Perpetually re-"editing" common knowledge
Join LibraryThing to post.
I've complained about this elsewhere but I thought I'd officially create a bug for it, since it's still happening. In fact, I gather a lot of people are working on cleaning up the Frommer's books because of how clogged my news feed is.
To get to the point, whenever anyone edits any field on a book's common knowledge, every other edit anyone else did on that book gets re-attributed and shows up in their news feed. Here's last Tuesday's screen shot (though it's continued since):
Can this be fixed? If it can't, or if it won't, at least can CK have a sub-feed like I suggested back in early July?
Actually, I think this is happening not when someone edits any CK field but when someone combines the work.
There is a bug report here:
See particularly here, after which Tim acknowledges that that is the problem:
On October 4, someone combined the work you have listed there:
12pm-1pm EST (17:00-18:00 GMT)
Collectorator combined Northern Italy: Including Venice, Milan, and the Lakes (Frommer's Complete) by John Moretti, Frommer's Northern Italy: including Venice, Milan & the Lakes by John Moretti (see work)
Collectorator combined Frommer's Northern Italy with Venice & the Lakes by Reid Bramblett, Frommer's Northern Italy: Including Venice, Milan, and the Lakes (Frommer's Complete) by John Moretti (see work)
Collectorator combined Frommer's Northern Italy: including Venice, Milan & the Lakes by Reid Bramblett, Frommer's Northern Italy: Including Venice, Milan, and the Lakes (Frommer's Complete) by John Moretti (see work)
Collectorator combined Frommer's Northern Italy, Frommer's Northern Italy with Venice & the Lakes by Reid Bramblett (see work)
Here's the CK changelog (history) for the original work number, where you entered the CK in 2010:
Here's the CK changelog for the current work number, which resulted from those combinations:
What's happening is that it's reentering CK data when something gets combined. Initially it was attributing it all to the person who combined it; now that part of the problem is fixed, but it's instead reattributing it to the person who originally entered it.
Yeah. The cause is known and is good - but this unintended consequence is making the (shiny new!) Feed pretty much useless. I don't look at my feed much, and I apparently don't do much CK on things that still need work, so it hasn't been a problem for me...but it's a major annoyance for specifically those people who do a lot of work for LT (and who look at their news feeds, of course).
Edited Common Knowledge on Nicholas Nickleby by Charles Dickens under "Series"
Yesterday, 4:54pm | Share this
Edited Common Knowledge on Nicholas Nickleby by Charles Dickens under "People/Characters"
Yesterday, 4:54pm | Share this
Edited Common Knowledge on The History of the Church by Eusebius under "Series"
Yesterday, 1:10pm | Share this
Edited Common Knowledge on The History of the Church by Eusebius under "Publisher Series"
Yesterday, 1:10pm | Share this
This all appeared in my News feed today and I've done nothing to these works now (or ever, I think).
Yeah, I really wish they would fix this. It's reattributing to you edits you've done in the past, probably because someone combined the work.
You're listed as having deleted some info from the series field in Jan 2010 (it had said Everyman's Library before), and as having entered or deleted something from the character field in Oct 2010 (unclear what, since the history only shows the top line of any field).
Bump. I've noticed it happening again, just not quite to the same scale.
Hi! There might be some reasons why this happens. Assume there is a work AAA in language A with CK A1 made by U1, i.e. A1-U1. Similar is A2-U2, A3-U3 and a work BBB in language B with CK B1-U4, B2-U5.
Now someone combines these works. The work page is either AAA or BBB. Let*s say it is BBB. This means that CK attributions from AAA needs to be duplicated in work BBB in order to be visible.
From now you have only one work BBB with a CK GUI and an CK url showing both histories. The history of AAA is still in the system. If you know what CK parameters you need for a manually generated a proper CK search you find all these duplicates for A1.U1 or A2.U2 or A3.U3.
In the CK search results the icon at the far left is a link to the relevant CK history either CK history of AAA or CK history of BBB.
One problem might be that you can not clear the CK from AAA from the system. It will be visible in all CK search results for matching these values.
The problem is also CK maintenance. Assuming you need to move a series to a publisher series. CK search does not indicate which are the actual works and what is accumulated history.
For example you may need to remove A3-B3 from CK(BBB) but as far as I know you may not remove it from and from CK(AAA). At least does the LT GUI not provide always direct methods to do this.
Bump. With the new focus on sharing and The Feed, this needs to get addressed. I just noticed it happening to me recently, and yes, it is quite obnoxious.
May I draw The Staff's attention to the fact that my original post was in October 2011, more than a year and a half ago? This has been a problem for a very long time. I'd haul in a donkey à la John Green's "long ass time" line on Crash Course, but I'm hoping the reference will be sufficient*.
* Unless, of course, you're the Mongols.
While this may not be working the way you like, it's not a bug. That's why it hasn't been fixed. It's working as it was designed to work since 2009 when CK was debuted. We even mentioned it a few times back at the debut.
The idea is that the last person to edit a record is also "approving" the previous edits.
Now, before you go crazy on me, I don't agree with that notion and I made it known way back when. But that's the way it's always been and it's not a bug.
That being said, I'll dive into the code and see if there is a way that I can unravel the attributions (the data isn't designed to keep track of it in that way) because we have a new feature that will be coming out soon that will work better with separated attributions. It's a good excuse.
That's not actually what's going on, though, or at least not the only thing; my feed is full of CK edits for works and authors that I've never even touched, or not for a very long time. Most recently it says I edited the Series information on http://www.librarything.com/work/5247196, a single-copy work (with no series information, though whoever did edit it may have removed it).
I know that "I did not do this" is not considered reliable information here, but I certainly didn't do anything to that book this week - I don't remember ever touching it, though I may have years ago.
In general I agree that this is obnoxious, and the way it is worded is just plain wrong. I didn't just edit CK; someone else just edited CK that I did maybe a couple of years ago, and have no idea any more what it was about.
On the other hand, for whatever reason, I looked at one of these more closely recently, and discovered that there was a bunch of combining needed and that whoever had updated my CK was understandably confused about the way the related works had once been divided. So in that case it was actually useful. (The old CK was also very out of date, having been written before some major LT update.)
I have no objection to seeing some sort of notification that my CK has been changed, in fact sometimes it could be a good thing - although with the danger of fullfledged edit wars breaking out. But 'share' is not the place for that. Also if there is a notification, it would be useful to also see when I had written the CK, and what it had said before this new edit. I really don't remember all the details about things I sorted out a few years ago.
11, you said, "The idea is that the last person to edit a record is also "approving" the previous edits."
If that were the case, then the last person to edit the record would have their name shown, and no reference to the previous editors would be generated.
Something is all fouled up in the logic here.
It may be that I edited something in 2009, and then someone else re-edited it or combined/separated the work and that somehow caused my edit to show up again on my feed. If that's true, then it's not completely misattributing things, but the date is still wrong - I certainly did not edit those works on those dates.
I'll look into this and figure out what's causing the feed to be pinged when new updates or combine/separates happen. I would guess it's the latter that is causing the problems.
I'm still not convinced that I ever edited some of the works in question. My feed claims I added CK to "The Life of Lazarillo de Tormes" on Oct. 23. While there were some CK edits in the logs on that date ( http://www.librarything.com/commonknowledge/search.php?q=%22Lazarillo%20de%20Tor... ) there aren't any, anytime, that are attributed to me. They're probably being re-pinged because of a combination (which will scroll off the one-week helper log soon, but I can see it now, on that date by arjaygee) but I have no idea why they're being attributed to me.
Agreed, it's likely due to combinations and separations of the work. It's a sticky issue but we'll figure it out.
Has this been fixed? I haven't seen any spurious CK "edits" in my feed since Oct. 30, which is encouraging.
Here is an example of how it is not fixed.
This evening I combined two works into one. Just now I saw the work in the CK. It says:
catalog Today: 6:50:46 pm
The past we sh…erican peoples
original publication date eng
profile catalog NinieB
Presumably, NinieB's feed would show her editing that work today, even though it was just me combining.
catalog Today: 6:32:58 pm
The gun that s…erican destiny
media reviews eng
Hobbies Magazine, Rhoades, Orille B.
profile catalog MsMixte
If you look at the history, paulstalder and I are both credited with two changes that I did not do. I went there after seeing it, and did remove a German publishers series from series. (All sorts of things were done by several people then)
I think the entries on January 18th are probably correct.
http://www.librarything.com/series/Die+Märchen+der+Weltliteratur is what I removed. The description there is by me.
Just now I added Diederichs Märchen der Weltliteratur to the German publisher series and deleted Grimms Märchen. Vollständige Ausgabe from German series. That is not what the history shows. The deletion is right, but it shows me entering different content.
News Feed today is right.
You mention January 18; are you seeing any incorrect data more recently than October 28? That's the latest that I see bad data in my own feed. I wasn't expecting a fix to be retroactive, so I'm not regarding the fact that I still see that as indicative of anything.
My feed says that I edited Common Knowledge for a book today at 4:28am. I was blissfully asleep at that hour.
I also have a couple edits from Nov 5 and Nov 3 that were not me.
24>See 22 and 23. Yes. Day before yesterday.
Nov 9th, 2013: 4:17:25 pm
The Complete G…'s Fairy Tales
original title eng
Kinder- und Hausmärchen
Edited Common Knowledge on Bunter the Caravanner by Charles Hamilton under "Series"
Yesterday, 5:20am | Share this
This appeared on my News Feed. I checked the series page and I am apparently the only person who has ever edited this series - I have a helping Score of 57 - and I last did some worked at 5.20am today. I am quite sure I have never visited this book page, until I checked the above entry, as I don't have a copy of this book or any others in the series, nor any other by this author. Previous phantom editing I could have accepted if I had ever done any editing for the CK of a book but I am quite sure I have had nothing to do with this book.
abbottthomas added author Chapman, C H to Bunter the Caravanner (Illustrator, secondary, some editions)
I found that in the logs.
Go to CK for the book, and at the top right 'view history'.
Yes, ten seconds later you edited series - except we know you didn't.
33, exactly. I am working on those NIVs and just did a big combo. So everybody's feed gets a bump.
Yeah, I got that one too. I did some work on Bibles a while back, but gave up since people added crap faster than I could fix it (stuff with the cover color in the title, etc.) Glad to see others are still fighting the good fight there.
I actually own several dozen Bibles, and only have one or two entered - if it ever gets to be less of a swamp, I might enter the others.
You are in the CK history as having added the series info to a work page for that work back in 2010; that work page has since (yesterday) been combined with another, which is why your 2010 edit is reposting:
If you're saying you didn't, back in 2010, add anything to the series page, it might be that you combined the work. Up until a certain point, all CK (for all relevant work pages) was being attributed anew to the person who last combined, and then at some point it switched to repeatedly counting it for the person who had credit for it before the combination.
#37 I've done plenty of combining so that's possible, but my name is on the series page as having done everything - there's no-one else - and I certainly haven't done that and don't deserve the credit. I should think the person, or persons, responsible for all the work must feel rather hard done by.
Edited Common Knowledge on Mortification: Writers' Stories of Their Public Shame by Robin Robertson under "Publisher's editor"
This appeared on my news feed this morning - a book I've never before heard of, edited by an author I've never heard of so you can be sure it isn't in my catalogue either. I'm sure I have books by some of the contributors but I really can't think of anything to link me with it in any way.
This topic is part of LibraryThing's in-talk bug tracking.
Join or watch Bug Collectors to get "Bug Tracking" under "The World" in Talk all the time.
Category: Common Knowledge
Assigned to timspalding
Reported by AnnaClaire
Oct 10, 2011, 1:43pm
2 years since last change
Reported. AnnaClaire (Oct 10, 2011, 1:43pm)
This topic is not marked as primarily about any work, author or other topic.