Collectorator: Explain your edit wars

I'm starting this thread to get Collectorator to explain any reversing or warring edits before they do them. They are to explain them step by step, so anyone can understand what they're doing and why.

I have informed them that, if they do not post here first, I will suspend their account. We need to get to the bottom of the argument they have with everyone's edits. And if they are unwilling to do so, they will be suspended for as long as is necessary.

Once they have posted some proposed edits, will then discuss the case, to see if we can come to some sort of understanding about process and what the correct actions really are.

Other members are instructed to do nothing on these edits either. The relevant authors and works are to be considered frozen until we hash the matter out here.

I ask members to be respectful in this thread. If a member fails to do so, I am going to treat it as a Terms of Service abuse.

>2 Collectorator:

It would be great if you could explain something you plan to do and why.

>2 Collectorator: Tropical Fishes as Pets was on that author page to begin with. I make it a practice to help new members who are authors to navigate the site. Part of that is to look at the author page before or after they’ve claimed it and make splits as required.
So some member(s) with that book had already catalogued it without the middle initial.

Three of the four members’ author links now go to the wrong page.
I maintain that my edits were correct.

Was Christopher W. Coates, rather than Christopher Coates, edited as the primary author of Tropical Fishes as Pets? It looks like the latter should be the calculated one.

>4 Collectorator:

Rufus Jones was likely split because there are multiple "authors" on the page.

There are books by and about the Quaker historian and theologian and then there are films in which the English actor with the same name appeared.

>4 Collectorator: see...

https://en.wikipedia.org/wiki/Rufus_Jones (multiple writers)

Okay, so my question for collectorator is, why de-split the author?

The author is evidently--indeed, pretty self-evidently--more than one person. Why de-split. We should insread figure out which works belong to which Rufus Jones, right?

>2 Collectorator:

You're omitting the initial step you took: assigning "Christopher W. Coates" as the primary author to Tropical Fishes as Pets. https://www.librarything.com/log_helpers.php?view=otherauthors

I find it difficult to credit this omission as anything other than dishonest. You may think the way to resolve split authors is to reassign authorship of specific works to a more specific author page, but that's not the policy as it stands:

"Changing a work's primary author on the "work" (ie., global) level is serious business, and should only be done when truly necessary."

This was not truly necessary, and was done in the face of other users following the actual policy that does exist.

I don't like the Coates example because the water has already been muddied with action. We can get into it, but I'd prefer if we talked over proposals to do things, not things already done.

>11 gabriel: I find it difficult to credit this omission as anything other than dishonest\y\.

Let's try to be nice here, please.

>12 timspalding: That's often how it starts though. C forcing authorship to suit their vision of how things ought to be is exactly what led to the "Unpleasant Run-In." Check the first post of that thread again.

>12 timspalding: But it's an example of a practice: force-edit the primary author so that it's not assigned to a split, then claim the split has no works associated with it and undo the split. Is it a desirable or even acceptable practice?

I'm still waiting to hear about Rufus Jones.

So, collectorator, is this the idea: De-split authors by assigning all their works away, so no split is necessary?

>15 Collectorator:

I'm not entirely sure all of the factors and their weighing. In theory it's the most common form of the author on the works by the author.

>18 Collectorator:

Well, it's clear he isn't one author. He needs research to identify who the splits are. Someone did a small amount of work--fingering the name as a problem requiring investigation. I don't see why we should reverse that. We should build on it.

>21 Collectorator:

"One click"? How so?

You have to click on "edit the division", click on "divided author" and then assign however many works there are to the correct authors, which could involve quite a few clicks, and then click on "save the page". However you figure it, that's more than one.

(I see that someone did split the page, and aliased one author, despite the admonition that "Other members are instructed to do nothing on these edits either.")

>12 timspalding:

My apologies for my conclusory language. However, it does remain the case, that when asked to account for his/her edits, C omitted the essential predicate for unsplitting that particular author page, and the only really controversial step. Unsplitting the author page would indeed be correct if that work had not naturally been assigned to Christopher Coates.

>21 Collectorator: Richard Chizmar is listed as the primary author, as Editor, in the other authors section for the work.

https://www.librarything.com/work/10069402/summary

Chizmar is one of the editors. He's also the magazine's founder.

https://www.cemeterydance.com/cemetery-dance-59-keene-vernon-langan-rainey.html

https://richardchizmar.com/biography/

So what's the problem?

>25 aspirit: For one thing most of those Main Authors should be Secondary, as Contributors. no?

>26 Crypto-Willobie: that section for the work does need cleanup, but Collectorator was asking specifically about the primary author field.

>21 Collectorator: As >25 aspirit: notes, Richard Chizmar is in the "Add/edit other authors" section as a primary author. Someone added it and/or confirmed it. That's how.

Now, I think we have a problem here in that we need a history for this section, so members can see who did it.

Does this answer your question? Are we all on the same page here?

Scroll down to "Add/edit other authors" and look at it. That's what's making the author what it is.

You want to know what it would be without that? The most common author. It's not the shortest-named author.

If I understand your question, I'll dig into the algorithm.

>32 Collectorator: Why? That author needs to be split anyway. What is the benefit of editing the primary author?

>34 Collectorator:

1) what makes you think the author name was changed? it just says 'confirmed,' and as far as i can tell, no name change has recently been logged.

2) this book was published with the author listed as Thomas Horn, which is also the name this author generally uses. why should it be changed to Thomas R. Horn?

Collectorator, thanks for the suggestions but there are several unresolved issues here. I think we need your response to >5 2wonderY:, >10 timspalding:, >35 SandraArdnas:, >38 I-_-I:, and especially >16 timspalding:. Otherwise we're just dancing around the reason this thread exists in the first place.

>41 Collectorator: Do you intend to discuss any of this when people have questions and comments? If not, it's really a monologue in which you're merely telling everyone what you will do

>42 Collectorator: That work's calculated author should be Thomas Horn, so someone must have edited the primary author.

For some reason 'recalculate for members books' and recalculate title/author' doesn't change it if it's been edited.

>44 SandraArdnas: If someone has manually entered a primary author to override the system selected one, you can delete that with the red x in on the primary authors line in the Edit other authors window.

>45 r.orrison: Indeed, you learn something every day. I also noticed only now that if the primary author has been edited it says 'confirmed' rather than 'calculated'

>48 Collectorator: The other ones simply mean Pope Leo too, and carry the exact same ambiguity.

>50 Collectorator: Regarding system choices when numbers are the same, I think the first entered is chosen (until some other option wins by numbers). It certainly works that way when it picks default covers

>1 timspalding: Is the intention that Collectorator should have to outline every contribution she makes to the site before it's carried out? For example, the combinations proposed in >50 Collectorator: are very straightforward and I think she should be allowed to continue contributing unhindered and without need for explanation on these sorts of things. Where I think the issue arises is with authors/splits/aliases etc.

>52 bernsad: I believe the aim is for her to communicate with other helpers when needed, whether it's splits or something else, but that obviously isn't happening, so yes, it's quite pointless altogether

>54 Collectorator: I agree those should be combined

>55 Collectorator: That's an odd one.

>55 Collectorator: See https://en.wikipedia.org/wiki/Arthur_St_John_Adcock

The ANZAC Pilgrim's Progres: Ballads of Australia's Army (1918) Lance-Corporal Cobber, editor

http://www.gutenberg.org/ebooks/author/42663 has 2 works but not this one.

At a guess, they are different people. He's not on Project Gutenberg Australia.

The Google Books entry seems to confirm this: https://www.google.co.uk/books/edition/_/gjIWAQAAMAAJ?hl=en

Yup, 2 different people: https://www.awm.gov.au/collection/LIB17676

If you zoom in on the title page of the Google Books entry, Lance-Corporal Cobber is credited as the author, and Arthur St John Adcock as the editor.

Going hiking for much of the day. Will circle up with these examples later on.

>52 bernsad: But Collectorator is so far refusing to engage re. the edits we’ve questioned... if that’s how it’s going to be, requiring that they propose (and obtain agreement to) any edits beforehand might be the only way to avoid the editing wars that got us here.

*edited for clarity*

My issue is not the edits per se. It’s the fact that should I undo them, or even just request an explanation, I will be subjected to ritual humiliation by Collectorator.

Okay, here are some I disagree with:

>32 Collectorator:

Meanwhile, this work does not need to be aliased. It needs the author name changed. All possibilities are single editions.
http://www.librarything.com/work/19917371/editions

I don't understand. Why would anything need to be done to this at all? The author's name is Rufus M. Jones. It's true that it has three editions, but it chose one, and as far as I can see from searching for "John Wilhelm Rowntree Rufus Jones," the M is the most common name. Why are we making any changes here at all?

>33 Collectorator:

http://www.librarything.com/author/harrisjulianearle
should not be combined with
http://www.librarything.com/author/harrisjulian&norefer=1
because the first is a split author.

I think I disagree. The author Julian Earle Harris is clearly often known as Julian Harris. If you don't combine them, then
1. Members who have books by Julian Harris will not go to the right place. See for example https://www.librarything.com/catalog/Chrissinger24&collection=-1&deepsea...
2. Some books by the author will simply go to Julian Harris ( example https://www.librarything.com/work/1313766 )

In order to make their link work, you'd have to split Julian Harris up into (1) who was aliased into Julian Earle Harris and (2) who was plane-jane Julian Harris.

>34 Collectorator:

The author name on this work was changed. If it was changed to Thomas R. Horn it would not need to be aliased. It would just go there.
http://www.librarything.com/work/18763475/editions

If the author wasn't changed, the author would be "Horn." That's winning 3 to 1 or 1.

Also, the name of the author as clearly pictured on the cover is Thomas Horn, not Thomas R. Horn.

>36 Collectorator:

>41 Collectorator: http://www.librarything.com/author/metcalfrobert&norefer=2
should not be combined with
http://www.librarything.com/author/metcalfrobertd
because the second is a split author.

I don't really see the force of "because the second is a split author." It's clear that there are multiple people named "Robert Metcalf." If there is an error here, it's that the code is "metcalfrobertd." But one way or another metcalfrobertd and metcalfrobert should be combined. The philosophy author in here clearly mostly goes by Robert Metcalf, but sometimes adds a D in the middle.

And here are some agreements:

I think it would be safe to combine
http://www.librarything.com/work/8027995/editions
with
http://www.librarything.com/work/18236704/editions

I think you're probably right, although it took some looking for me to be assured of it. In any case, both are pamphlets by the Protestant Truth Society.

>37 Collectorator:

I think it would be safe to combine
http://www.librarything.com/work/25026713/editions
and
http://www.librarything.com/work/25026742/editions
and
http://www.librarything.com/work/25026740/editions

Also agree. I'm not sure I'd do such combinations, as it's possible to be wrong, but I think unlikely here.

>39 Collectorator:

http://www.librarything.com/work/23589639/editions
and
http://www.librarything.com/work/25254330/editions
could be safely combined.

http://www.librarything.com/work/19625303/editions
and
http://www.librarything.com/work/22809877/editions
also.

Definitely to both.

>60 Collectorator: I would change the secondary contributor name to James S. Ackerman:
http://www.librarything.com/work/719631

The author is described as James Ackerman on all the blurbs I find. The purpose of the system is to allow people to be called what they are called. The purpose is NOT to change people's names to make them unique, despite what the book says. The point is to allow the author to be what the book says it is, and then deal with it. So, no, I don't agree with this.

I would select the option to change the primary author to James S. Ackerman:
http://www.librarything.com/work/17781861/editions
All names are "ones."

It's tough, but, yes, James S. Ackerman is the most common name by both editions and copies. It was overriden manually into James Stokes Ackerman.

I would combine
http://www.librarything.com/work/9644760
with
http://www.librarything.com/work/20908966

Yes. The latter is ratty data, but it's clearly the same book.

>48 Collectorator:
>49 Nicole_VanK:

I will also chime in on this one: it seems to me that ambiguous names should always be combined, even if the particular name doesn't currently have works belonging to two or more authors. Firstly, many users won't be able to identify where a particular work under an ambiguous author name ultimately belongs - but they can combine the ambiguous author names, which allows other users to see it and assign it to the proper, distinguished author. Secondly, further works may always bubble up on such ambiguous author names, and it's extremely hard to identify and correct wrongly assigned works if they are buried way down on a list of a hundred works under "Leo XIII".

>67 Collectorator:

Let's say you have two authors.

Author number one is named John Smith.
Author number two is also named John Smith, but some of his books are published under John P. Smith.

I do indeed think you should combine John Smith and John P. Smith, and then separate them into John Smith (1) and John Smith (2).

You think otherwise, right?

>68 timspalding: I wrote a longish post on this but my browser somehow swallowed it, so I'll just point out that combining in case of >33 Collectorator: resulted in both authors ending up as Julian Earle Harris, even though it fits only one of them. Please see https://www.librarything.com/topic/325870#7326978

>63 timspalding: In order to make their link work, you'd have to split Julian Harris up into (1) who was aliased into Julian Earle Harris and (2) who was plane-jane Julian Harris.

This is what helpers assumed was official policy and how we normally split and alias

>68 timspalding:

I definitely think otherwise!

As I said somewhere else:

If the John Smith page is split and aliased, then someone who has entered "John B. Smith"'s book as by "John Smith" will get to that page, and see that their author also uses "John B."

But if the pages are combined and then split, someone who has entered the book as by "John B" will be led to the wrong page, one they did not expect.

Often an author who has never published under any name but John Smith is aliased by Collectorator into John R. Smith because research has unearthed his middle name as Randolph and this will distinguish him from the other John Smiths. Why this is preferable to John Smith (10) is a mystery to me.

>68 timspalding: I do indeed think you should combine John Smith and John P. Smith

And what about John D. Smith? And John Reginald Smith? Should they all be combined in with John Smith?

What do you do if John Reginald Smith is the most popular author, and ends up being the main name on the page?

You end up splitting into
John Reginald Smith (1)
John Reginald Smith (2) is John P. Smith
John Reginald Smith (3) is John D. Smith

This is not hypothetical, it's exactly what we now have on https://www.librarything.com/author/harrisjulianearle - "Julian Earle Harris (2) : Julian Harris".

The author page for Julian (no middle name) Harris (2) is now a split page of Julian Earle Harris and says "Julian Harris is Julian Earle Harris (2). For other authors named Julian Earle Harris, see the disambiguation page." Do you seriously think that's right?

The "Split author" page starts out saying 'Authors with the same name, such as "John Adams" or "Tom Snyder" can be divided by specifying the works each individual wrote.' Are you claiming that Julian Earle Harris and Julian Harris have the same name? Whoever wrote that text didn't believe this was what author splitting was for.

(I find myself in stronger disagreement with Tim than Collectorator! Though I do still disagree with Collectorator's practice of renaming authors on works instead of split-then-alias. At least that doesn't generate nonsense like "Julian Harris is Julian Earle Harris (2)".)

>68 timspalding: I would disagree with that, personally. I think that's primarily what aliasing is about. But, in my opinion, the same goes for James Ackerman vs. James S. Ackerman.

Edited because I was earlier referring to another reaction on this thread. My bad, sorry r.orrison

>68 timspalding: In general, by that line of thinking, why not combine all people surnamed Smith and take it from there?

ETA: I'll comply if that's what you decide, but I think it would be silly

>75 Nicole_VanK: I would disagree with that
What would you disagree with? My interpretation of Tim's position, my disagreement with Tim's position, or something else?

>74 Collectorator: Not sure if this is exactly relevant, but i just combined two editions, one of which only listed the author's last name, and the system defaulted to the longer one
https://www.librarything.com/work/12514440/editions

>77 r.orrison: Oops, sorry. I seem to be referring to another post than I meant to do. Will edit when I find the one I did mean.

ETA: reaction numbers seem to change around here. Next time I'll make a screen shot, and if so I'll report in bug collectors.

>79 Nicole_VanK:
>78 I-_-I:
>77 r.orrison:

aspirit made a dedicated thread to the aliasing/combining discussion in Combiners:

https://www.librarything.com/topic/326785

I think we should migrate the conversation over there.

>81 Collectorator:

At its core the algorithm look at all the books that fall under the author, descending to individual member catalogs to do it. It picks the most common one, as if each book in each catalog were "voting."

* CK overrides it absolutely
* Secondary author data on the work level gets factored in as a vote.
* It attempts to chase down all the various splits, combines and aliases.
* If the book has more than 50 copies, it uses partially cached data.

>84 Collectorator:

I'm confused. I didn't leave out "most common." The second sentence is "It picks the most common one, as if each book in each catalog were 'voting.'"

Tim, you may or may not have noticed that Collectorator has been responding only to you, and even then, selectively. Questions that would get to the crux of Collectorator's editing wars are being ignored. I'm not saying this algorithm discussion isn't important, but we're getting sidetracked, and it seems pretty clear that Collectorator will continue to ignore us unless you clarify that dialogue is expected.

edited to add a link to related concerns in another thread: https://www.librarything.com/topic/325870#7329482

>74 Collectorator:

I looked at the two copies on LT of Using Powerpoint F/windows 95 (User Friendly Reference).

Reding was added to LT on 02-01-2008 from Amazon, and contained a ISBN.
Elizabeth Reding was added to LT on 03-16-2008 from Amazon, and contained the same ISBN.

I've seen this happen before... where an incomplete/incorrect entry was added to LT first, and the 2nd/3rd complete/correct entries added don't change the first occurrence. I've always thought it had to do with the first occurrence taking precedence. Lots of users don't know to "Recalculate Title/Author," which sometimes clears some errors, but admittedly, not usually the author.

edited for clarification

>88 Conkie:

When there's only two entires, it has to make a choice. It may well choose the first. It might choose the one with the lower internal id number. But there's no answer.

>89 timspalding: Thank you for the explanation!

>93 Collectorator: I agree I wouldn't be happy with that

>95 Collectorator: Since it's currently attributed to J. Schlosser, I don't agree.

>97 Nicole_VanK: I just looked at that one and clicked on "Recalculate title/author" and it changed to Julius Schlosser. Which is an example of what I was saying in >88 Conkie:

>99 Collectorator:

Why not? The author's LiveJournal and his website refer to him as "Robert A. Sloan":

https://robertsloan2.livejournal.com/
https://robert-sloan.com/

His name on the book is "Robert A. Sloan": https://www.amazon.com/Raven-Dance-History-Utopian-Revolution/dp/0595164730

>100 lilithcat: I think she means the primary author should be with the middle initial, thus going directly to the aliased page with no need for a split one, which would remain empty of works. However, this way both of the members who have it catalogued get them on their author pages. I'm not sure what would be achieved by editing the primary to Robert A. Sloan, other than one less split on Robert Sloan page

>100 lilithcat: >101 SandraArdnas: exactly - aliasing is the best way to handle this.

>104 Collectorator: Would you kindly comment on other members' comment on that. It is extremely strange (and rude) to consistently ignore anything anyone has to say. It is also strange (and rude by necessity) to comment as if you're not here to explain your view of the issue yourself

>112 Collectorator: Rufus M. doesn't need to be split, but Rufus Jones does. I understood that you wanted to change the author from Rufus Jones to Rufus M. Jones to avoid splitting Rufus Jones.

The situation has changed meanwhile because someone did a required split of Rufus Jones, together with clear DN about the three of them, as well as recalculated the work author, which you'll be happy to know resulted in assigning it to Rufus M. (Recalculating also seems to prove shortness of name is not a factor since it actually chose the longest of the three - Rufus Matthew Jones)

However, we can still discuss the principle and come to some community policy. Do we edit the calculated author and in what circumstances? Since in this case the calculated name resulted in unsplitted Rufus M. Jones, I'll pose the reverse question. Would it actually be better if it was assigned to Rufus Jones split, because that way all three of the members see it on their author page, rather than just the two who have Rufus M.Jones and Rufus Matthew Jones?

This is a fringe case with only 3 books catalogued, but these issues arise with hundreds of members not seeing the work on their author page too, so we weigh the decisions among other things by usability to them too IMO. The main point, however, is that when disputes or concerns arise, they are discussed and attempt is made to find the best possible solution for a particular case, and if possible, some general principles for the future. So talk to the rest of combiners when needed. We can't divine your reasoning, you have to state it, and in discussing you might find out it actually helps to clarify things both to you and others.

And thank you for responding

No responses for me though! I suppose C did warn me that anything I wrote would be ignored https://www.librarything.com/topic/325870#7322807
¯\_(ツ)_/¯

>108 Collectorator: I don't believe the shortest name is winning in those examples. The first author name to an edition with an ISBN is chosen. That happens to be Aurel, the shortest name.

>123 Collectorator: to use your phrase...

2 > 1

There are two entries for "Ackerman, James" and one for "Ackerman, James S." Just like before (>122 aspirit:) this does not look like an example of the shortest name winning.

The author primarily goes by "James Sloss Ackerman" or James S. Ackerman", which is what his "James Ackerman" page is pointing to. Can you explain what you mean by the need for a primary author change?

>128 Collectorator:

Looking at the history, it looks like someone made a hash of it.

There was another book on the page by a Canon J.B. Leslie, so someone split the page. Which makes sense, except, they then seem to have managed to get it under James B. Leslie, without undoing the split. And, instead, split Canon J. B. Leslie, and James B and aliased them to each other.

My head hurts!

>131 Collectorator: see, i don't think it always does select the shortest name. we've posted some counter-examples but you've ignored them.

>131 Collectorator: just like another example above, the name that was picked appears to be the first to an edition with an ISBN.

Regardless of the reason why you want to change the author, can you explain how you would in this situation?

>133 Collectorator: possibly, yes. What do you want to edit?

https://www.librarything.com/author/riordanjames

>134 Collectorator: no, "Gioconda. Emilio & Mussolini Willems" shouldn't be combined with "Emilio Willems".

Please look at the unaddressed questions above.

>131 Collectorator:
https://www.librarything.com/work/8808856/editions
Does not need to have system selected author name overriden by Other Author to suit one inidividual's preference.
https://www.librarything.com/author/willemse is correctly split and aliased.
No user will be shorted by this action because their book will link to either https://www.librarything.com/author/willemsemilio or https://www.librarything.com/author/willemse-2 which directs them to the other page.

>133 Collectorator:
Looks like a case of Tim's combine-then-split. As a general rule, I disagree with the practice and wouldn't have done it, but in this particular case Jim's works are very often cataloged as James so I wouldn't have messed with it if I happened across it.

>139 Collectorator: the Stanley Schmidt (2) and Stanley F. Schmidt pages should be linked. How would you like that done?

Also see the other questions that remain unanswered in the thread.

### Agreement:

>91 Collectorator: Agreed. So far I don't think anyone has disagreed with you about work combinations. Anyone disagree?

Note, however that "If you look.at.the.editions it gives you a hint" is unnecessarily obnoxious.

>92 Collectorator: Agreed. Separation. Similar lack of problems.

>96 Collectorator: Agreed.

>105 Collectorator: Agreed.

>121 Collectorator: If you looked a litte and found that, yes, that's the polish, then great.

>125 Collectorator: Yes.

### Thorny:

>95 Collectorator: This work does not need to be aliased:
http://www.librarything.com/work/4782840/editions

I'm confused. The work is not aliased. The author "J. Schlosser" is split--because there are two J. Schlossers--and then one is aliased into Julius Schlosser. This is clearly correct.

What is your proposed solution? Are you proposing to change the primary author on the book? In such a case, where "J. Schlosser" isn't a real name, I'd be fine with that.

>97 Nicole_VanK: See above; what do you think?

>99 Collectorator: Can you explain what you mean? Are you proposing to change the primary author? (See >100 lilithcat:)

>100 lilithcat: >101 SandraArdnas: So I think I'd prefer if changing the primary author was a rare thing. It's very brutal. The general way LT works is for the correct bibliographic answer to emerge from a lot of data. Maybe the author's books are mostly without the A. Maybe they're with. I'd rather not FORCE an author into one or the other. It shortcuts the natural accumulation of the right answer in favor of what some member thinks is right and which sticks forever.

>102 I-_-I: What do you think of my argument right above?

>113 Collectorator:

I believe that when I say a work does not need to be aliased, I shouldn't have to spell it out any further to this audience. I'm including a link to the editions page, which is one of two places on this site that a user can get the actual facts of the matter. When every edition has a "1", they're all equal. That's why I keep saying "one equals one."
Also, it is clear from this discussion that the system has an unknowable algorithm guiding it. I question it more now than ever before.
Regardless, there is no reason to alias a work to a name that is equally represented in the choices. Just fix it so it goes there.

So here I think we get to the nub of it. As stated above, I think name-forcing is a last-ditch thing, not a regular way to organize things. These situations are mostly ones where there are very few editions. We really have no idea what will happen if more editions were added. Overall, more should lead us to better answers.

That's why I keep saying "one equals one."

Yes, but that's not an argument. It needs to be made clear or you're just pounding the table.

>70 SandraArdnas: SandraArdnas: I think I have made it clear that I am completely against combining dissimilar names with split authors.

I think we need to hash this out, because I don't think your opinion is universal.

>119 SandraArdnas: However, we can still discuss the principle and come to some community policy. Do we edit the calculated author and in what circumstances? Since in this case the calculated name resulted in unsplitted Rufus M. Jones, I'll pose the reverse question. Would it actually be better if it was assigned to Rufus Jones split, because that way all three of the members see it on their author page, rather than just the two who have Rufus M.Jones and Rufus Matthew Jones?

Do we edit the calculated author and in what circumstances? — My view is that we do it when the name is simply wrong, or when there's no other good way to get the right answer. We do not do it merely to get works to go to the right split/non-split.

Would it actually be better if it was assigned to Rufus Jones split, because that way all three of the members see it on their author page, rather than just the two who have Rufus M.Jones and Rufus Matthew Jones? — This is where we need to distinguish between theory and features. By "see it on their author page" you mean what, exactly? Which page? If it's getting lost, then we need to fix the feature, not muck with the data so the feature works.

### Questions

>110 Collectorator: Does not need to be aliased:
http://www.librarything.com/work/6818410/editions

I don't see what you're talking about. Please be clearer.

>126 Collectorator:

I'm scratching my head to understand what was going on here. AFAIK, the author is unitary. Was something else going on here? shortride, collectorator, krazy4katz, realsandy, casaloma, lilithcat are all involved.

Whatever is going on, I don't like the fighting. Many members seem to have thought the author should be split. Anyone could see there was a concerted effort to work on it. collectorator should not come in and just reverse it all without some sort of discussion.

>128 Collectorator: >129 lilithcat: So are collectorator and lilithcat in agreement here?

>130 Collectorator: Somebody is going to have to make rules about who gets added to split authors, and what happens when works have equal choices for the author name.
It's not right that we're spending so much time disagreeing on these things, when a simple decision made by Tim and announced to all would fix it right away.

1. I can see arguments for both methods people use--combining and splitting, and aliasing author variants around. I am NOT in favor of using names other than the main, common name authors are known by. That is, I do no agree that Joe Schmoe who is called Joe Schmoe most of the time should be shunted off to Joe R. Schmoe in order to put things in order.

If someone has a principle to decide when to combine and when not, I'm all ears.

2. what happens when works have equal choices for the author name What is your question? I do not favor forcing author names unless the current result is pig-wrong.

3. I do not agree that this is just about policy. It is also about being sensible, proportionate and willing to discuss things.

>131 Collectorator: http://www.librarything.com/work/8808856/editions
Does not need to be aliased.
Needs author change to move it from The Shortest to the desired.
No user will be shorted by this action because the two choices are equal.
The system used to select The Reasonable, but now it selects The Shortest.

I don't see that any system can make the right choice here. I wonder if the answer isn't that when you recalculate the title, and the answers are tied, it gives you the choice. On the back end this would not be a force but a preference--that E. Willems is worse than Emilio Willems here. As it would be a choice, ntoa force, it would only apply if they were tied. Would this be sufficient?

>93 Collectorator: Indeed I would. The author is clearly sometimes known without her middle name. I don't agree with the policy of, basically, never combining unless the names are dead-identical in all cases.

>133 Collectorator: James Riordan (disambiguation)
"James Riordan" is composed of at least 2 distinct authors, divided by their works.

Includes the names: Jim Riordan, James Riordan, Professor Jim Riordan, Professor Jim Riordan

IS THIS RIGHT TO DO OR NOT?

I say absolutely yes. What is the counter-argument?

>134 Collectorator: http://www.librarything.com/author/willemsemiliomussoli&norefer=2
IS THIS RIGHT TO DO?

Yes.

>136 aspirit:

>134 Collectorator: Collectorator: no, "Gioconda. Emilio & Mussolini Willems" shouldn't be combined with "Emilio Willems".

Why not, exactly? Emilio & Mussolini Willems is evidently ratty data for Emilio Willems. Why harm is there is recognizing that?

### Shortest

>108 Collectorator: Ah, okay. So the calculation of a work's author is different from the calculation of the best form of an author overall. But in any case, I don't think there's a shortest rule here. There's just only two options and its picking one.

>123 Collectorator: Not shortest. Most common. James Ackerman is two to one against James S. Ackerman.

As a final note, I'd be glad if we ended up with some rules and (more) rules of thumb. And some feature changes.

But the end result cannot be members going off and reversing edits pell mell and without explanation or with only a rude reply. Blocking combiners members who just want to understand or to talk cannot be allowed. And if any sort of "stalking" is taking place, it needs to end.

Okay, one last "final" note. We are not going to come up with a set of iron rules for every circumstance. Combination, aliasing and name-changing are all tools. Even if one is preferred, if someone managed to make the situation work well in one way, I see no compelling reason for another member to come in and change everything so that it's done another way. That is, big, destructive edits of other members' work need to have a rationale other than "the rules." They need to be have some concrete problem in mind, and respect others' work.

>136 aspirit: "Gioconda. Emilio & Mussolini Willems" shouldn't be combined with "Emilio Willems".

>141 timspalding: Why not, exactly? Emilio & Mussolini Willems is evidently ratty data for Emilio Willems. Why harm is there is recognizing that?

It's "Gioconda. Emilio & Mussolini Willems". It was originally "Willems, Emilio & Mussolini, Gioconda." from Buzios Island: A Caicara Community in Southern Brazil, and it's clear it's supposed to mean two people.

>144 prosfilaes:

So what should be done with "Gioconda. Emilio & Mussolini Willems"?

and it's clear it's supposed to mean two people

Right. And that's impossible in the system. So we need to do something with it. Choose the first author as the real author and move on.

>145 timspalding: Why do we need to do something with it? Why not leave it uncombined? It's only part of a zero-copy edition, so absolutely nothing bad is going to happen with it.

And that's impossible in the system.

You made the system. If something has to be done with it, make it so it can be marked as an alias that means two people.

>133 Collectorator:: Personally I don't see titles as part of names. So I think the addition of "professor" should be ignored. (I could be convinced to do things an other way though).

I would be reluctant to combine "Jim" and "James" though.

>145 timspalding: The basic consensus seems to be to leave these things alone. The exception is if there is more than one such record featuring both authors, in which case many consider it good practice to combine those records. Any works found in the course of rolling up the author records are then (ideally, time permitting) Other Authored to the appropriate single authors, and/or combined with the appropriate work records found under said single authors. Last, the truly conscientious among us will finish with helpful disambiguation notices instructing people to Other Author any work that subsequently lands on the multiple author record. (Personally, I'm usually moving too fast for that).

>145 timspalding:

Choose the first author as the real author and move on.

But one of the guidelines for author combining is: An individual with a collaboration ("Neil Gaiman" should NOT be combined with "Neil Gaiman & Terry Pratchett")

https://wiki.librarything.com/index.php/HelpThing:Author

>141 timspalding: Do we edit the calculated author and in what circumstances? — My view is that we do it when the name is simply wrong, or when there's no other good way to get the right answer. We do not do it merely to get works to go to the right split/non-split.

Yes, that IS the general community policy, I was mainly making a point (in the entire post) that WHEN editing the author, rather than leaving the calculated one, there are going to be different arguments why chose one or the other. It is my understanding Collectorator favors simplifying/avoiding splits and aliases and would often edit the winning primary author to achieve that end. This isn't a common practice. In fact, we commonly leave the author as is and do the splits an aliases based on that.

>141 timspalding: Would it actually be better if it was assigned to Rufus Jones split, because that way all three of the members see it on their author page, rather than just the two who have Rufus M.Jones and Rufus Matthew Jones? — This is where we need to distinguish between theory and features. By "see it on their author page" you mean what, exactly? Which page? If it's getting lost, then we need to fix the feature, not muck with the data so the feature works.

In case of split authors, the aliased page always has all the works of that author because that is how the feature works, but those who have entered the shorter version of the name (the split) go to that page when they click on their author and there see only the works where that name is the primary author.

Ideally, the works from the alias should go to the split in the same way those assigned to a split go to the alias. That way, everyone would see all of them, regardless whether their author page is the split or the alias. This would end many arguments over what should or shouldn't be the primary author because the work would end up on both the split and the alias anyway and regardless of the version of the name used all members would see all works on their designated author page.

Even more ideally, splits would be independent author pages that we could combine with what is now the alias, thus having a single author page for every author, but that might be a unicorn

>147 Nicole_VanK:

I don't see titles as part of names.

Agreed.

Professor, Ph.D., Bishop, none of those are parts of names.

I would be reluctant to combine "Jim" and "James" though.

Me, too. If they are different people, even if "James" uses "Jim" occasionally, if "Jim" never uses "James", it would be weird to see him on "James"' page.

I don't think we can draw too hard a little re titles. Edward Plunkett's author page ought surely be "Lord Dunsany", frex. I'm not aware of anyone who actually published as Professor something-or-other, but it wouldn't surprise me if there's someone.

>153 AndreasJ: Plenty of non-fiction comes with MD, PhD and such on cover and title page. Publishers like to make author's credentials prominent. Libraries do not normally include it in author name. Not sure about amazon, which most likely doesn't have a clear policy one way or the other. However, a fair amount of records on LT include such titles in author name, whether because imported from amazon or because people prefer entering what's on title page I don't know.

>153 AndreasJ: Hard rule?: no. Lord Dunsany, Lord Byron, Dalai Lama etc. - commonly known under those names, and it would be overkill to "correct" that too.

But yes, some publications do list authors with their academic titles, or military ranks, or monastic order, or whatever, while some don't. And sometimes the titles change during their lifetimes. For example, my copy of The Recovery of Jerusalem calls Charles William Wilson a Captain on the title page. He later also published as "Colonel" and as "Sir" without ever changing his name.

>147 Nicole_VanK:
>151 lilithcat:
>153 AndreasJ:
>154 SandraArdnas:
>155 Nicole_VanK:

Regarding titles in names, imho they can be used for disambiguation purposes & aliasing. So if we have two James MacDonalds, but one was a baronet, you can combine all the Sir James Mc, James Mc, bt., etc., then split the simple James MacDonald and alias works pertaining to the Sir James to the particular name.

>142 timspalding:

I'd be glad if we ended up with some rules and (more) rules of thumb. And some feature changes.

One feature change that might be helpful is disambiguation warnings showing up when combining authors with disambiguation notices. If I understand correctly, that's one main concern about using canonical author to deal with the Gary D. Chapman situations - the lack of a disambiguation warning will result in the authors being combined eventually by a well-meaning editor who doesn't see what's been done.

>156 gabriel:

Yes, that was the point.

>146 prosfilaes: Why do we need to do something with it? Why not leave it uncombined? It's only part of a zero-copy edition, so absolutely nothing bad is going to happen with it.

Your second point is true enough. But this happens with multiple-copy works.

You made the system. If something has to be done with it, make it so it can be marked as an alias that means two people.

I just think this way lies madness.

>148 davidgn:

Would some sort of "this is combined, but the data is crap" flag be helpful?

>149 lilithcat:

But one of the guidelines for author combining is: An individual with a collaboration ("Neil Gaiman" should NOT be combined with "Neil Gaiman & Terry Pratchett")

Well, fundamentally, what's going on here is very crappy data. "Willems, Emilio & Mussolini, Gioconda" isn't a collaboration, it's a mess.

I think this raises the issue of a "ratty data" flag, and the capacity to do as banks do when they can't read a check--"Read as…" But I fear if I create a "read as" capacity it will be massively abused to fix all sorts of problems that aren't this kind of thing.

Lastly, the reason why we don't combine "Neil Gaiman & Terry Pratchett" into either is that it would end up getting combined into both and madness would ensue. I don't see that happening with "Willems, Emilio & Mussolini, Gioconda."

>150 SandraArdnas:

timspalding: My view is that we do it when the name is simply wrong, or when there's no other good way to get the right answer. We do not do it merely to get works to go to the right split/non-split.

SandraArdnas: Yes, that IS the general community policy

Okay. So here, I think, is a point of real difference with Collectorator. Right?

In case of split authors, the aliased page always has all the works of that author because that is how the feature works, but those who have entered the shorter version of the name (the split) go to that page when they click on their author and there see only the works where that name is the primary author.

Okay, so would members want it to "go all the way"? That is, if Robert Jones goes to Robert Jones (3) and Robert Jones (3) is aliased into Robert Q. Jones, the link should go to Robert Q. Jones?

I don't see titles as part of names.

Agreed.

Professor, Ph.D., Bishop, none of those are parts of names.

So you don't think Prof. John Smith should be combined into John Smith, or are you merely saying it shouldn't "win." If the latter, I agree. Except for rare cases, like Dr. Dre.

However, a fair amount of records on LT include such titles in author name, whether because imported from amazon or because people prefer entering what's on title page I don't know.

I bet its mostly Amazon.

Lastly, the reason why we don't combine "Neil Gaiman & Terry Pratchett" into either is that it would end up getting combined into both and madness would ensue. I don't see that happening with "Willems, Emilio & Mussolini, Gioconda."

But it could. Both Emilio Willems and Gioconda Mussolini have separate pages. What's the standard, then, for deciding when a collaboration and a single author get combined?

>159 timspalding: Re the crap data flag idea, it might be helpful in one sense: if there was thereby generated a publicly accessible list/queue of such flagged combined "authors" containing works needing Other Authoring (ETA and/or combining) that the community as a whole might work on.

Related: been meaning to bring this up for a while, but another good idea would be a publicly visible list/queue of authors with combination issues. I will freely admit that my author combining tends to produce those, and I rarely take the time myself to go into the CK fields and resolve them, since that's usually out of scope for my combining runs. I tend to leave that as an exercise for other users.

(Similarly, I often do not combine works when combining authors. My theory is that making the work records stranded on uncombined author records visible by rolling them all up properly into one author record tends to stimulate work combining by others. Also, it sets them up for combining in the next semi-automated work combining queue, if you ever set that up again).

Guess we're into RFI territory, so I'll write this up there when I get a chance.

>161 davidgn: "Also, it sets them up for combining in the next semi-automated work combining queue, if you ever set that up again"

I've noticed that, years later, the "Combination Opportunity" page link that was used for that operation in early 2018 still appears on the confirmation page after some combinations/separations, despite never having anything available on that page anymore. ("It appears that's the last one!")

>162 Opteryx:

I'll look into it.

>various

Shouldn't "Willems, Emilio & Mussolini, Gioconda", which after all are just a couple of collaborators be struck out and replaced with Willems, Emilio as a Primary author and then Mussolini, Gioconda as a Main author? I thought that was the standard way to list collaborators.

>164 Crypto-Willobie:

Shouldn't "Willems, Emilio & Mussolini, Gioconda", which after all are just a couple of collaborators be struck out and replaced with Willems, Emilio as a Primary author and then Mussolini, Gioconda as a Main author? I thought that was the standard way to list collaborators.

Ah, I thought the Gioconda was part of the title creeping in. My apologies.

In this case, though, there is no work. It's just a free-floating 0-copy name.

>165 timspalding:

I thought the Gioconda was part of the title creeping in.

See >160 lilithcat: and Gioconda Mussolini

>166 lilithcat:

Yes, I know I was wrong :)

>167 timspalding:

Even Homer nods.

>159 timspalding: Okay. So here, I think, is a point of real difference with Collectorator. Right?
That is my best deduction from the cases I've seen, but I'm not even sure my deduction her aim is to avoid splits is correct, so I'd say the the root of the problem is lack of communication. Even when she has better arguments than whoever is disputing her edits, we're all supposed to deduce them on our own.This isn't really working even with those of us familiar with ins and outs of combining conundrums, let alone others

>159 timspalding: Okay, so would members want it to "go all the way"? That is, if Robert Jones goes to Robert Jones (3) and Robert Jones (3) is aliased into Robert Q. Jones, the link should go to Robert Q. Jones?

Yes, if all works assigned to Robert Q Jones were also seen on Robert Jones (3) (just as those assigned to Robert Jones 3 are seen on Robert Q Jones currently), everyone would see all of them on their author page. This would, among other things, make it irrelevant which of the two is calculated primary author, so no reason for anyone to edit it.

>169 SandraArdnas: Yes, if all works assigned to Robert Q Jones were also seen on Robert Jones (3)

How would that work in a case like this:
There are multiple authors named Al Smith
There are multiple authors named Bob Jones
Al Smith (2) is a pseudonym of Bob Jones (3)

What we'd want to do is link Al Smith (2) and Bob Jones (3) so all his works show up on both pages.

>170 r.orrison: I'm not sure, I'm just suggesting that when something is aliased, the flow of works assigned to one or the other should, if possible go both ways.

Is it even possible to alias two splits as in the example? I thought they can only be tied via a third unsplit author, if there is such

>171 SandraArdnas:
I agree that it would be great for it to be able to work both ways, but I think the problem that I highlighted would need to be fixed first.

Thinking through it more...

If you alias Al Smith (2) to Bob Jones, then all of Al Smith's works would appear on Bob Jones's page and would need to be manually assigned to Bob Jones (3). We can do this now.

If the reverse link was implemented, so that Bob Jones's books appear on the Al Smith (3) page, it would include all of the works by all of the authors named Bob Jones (since the alias can only be to jonesbob), not just the works of Bob Jones (3).

I'm pretty sure you're right that you can't alias a split to a specific other split. That would need to be fixed before the suggestion in >159 timspalding: could work.

>172 r.orrison: Ah, I see what you mean. But, yes, in the current system we can only combine them, which is extremely messy because it can only combine all Bob Jones splits with all Al Smith splits.

Collectorator is gone... zero books cataloged.

Has this exercise worked? Everyone has stopped posting... is everyone bored or has a resolution been made? If resolved, shouldn't it be posted here? If not, shouldn't a status update be posted?

I've monitored this string for various reasons, and I imagine there are many other LTers who've done the same. A lack of knowing what is going on or going to happen to a fellow LTer. Well, it doesn't feel good.

>174 Conkie: I'll ask Tim for an update. I think things have gone a bit stagnant here with the holiday craziness of SantaThing et al. but I'll ask. Thanks for your concern.

>174 Conkie:

I'm sorry that Collectorator left, but she left. I did not delete her. Clearly the attempt to get her to play well with other members not, as many urged, simply remove her from the site, did not work.

It's possible if I had been faster to respond to her demands, and granted more of them, she would have stuck around. But she was responsible for how she edited on the site, and was extremely unwilling to listen to me or others, and was unwilling to communicate her thinking to others. She probably made the right choice.

>177 Collectorator: I did not read that as 'bad edits', but as 'controversial edits', meaning leading to disputes with other helpers, which never seem to be resolved through short discussion. You seem blissfully unaware how entitled it is to expect the site owner to step in to resolve this.

In reality, yours are not the only controversial edits around, they just stick out like a sore thumb because of the way you choose to handle them - by merely insisting on re-editing your way, without explaining why that is preferable. It is my impression you think there are hard and fast rules for everything, but quite obviously cases arise quite often where things are not that black and white and members disagree about the best way to handle a particular case, while sticking to established community policies.

>177 Collectorator:

Glad to hear of it. Can you please engage with us? My last substantive message to you was >141 timspalding:. It was lengthy, and contained all manner of discussion as well as direct questions for you. You have said nothing in reply, and nothing at all to others. Others too have noted you're not responding to people generally.

I'm not really a combiner but I'm trying to follow this thread and Collectorator keeps mentioning aliasing works. But as far as I know aliasing only exists for authors. Can someone explain what aliasing means for works?

>181 jjwilson61: Nothing for works in particular, as far as any of us can infer, it refers to aliased authors

>181 jjwilson61:

Collectorator keeps mentioning aliasing works. But as far as I know aliasing only exists for authors.

Collectorator is correct. It is the works that are aliased to an author page.

So if John Jones is split, and some works assigned to John Jones (17) are by John Q Jones, the works get aliased to John Q Jones, but also continue to show under John Jones (17).

That's why you'll see, on a split author page, the notice Works have been aliased into . . .

>176 timspalding: "if I had been faster to respond to her demands, and granted more of them, she would have stuck around."

Do you know Kipling's "Dane-geld"?

>183 lilithcat: Yes, but it sounds confusing if it is implied that individual works are aliased. Either all of them on the author page are aliased or none.

>183 lilithcat: But to say that you're aliasing a work by John Jones (17) to John Q Jones implies that you might alias another work by JJ (17) to a different author which, as I understand it, isn't possible.

ETA: It just makes no sense to alias a work to an author. To alias something means to change it's reference to another one of the same kind of thing.

>184 DuncanHill: I didn’t know that one but very apposite.

Ten days silence after Tim’s very long and comprehensive post then a response in eleven minutes after his next comment. What does that tell us?

>187 abbottthomas: It tells us that LT is in SantaThing mode. They rarely get anything else done in December.

>188 jjwilson61: I refer to Collectorator's long silence and lightning response - nothing to do with LT staff activity.

>177 Collectorator: "That's the second time you have alluded to my 'making bad edits' yet you have not once posted even one example of it. It's a story, a fabrication, a delusion on your part. I know you wish it was true, but it is not true, Tim Spalding."

Collectorator is a perfect editor with a 100% accuracy rate. Any insinuation that Collectorator has made a single mistake, ever, is a lie. We're all implicated in collective delusion or clever conspiracy. Maybe both!

Nearly 200 messages into this thread, the member has yet to acknowledge any issues with her edits or - more importantly - her handling of disagreements and her treatment of other editors.
She now repeats the refrain that no one has yet provided an example of a Collectorator error - an amazing contention given that LT now has at least 5 talk threads devoted to exactly that. This suggests bad faith, either in the Sartrean sense of self-deception, or the more traditional sense of intent to deceive others.

>191 Collectorator:
There is no reason to change it. The current, system calculated, author is not wrong. The work is linked to Richard Bosworth (3) which is aliased to Richard J. B. Bosworth. There is nothing more that needs to be done.

Surprising how a short while after I sort out the various forms of R J B Bosworth's name and combine a few works as necessary, Collectorator decides that she needs to change the name on one of his books for no reason.

https://www.librarything.com/author_split.php?author=bosworthrichard&page=as...

https://www.librarything.com/author_split.php?author=bosworthrjb&page=assign

>193 DuncanHill: To be fair, a lot of us watch the helpers log and find things there that need (re-)fixing. It's just that most of us go about it more collegially.

(ETA: i'm not implying that you made a mistake here.)

>191 Collectorator:

I have always taken the warning on the edit other authors pop-up to be the guiding policy in this area:

"Warning: Changing a work's primary author on the "work" (ie., global) level is serious business, and should only be done when truly necessary."

In my view, necessary is when the calculated author is incorrect or it creates some absurdity. That is not the case here. A calculated author does not have to be changed merely because it directs to a split author.

I see that Collectorator made some edits yesterday, including changing the author name on a work i had aliased. It's exactly the kind of questionable edit that has led to conflict in the past. More to the point, i thought Collectorator was supposed to propose edits here *before* making them? >1 timspalding:

Abyssal environment and ecology of the world oceans

Another edit war and a classic personalized disambiguation notice: Donna Cross

>197 I-_-I: And the biggest name on Pope Joan: A Novel is Donna Cross, not Donna Woolfolk Cross.

I have just read this whole thread and the author combinations thread and become even more confused about splits and aliases than ever, but you all have washed up (for the second time in this thread...) on a book I own, and that has clarified things.

So for anyone else reading along who hasn't been neck deep in aliasing: Yes, it is extremely confusing that when I currently click on the author name Donna Cross on my copy of Pope Joan (on the works page or in my catalog), I am taken to a page on which I do not see my book and don't see myself listed as having any books by the author.

Meanwhile if I go to Donna Woolfolk Cross, I see myself listed as owning one book by the author, but if I click on the author link on that page, I am taken to an empty catalog page that says I have no books by the author. (But if I click on the work link on the Donna Woolfolk Cross page, I am taken back to my work page from which I can't get back to the Donna Woolfolk Cross page, only the Donna Cross page. Which say I have no works by that author.)

This is probably something that should be fixed on timspalding 's end at some point, because that's just. really bad, from a usability standpoint, something like that should not happen regardless of the status of the combining. (It has been extremely annoying even in the process of writing this comment, because if I click from the author page to the work page in an attempt to understand what's going on, I can't then easily get *back* to the author page. Or vice versa.)

But in the meantime, the current combining "fix" has not made anything better.

I think, from reading upthread, that if Donna Cross was split and then Donna Cross (2) was aliased to Donna Woolfolk Cross, this problem would go away, and I could click from my catalog to the correct author page in at most two clicks? Or would one also have to do something to get the DWC works onto the Donna Cross author page in order to do the split + alias? Or would there just be an empty split on the Donna Cross page with a link to the alias (Which would still be an improvement on the current situation.)

(I know, I should just edit the author name on my copy, but I want to understand what's going on first, and that wouldn't help the 900+ people who also have the Donna Cross edition.)

(...and now I have lost my touchstones by editing and forgotten how to get them back.)

>199 melannen: Except in this case there is no split and it is unclear why the two shouldn't be combined. Perhaps Collectorator will grace us with an explanation for a change.

>200 SandraArdnas:

If, as I understand it to be the case, "Donna Cross" and "Donna Woolfolk Cross" are two different people, then they should not be combined. I think neither of the books currently on the "DC" page are by "DWC", but are edited by "a professor at the School of Exercise, Biomedical and Health Science, and the director of the Child Health Promotion Research Centre at Edith Cowan University, Western Australia.".

In the past Collectorator has seemed to prefer distinguishing authors with the same name by adding a middle name or initial(s) to one or more and aliasing him/her/them, even if this form has never been used by the author or authors on any publications. I am not sure why Donna Woolfolk Cross, who has apparently used this name, or Donna W. Cross, on all her books, was an exception. As you can see, my attempts to change Donna Cross (1) to Donna Woolfolk Cross earned me a rebuke, by name, from Collectorator. Payback for the egregious “François Voltaire” ?

>201 lilithcat: I did just a quick search to see what's it about and to me they seemed the same person. Either way, 'do not combine' an a personal rebuke is not a disambiguation notice. It gives no information whatsoever, but merely attempts to assert the indisputable authority.

booksaplenty1949 and Collectorator have reportedly deleted CK data that distinguishes the authors from each other. That looks inappropriate on both sides and is making the situation harder to sort out. (...I don't know if it's the system making the deletions as the authors are aliased then separated.) Update: Fields that were empty several minutes ago are now showing as having been refilled seven hours ago! Strange....

The data that's been deleted shows thatAnyway, the Australian Donna Cross currently linked in the touchstone above is a different person than the American Donna Woolfolk Cross.

https://www.librarything.com/author/crossdonnawoolfolk

That's also confirmed by info in the external links that remain on the author pages. What might be contributing to confusion is that the American Cross is pictured on the Australian Cross's page. The photos need to be moved.

>203 SandraArdnas:

'do not combine' an a personal rebuke is not a disambiguation notice.

That was just rude.

>204 aspirit:

What might be contributing to confusion is that the American Cross is pictured on the Australian Cross's page.

I think that at some point they were combined (perhaps some of DWC's works were assigned to DC?), and that's when the photos landed there.

If you look at “Flagged pictures” you will see halfway down a picture of Donna Woolfolk Cross (in a brown dress) identified as Donna Cross (1) or maybe (2) and a series of messages about whether the author is or isn’t split. Votes to remove or keep have become meaningless as the author pages keep morphing. Recently the Donna Woolfolk Cross page had two copies of it and it was flagged again. I added and deleted CK and links as the pages came and went, all along thinking that the answer—-having a page for Donna Cross the Australian professor and a page for Donna Woolfolk Cross the American author—-was the more than obvious solution. But She Who Must Be Obeyed thinks otherwise, apparently.

Yes, I am 99% sure that the medical author and the Pope Joan author are different people, and the medical author is not DWC.

So the argument is over whether to combine DWC with DC, and then split the DWC works from the DC works, or split DC and alias the DWC works to DC, right? We all agree the current setup is bad? And either of those solutions will result in be being able to go from my works page to an accurate author page and back?

The effective difference, usability-wise, is that if you combine DWC and DC, there is no longer a separate author page for DWC at all. And there is a chance that DC's works will end up getting labelled as by DWC. Does this have other downsides? It doesn't seem to affect things like search. (DWC is the daughter of Joanna Woolfolk and William Woolfolk who wrote on similar topics, so the "Woolfolk" matters, which is presumably why she always used it - I want to be able to bring her and her mother up on the same search, etc, but that doesn't seem to matter in terms of combining...)

What is the downside if you split and alias without combining - does that mean that all of DWC's works won't appear on DC's author page, only the ones with DC editions? I'm still confused by the comments upthread about works not appearing on the right author pages if you do the split/combine/alias wrong.

It seems to me like this is an obvious case for split and alias don't combine, since we have an author who always goes by DWC and already has an author page for that, but ratty data has linked some editions of her works to a different author page. But I am probably still missing some reasons?

(Sorry if I'm being pedantic but it's so nice seeing people at least trying to explain this stuff. This is why I basically haven't touched authors since splits arrived, so confusing and no clear standards for a lot of it. We really need authors to go to unique database IDs not truncated strings, but I do get that it's late in the game to easily shift the whole database...)

>210 melannen:

So the argument is over whether to combine DWC with DC, and then split the DWC works from the DC works, or split DC and alias the DWC works to DC, right?

No. Neither of those is appropriate. We do not combine different authors with different names.

We all agree the current setup is bad?

The current "setup" is the way it should be. Two different authors with different names, each with her own author page. What on earth is wrong with that?

Lilithcat expresses my sentiments exactly. If we could post additional picture of the more mature DWC, which someone took the trouble to locate and upload, on her page, all would be even better. There is a thumbnail of Donna Cross the Australian academic available on the two “Academic” links, if anyone knows how to add it to her page.

What's wrong with that is what I outlined above - right now, if I click on the author name (DC) on her work in my catalog, it takes me to an author page (DC) that says I have no works by that author. Meanwhile, if I go to the DWC author page, it says I have one of her works, but when I click through to my catalog (DWC), it says I don't have any works by this author.

If I go to my Pope Joan work page, and click through to the author page from there, not only does it tell me I have no works by that author, I also have no obvious path to get from that author page back to the work page I came from.

This is wrong and bad.

Maybe not on an abstract data level, but on an "I would like to catalog my books" level, it is wrong and bad.

The wrongness and badness is not as apparent if you don't own one of her books, but it is very, very apparent (on the 'I tried to use LT and it was broken and made no sense so I quit' level, or on an "I wanted to fix my catalog so I made the obvious combination" level) for an individual user who owns the book.

And it would be one thing if I had cataloged the author as "adkc mickey mouse" or something, but I own a 900-copy edition.

The wrong and badness is mostly to do with things that are messed up with the way LT-as-a-database handles authors vs editions vs works, and site-level vs. catalog-level data, but it is still wrong and bad, and fixing things like that on the level of using their catalog is the primary goal of most LT users, so if our combining choices make them worse, that is also bad.

I don't actually know what is the best way to use the combine/split/alias system to fix this problem, but I would like to learn.

ETA: I suspect this is why Tim came in with an 'always combine' assumption - because database-wise, that is probably the most effective way to correct for the ways the database isn't well-designed for authors, and he is thinking in terms of the database.

Far be it from me to tell anyone how to catalogue his or her books, but my own preference is to use the name which the author, editor, etc chose to put on the jacket/title page, which in the case of Pope Joan is “Donna Woolfolk Cross.”

>215 booksaplenty1949: I did say in my first post that the obvious answer is to fix my author, which I would have done if I wasn't using it as an example in this thread, but there are 1000+ other copies of this book under that author name, the author name isn't obviously wrong, we can't fix all of them at the catalog level, and most of the people who own them aren't reading Talk threads in the combiners group and understand the database even less than I do.

The goal of doing Helpers tasks should be to make the site more usable for all users - those who are navigating in from elsewhere and just want to learn about the authors (for whom the current setup is probably good); those who use the site at a high level (who have probably fixed the authors in their catalog and are arguing on combiners!) *and* those who just want to be able to add things from the app and then use their catalog to organize and find them (probably the vast majority of people who are actually going to those author pages.)

ETA: Also, IIRC I had actually looked at this book earlier in the year, when I was recataloging some of her mom's stuff, and debated editing the author name, but the state of the author combining at the time actually meant it made more sense to leave it as it was. So, another vote down for edit wars on this stuff.... and the fact that LT shows "your" author name on all work pages makes it really hard to understand that you have the "wrong" author name if you aren't super experienced with LT.

>216 melannen: I recently acquired a print-on-demand copy of Tracks in the Snow. The cover gives the author as Lord Charnwood. I entered it as Lord Charnwood Godfrey Rathbone Benson in my catalogue because there could be many Lord Charnwoods, but the “Main page” shows the author as just Lord Charnwood. When one clicks on that name on said “Main page,” however, “Godfrey Benson” pops up, with the note that this author includes eleven other names. Is this a bug or a feature? If you look on the DWC “Combine works” page you will see several copies of Pope Joan entered under “Donna Woolfork Cross” Presumably if their owners are frustrated in their attempts to follow up on this author they will click on the book title and get led back to DWC. Those who chose “Donna Cross” have the same option.

Huh, so apparently the reason I've been so confused by author combining is that a lot of you have an entirely different aim in doing it than I had been assuming was the purpose.

>218 booksaplenty1949: If I click on Lord Charnwood \Godfrey Rathbone Benson\ in your catalog, it takes me to an author page that shows you as owning a copy of Tracks in the Snow by that person. If I click on your works by that author from the Godfrey Benson author page, it takes me to a page in your catalog that shows your copy of Tracks in the Snow.

As LT does not, at any point, flatly contradict itself as to whether you own the book or not, it is indeed wildly preferable to the situation with my book.

ETA: Donna Woolfork Cross comes up as an automated combination suggestion for Donna Woolfolk Cross. As it's also an obvious typo and there is no other author it could be confused with, before I read this thread I would have combined without a second thought. As that kind of thing is exactly what author combining was originally meant for. I'm going through the rattier data end of my catalog right now and looking for other examples and can't find any because people have already combined most of the ratty authors. Which is great.

>217 Collectorator: That's not really true. The majority of authors with multiple forms of name are combined because the shorter form is also distinct, not split. From among the rest, the vast majority is split-aliased, which again at the very least provides a link to the page gathering all author's works. This case where the shorter form is not split, even though it is used for two different authors is pretty rare.

Either way, disambiguation notice should have pertinent info. Do not combine is not an explanation for people who want their record to show on the correct page

>221 Collectorator: I love that condescending tone of yours. How appropriate given that you can't follow an argument by your own admission. Yet, you expect others to decipher your posts like >209 Collectorator: . We do not want to decipher. Form a coherent argument about what should be done and why.

>223 Collectorator: Can you explain some more about both works and authors being aliased? I am really truly confused about that. Is >185 SandraArdnas: disturbing because it's incorrect or because it implies other people don't know it is correct?

I've never actually done an alias due to not understanding them, but it looks to me like what gets aliased is an author division, which is neither an author nor works, really, it is defined set of works that have been associated with an author. Is there some other way of aliasing?

>223 Collectorator: We can alias authors, we can "correct" author names on works. The two are not the same.

>223 Collectorator: If you keep re-editing someone else's edit, than you most definitely 'have an argument' with that person. You just refuse to discuss it.

lilithcat isn't the one playing the bully in the playground, so quit making it about someone else. Did lilithcat add the reprimand with no explanation whatsoever or did you? If practically everyone who looked at that case is confused, quite obviously a more informative DN is needed, but you're so hung on who's right and who's wrong that proceeding in a way that's actually helpful never figures into it.

>223 Collectorator: I'll repeat that it really is important to understand that both authors and works can be aliased. This post is disturbing: http://www.librarything.com/topic/326689#7344568

Really, enlighten me - how does that work. Help out a fellow helper. You can alias authors without aliasing works on that page or vice versa? It's news to me, but I'm eager to learn.

When I first looked at Donna Cross, apropos of the flagged picture back in October or November, I saw that Mediaspeak, Word Abuse, and Speaking of Words were wrongly included with the two books by the Australian professor. So I split the author. Then I discovered that those three books, identified on their covers as authored by Donna Woolfolk Cross or Donna W Cross, were by the same author as Pope Joan, but when I tried to combine Donna Cross (1) or (2) (I forget which was which) with Donna Woolfolk Cross I incurred the wrath of She Who Must Be Obeyed.

>228 booksaplenty1949:

I don't know about wrath, but you can't combine a sub-author on a split page into another author page. It doesn't work.

And combining the other way (i.e. DWC into DC) is wrong because the DC page is split.

Hence, aliasing.

Setting up an alias happens with/to the author. Aliasing an author affects the works associated for that split author (see Rudyard Kipling (1)).

Saying "we can alias works and authors," seems to suggests that aliasing can happen with/to a work separately from the author. AIUI, that is not possible.

ETA: It may not help that we use overlapping terms: we can combine/separate authors, and we can combine/separate works. We can split -- and alias -- only authors.

>230 karenb:

We can split -- and alias -- only authors.

Then why, when you look at a split author page where there's been aliasing, do you see "Works have been aliased into {author whoever}"?

That language suggests that you split authors and alias works.

>231 lilithcat: That text is misleading and should be changed. While it reflects the fact that technically it's the author for that work that's being aliased, the UI only permits you to alias all the works in that split together. So from the perspective of an actual user it only makes sense to talk about aliasing authors, or maybe just author splits.

The alias page says you are aliasing "divisions". If we're being precise and pedantic about terminology, it makes a lot of sense to me to say that you can't alias works *or* authors - only divisions.

(Whether they are primarily divisions of authors or they are primarily divisions of the works assigned to an author seems like sophistry, but I suspect the database backend treats them like divisions of works assigned to an author, which is why it refers to them as works elsewhere.)

>234 Collectorator: I If the way it is described differs from the way it is used then we are all just running around in circles.

>234 Collectorator: As I see it, aliasing is a way to say that a work that has been attributed to J. Smith is really by Jeremiah Jebediah Smith, which is just a renaming not adding a second author.

>237 Collectorator: Are you under the impression that an author's name and the author are the same? If I say that for a given work that J Smith actually refers to the author known elsewhere as John Smith, I'm not talking about two different people.

It's a way to have one record for an author even though they may have used different names for themselves on different works.

>237 Collectorator: AFAICT It's "A way to link part of one author page to a different page for the same author."

That is certainly the problem I was having on the work where it hasn't been done - there is no connection in the system between the two author pages for the same author, and that makes site navigation and understanding the author's works difficult. So you alias to create the link.

Having two pages for the same author isn't ideal, but the aliasing system doesn't (or at least shouldn't, if I'm understanding it correctly) create the duplicate author pages, it just attempts to ameliorate the problems it causes.

Collectorator has been suspended as a member for violating the terms of her warning.

If I had a way of preventing a member from making helper edits, but otherwise keep their account, I would do that. But I made the consequences clear, and at least Collectorator has no books.

It's clear to me that we need to on both tools and policy. I think she raised some important points, and, indeed, did much good work. But I cannot ignore the violations of the terms, and the stress she has been putting on other members and on staff.

>241 timspalding: Thank you for informing us of your decision. I am grateful you acknowledge that she did a lot of good work here on LT; it is very easy to let frustration overrule all other emotions. Indeed you, your staff, and many LT members (me, too!) have had run-ins with her, and unless we were in agreement with her, rarely did it end well.

I'm just sorry none of us could help her see there are better ways to communicate than what she chose. My hope is that she experiences some personal happiness in 2021... the same hope I have, for all LT members, staff, and indeed, humankind.

>241 timspalding: May I ask, is that a permanent suspension or temporary?

>243 bernsad:

I'd be happy to make it temporary, but it would require the sort of willingness to work with me and others that has not been forthcoming, and, probably it would require new tools for members to police, dispute and lock down edits--tools that I can't spend the next two+ weeks making at present. My sense is she won't ask to come back.

Just speaking in the abstract, no longer having any books on the site but continuing to edit, combine, alias etc seems a bit like remaining in the PTA after your children have graduated. I would be a bit concerned about someone like that.

>244 timspalding: That's unfortunate that it had to end this way. Thanks for your patience and input with this.

>241 timspalding: I reckon you have handled this difficult problem with patience and tolerance verging on saintliness. You must be a very good father! It all goes towards making LT the place so many of us appreciate so much.

>247 abbottthomas: I'll second that!

>247 abbottthomas: Hear, hear!

>241 timspalding:

I'm very sorry it came to this. Collectorator, as you say, did contribute a lot to the site. I wish she could have seen her way to explaining why she did some of the things she did.

Thank you for your efforts and your grace in handling this difficult situation.

>250 lilithcat: “Seen her way to explaining” why she lurked the Helpers’ Log and left insulting messages for other members? That seems unlikely. People don’t usually have that level of self-knowledge, and in any event this isn’t the forum. I do recall that at one point the “About Me” line on her Profile page read “I am perfectly odd.” Suggestive.

>251 booksaplenty1949:

I was referring to her combining/separating choices.

>252 lilithcat: I knew that. But that wan’t the real problem. Cf the talk thread “An unpleasant run-in...”

A “small” sample from that thread: “A very vague compliment followed up by a bold, false, pointless, proofless accusation ‘you sometimes have a tendency to over-correct things’. Just exactly what Tim Spalding posted recently. I told him he was sad and common. I neglected to say small. Dang. I regret that. Small, small, small.”

>254 booksaplenty1949: why don’t we consign this nastiness to history. It is really sad that someone feels this way, how can she have any happiness? >244 timspalding: doesn’t think she will come back. I wonder. How will she fill her time?

It is very sad. At one time she did have a catalog and participated in several groups. We exchanged books and cards even. But the editing obsession and temperament changes swept those aspects away. I’m breathing a sigh of relief for life on LT; but I do still worry about her personally.

>256 2wonderY: My sentiments exactly.

?? with this Case

CORNEILLE
there are 3 authors but 4 entries : A and A bis is the same with some pictures on each....

Common name used for Pierre Corneille is CORNEILLE so A bis > A

So how could we do for this (thanks to keep CK and http and pictures... it's a lot of time, searches and works)

xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx

A
https://www.librarything.fr/author/corneillepierre
Pierre Corneille (1606–1684)
225 works
-------------------------------
A bis
https://www.librarything.com/author/corneille-1
Corneille 1 (is pierre corneille)
0 works................. !!!!!!!!!!!!

xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx

B
https://www.librarything.com/author/corneille-2
Corneille (is Beverloo, Corneille Guillaume)
2 works

xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx

C
https://www.librarything.com/author/corneille-3
Corneille (is Nyungura, Cornelius)
1 work

>258 JMK2020: I put 3 stranded in "unknown" in "A bis"

Obviously this does not fix the underlying problem though

>258 JMK2020:

Please let this thread die.

Thank you.

Right. The correct thread is this one:

https://www.librarything.com/topic/327130