< Earlier Kibitzing · PAGE 595 OF 595 ·
|Oct-19-16|| ||Tabanus: Yes Smyslov - Hübner Candidates Quarterfinal (1983). On my work computer: HP EliteDesk, Windows 7 Enterprise, Google Chrome.|
Also after reboot. But when I switch to IE, it works!
|Oct-19-16|| ||Tabanus: Albin - Marco (1901) also has the " instead of 1/2.|
|Oct-19-16|| ||chessgames.com: OK thanks. Please, nobody edit those pages for now, I want to carefully examine those pages first. Editing may have side-effects.|
|Oct-19-16|| ||Tabanus: <CG> Before leaving work, I noticed that it (the 1/2) occasionally was fine after refreshing the page (but only about 1 of 3 or 1 of 5 times). But clicking on the refresh button again, it was back to ".|
|Oct-19-16|| ||zanzibar: <chessgames.com: OK thanks. Please, nobody edit those pages for now, I want to carefully examine those pages first. Editing may have side-effects.>|
<CG> should have a lock-down mode for freezing edits under such circumstances.
In other words, you shouldn't be asking, you should be notifying that edits are locked out (for a set duration or until further notice).
|Oct-20-16|| ||Tabanus: <Editing may have side-effects.>|
One being that my editing of Schlechter - Janowski (1902) caused no less than 196 versions to be saved, see Edit History: Schlechter - Janowski (1902).
|Oct-21-16|| ||chessgames.com: <196 versions> I don't believe we've discussed this yet, but the idea is NOT to save every single version of every edit. If somebody comes in and inserts a comma where one belongs there's no reason to archive that for all time.|
To address the proliferation of versions, this there is a program called
"pruneversions" which is run daily to remove anything except the "milestone" entries. When big things change, or when the editor changes, or under various other circumstances, we preserve a version virtually forever. But the milestone versions are but a small subset of the hundreds of even thousands of versions which are likely to be produced.
Why then, you may wonder, are there almost 200 versions of Schlechter - Janowski (1902)? Is pruneversions not working?
No. I thought it would be prudent to put a clause into pruneversions so that nothing in the past 7 days is ever deleted, no matter what. This way when the article is being currently worked upon, there is no chance of this software removing something that might be of short-term value.
But once the edits are over a week old they are all candidates for deletion.
|Oct-21-16|| ||OhioChessFan: <cg.c: If somebody comes in and inserts a comma where one belongs there's no reason to archive that for all time.>|
I noticed that.
|Oct-22-16|| ||Tabanus: <CG> That sounds great. For the 30,000 or so bios ;)|
Ohio probably wants <If somebody comes in, and inserts a comma, where one belongs, there's no reason to archive that, for all time.>
|Oct-22-16|| ||OhioChessFan: I already experienced it. It doesn't show up on the recent edits list. I suspect that was the reference cg.c was making, and not a mere hypothetical.|
|Oct-22-16|| ||Stonehenge: <CG>
I have just added a link to Wikipedia here but it doesn't work properly:
|Oct-22-16|| ||chessgames.com: Yes, we lost that functionality, see chessgames.com chessforum (kibitz #26261).|
|Oct-22-16|| ||Stonehenge: OK
Two penalty points and don't pass Go.
|Oct-22-16|| ||chessgames.com: Sorry. Believe me, it bothers me to no end. I was ebullient when I finally had it working — but then, on the day of the server move, I fractured it like one would drop a vase.|
Of course I have backups, but that's the problem—a dozen versions of the code, no one of which actually works. There must be some combination of the little changes from each file than can be assembled in the most delicate way will bring it back.
Anyhow, I'll let you know when I rediscover the solution, but here is the current state of affairs:
Things worked like they always have, except (1) the "\ forces a line-break" feature works, and (2) wikipedia links now break in a way different than they used to.
|Oct-22-16|| ||chessgames.com: <OhioChessFan: I already experienced it. It doesn't show up on the recent edits list> To be clear, any edit you make, no matter how trivial, will show up on the recent edits list for 7 days. After that, the minor changes may be discarded while "milestone" changes will be preserved.|
Note that the milestone changes will incorporate within them the previous smaller changes, if you see what I mean. For example if the version archive contains this
<OhioChessFan Oct 8, 2016 5:38 — Deletes a comma.>
<OhioChessFan Oct 8, 2016 5:42 — Inserts a giant paragraph.>
The first version will be removed, leaving us with this
<OhioChessFan Oct 8, 2016 5:42 — Deletes a comma and inserts a giant paragraph.>
Also note that if the two editors were different, the edits would not be merged. We wouldn't want somebody accusing OCF of being anti-Oxford comma if he's actually pro-Oxford comma!
|Oct-22-16|| ||OhioChessFan: <To be clear, any edit you make, no matter how trivial, will show up on the recent edits list for 7 days. >|
No big deal, but are you sure? I am pretty sure on one edit I literally only added a comma and it didn't show up on recent edits. Maybe I didn't save it.
|Oct-22-16|| ||chessgames.com: <but are you sure?> Fairly sure. If I'm wrong, something is horribly broken. <Maybe I didn't save it.> That would explain it. Feel free to try to reproduce that effect; I am guessing you won't be able to.|
|Oct-22-16|| ||OhioChessFan: I just added a comma and it showed up in recent edits. I must not have saved the last time.|
|Oct-22-16|| ||chessgames.com: Here's some good news, I think I've re-engineered the "wikipedia hook", testing now.|
It's Frank Marshall approved. I'll go test a biography next.
|Oct-23-16|| ||offramp: Janowski - Marshall, Match 5 (1916)|
The intro, "Marshall drew the 8th and last game", and the game list do not agree.
|Oct-23-16|| ||MissScarlett: Does anyone have <Pasadena 1932 International Chess Tournament> (The House of Staunton, 2011)? Apparently, it has seven game scores not in this collection: Game Collection: 1932 Pasadena|
What are they? By when can you upload them? Thanks!
|Oct-23-16|| ||Tabanus: <offramp> Fixed. One of the games had month July instead of June.|
|Oct-23-16|| ||Tabanus: <Maybe I didn't save it.> I just updated a bio ("edit write-up") (to see if the Notes list was affected by the \) <without> changing anything. Even then it shows up on the Recent Edits Page. And it seems the \ now works both with and without space between it and a link.|
|Oct-25-16|| ||chessgames.com: MissScarlet wrote: <Happy to be on board, but unhappy to find I can't edit the Event field. For the purpose of normalising tournament collections, that's imperative.>|
I'm not against that feature but some discussion/planning needs to take place. We need to define exactly what we mean when we say "edit the event field." Believe it or not, it's not that obvious. We aren't just editing text files here, we're editing database tables.
For example, suppose there is a game that reads "Moscow" and you come along and rightfully point out that one game is better described as "Moscow 50 Board Simul" with a site tag of "Moscow RUS". OK, sounds fine.
Now there is a tournament table with an entry of 10053 that says "Moscow" with 100 games associated with it. If we simply change that single record to read "Moscow 50 Board Simul" we change 99 other games—many of which were obviously not part of this simul.
So we could make a new tournament record, call it "Moscow 50 Board Simul", and assign this singular game to it. Meanwhile another editor finds a game from the same event and types in "Moscow 50 Board Sim". Also perfectly acceptable, but since it's not a byte-for-byte match with the previous tournament, a new record will be created. The job of normalizing the tournament names has been damaged as they are no longer normalized.
Meanwhile in other situations, perhaps we find a tournament simply misspelled or we don't like the transliteration, e.g. "Moskow 50 Board Simul." Here we want to just fix the spelling of the record that exists and not create a new one.
All the while, you have the tool to fix the problem once and for all in a very rigorous way. You could create a game collection containing all of the games from this hypothetical 50 board simul, put them into it in whatever order you think best, and put it up for a vote at the Biographer Bistro. If you do it that way, you are guaranteed to see it normalized in the end, with all sites and events being "byte-for-byte identical."
So while I agree that some tools could be expanded, and there are cases where creating a tournament and putting it up for vote is just a waste of time (or even wrong), what you are asking to do is not impossible with the tools in your hands at this time.
|Oct-25-16|| ||MissScarlett: <Now there is a tournament table with an entry of 10053 that says "Moscow" with 100 games associated with it.>|
You lost me. What if there are already ten tournaments with the Event field designated 'Moscow' or a thousand separate 'Simul's? How are they distinguished from each other at present?
<You could create a game collection containing all of the games from this hypothetical 50 board simul, put them into it in whatever order you think best, and put it up for a vote at the Biographer Bistro.>
Most tournaments are only represented by a fraction of the games, rendering them unsuitable for formal tournament collections. But for anyone looking through the DB, normalised Event names would be a big help.
< Earlier Kibitzing · PAGE 595 OF 595 ·