually replacing all of the WCC intros with new drafts, many of which are written but not "finished" in terms of vetting. Here are the new intros already on the official WCC pages: Steinitz - Gunsberg World Championship Match (1890)
Lasker - Janowski World Championship Match (1910)
Lasker - Capablanca World Championship Match (1921)
FIDE World Championship Tournament (1948)
Botvinnik - Bronstein World Championship Match (1951)
If you or anyone else wants to help us with research or especially with writing new drafts, please drop into our forum. |
|
| Aug-02-14 | | zanzibar: <RE: Dubai Rapid WC 2014> OK, there are the previously discussed issues.
But in addition:
(A) There are 25 games with the same header info, but where FIDE contains additional moves. E.g.
2014.06.17 (R7.34) = Maxim Matlakov -- Jahongir Vakhidov (E62) 21 2014.06.17 (R7.34) = Maxim Matlakov -- Jahongir Vakhidov (E62) 167 Looking at a sample, the <CG> games typically end with an even eval, whereas the FIDE games show a clear winner on the board. As expected if the latest FIDE is the "correct" version. (B) There are 227 games missing from <CG> which FIDE has. 1) 226 of them are "stub-games", with no moves, only PGN header. 2) There is one missing game with a movelist:
2014.06.17 (R8.51) 0-1 Basheer Al Qudaimi -- Abdulla Faisal (B43) 36 (I think I've submitted it, but can't exactly remember). * * * * *
Now, if <CG> doesn't do stub-games, this raises a substantive question as to the updating of leader board. Presumably, this could involve updating the leader board by hand - for 2x226 entries. As it turns out there are only 108 unique players involved in the missing games. Still, that's a lot of work to do by hand. I suppose the official FIDE leader board could be copied over and shoe-horned in. But it does show a possible great use of having stub-games, at least until the leader-board is constructed. |
|
Aug-03-14
 | | Stonehenge: I believe E Guo vs J Wanjiru Wambugu, 2014 = Riya Shah vs B Dekic, 2014 and the other way around. S Yu vs Thitu Winfred, 2014 is the right game, not S Yu vs Thitu Winfred, 2014, which is G Jumba vs I Berezina, 2014. That last game has the moves of S Yu vs Thitu Winfred, 2014. |
|
Aug-03-14
| | Stonehenge: ...and there are more wrong games. |
|
Aug-03-14
| | chessgames.com: I just swapped < E Guo vs J Wanjiru Wambugu, 2014 = Riya Shah vs B Dekic, 2014 > but haven't looked at the others. Are you getting your information from https://chess24.com/en/olympiad2014... or is there a better source? |
|
Aug-03-14
| | Stonehenge: I looked at chessbomb.com and checked it with chess-db.com |
|
 |
|
ARCHIVED POSTS
< Earlier Kibitzing · PAGE 729 OF 1118 ·
Later Kibitzing> |
Kibitzing
