Author gender

We made a change that got somewhat buried in a long thread on the topic. For the State of the Thing we wanted to have something clean and non-complex to link to.

The change is that author gender has moved from a menu with five options to a free text box.

Here's what the field looks like:


For now, the field:

1. Is free text
2. Has the suggestion text listed above
3. Is unitary, unlike some CK fields which can contain multiple lines.
4. Not translated. I don't currently have a good answer for this. I don't want to split the field up by language, but I also can't translate free text. (If I did so, and someone entered a typo or a weird phrase it will go onto the translation list of 30 languages. It a recipe for chaos.) For now, we have to use English.

I have NOT changed the stats page. I will be changing it when the new design reaches it, but it's not in the first wave of new designs.

My reasoning for a free text field was spelled out in that long topic. A complicated UI wasn't going to happen, for multiple reasons. Nor do I feel comfortable coming up with an official, or above-the-fold, genders. As discussion showed, informed, intelligent and respectful people will not agree in every respect. =

You can see recent changes here: https://www.librarything.com/commonknowledge/search.php?t=2&f=5
You can see a "cloud" of currently-assigned genders here: https://www.librarything.com/commonknowledge/clouds.php (cached)

As I see it, the problem now is to maintain some uniformity. We went with a free text box because LibraryThing isn't qualified to come up with a list of official, accepted genders. Proposals along those lines invariably met with opposition. But we need to be careful to avoid pointless differences, such as "non-binary" vs. "binary."

>1 timspalding: something clean and non-complex to link to

Is that a complicated way to say "simple"? ;)

Thanks for making this change!

The translation problem is coming up with männlich, if anyone wants to fix it.

I suspect the solution to this is to get local translations for the top X genders, so genders show up right in other languages, and if someone uses them, it translates them back into English.

Is that a complicated way to say "simple"? ;)

It's also easy.

I'm not sure if this is new but on the author gender page I'm getting a few authors under "not set" who have set genders on their author page.

Corsino Fortes, Hilda Hilst, Bjørg Vik, and Per Wästberg

I just set genders for some other authors who really were "not set" and they migrated to other categories, as usual. That page is very important to me - I hope it makes it through the changes.

>5 Settings:

Thanks. I'll check on it.

>6 timspalding:

I have the same issue as >5 Settings:. It's been like that since you changed the field.

ETA: Paul Claes, Victoria Charles, Alex Bolckmans, Maria May

Is there a reason that clicking on some of the less-frequently used genders doesn't result in a list of authors who have recently been edited to that gender? It says "1–0 of results."

I wanted to clean up the capitalization to help standardize the fields, but I could imagine someone might want to read from gender fluid authors and be upset not to be able to find any results.

Also, it appears someone is going around removing genders from authors (among other things). I reopened the ck thread on "bad edits" this helper has previously made... is there a way to stop or revert their edits? It appears to be a longstanding issue.

>8 parlerodermime:

You might also try searching, but my bet is that you were looking at the cloud and the cloud is cached. Someone already edited them away.

>8 parlerodermime:

I know there have been a lot of complaints before, but looking at their recent edits, I see only the removal of genders from the "root" author of author splits, and from splits that no longer exist. Which ones concern you?

Yeah, because they edit with such volume, it can be hard to glance at their edits and see what's wrong: a lot of what they do is editing over information that's already there to simplify it or remove it altogether. Off the top of my head, I can't weed through the gender field and remember which ones were for the legitimate purpose you named vs. which ones were for real longstanding author pages that they decided shouldn't have that data anymore. I definitely saw some last night though, and I can go looking again if it's important.

Here are just a smattering of general edits from the last week that don't make sense:

If you look at the Edwin A. Abbott page, you'll see that one of the entries for his education (added in 2015) used to be: "University of Cambridge (St John's College)." This was nice, since it indexed to the list of authors who also attended the University of Cambridge, while also providing the extra information that he belonged to the St John's residential college there. They changed it to "John's Collegle, Cambridge" which is spelled incorrectly, makes it sound like it's a small college which happens to be in Cambridge (MA? UK? unclear) rather than being a prestigious part of the University of Cambridge, and doesn't index to anything.

For whatever reason, they also decided that Abbott's occupation "Anglican priest" needed to be changed to "cleric," which I guess is still true, but at least to me, is not as interesting.

They've been removing details about cause of death, changing many subtypes of cancer just to cancer, or xxxtentacion's cause of death from gunshot to just "shot," which is ambiguous as amanda4242 pointed out in the cause of death discussion thread about standardizing this stuff.

Oddly, it's not just simplifications that they make though. They edited Rachel Joy Scott's cause of death from "school shooting (Columbine High School Massacre)" which was built to index with other school shootings, in case that became relevant, to "shot to death at Collumbine High School" which wouldn't index to anything since it's so specific, and, again, is mispelled. They also removed the secondary data field which listed gunshot and where she was shot in parentheses.

I also see lots of link removals and author removals. For instance, in the last couple of days, they removed the two coathors of Revolution, Resistance, and Reform in Village China ...it's an academic book, they should probably all have equal billing, like they do on the cover, but it's just the one default primary author now instead.

I could look through their history more and come up with a longer list. Personally, I'm at a point where I've seen enough weird edits from them that I don't trust the data they do input to common knowledge. There are some trends in what they do, but also a lot of contradictions, and it's a private profile, so there's no way to contact them.

>11 parlerodermime: Someone ought to fix some of the edits you list in your post, for spelling errors if nothing else. Putting John's Collegle, Cambridge instead of University of Cambridge (St John's College) is just wrong in my opinion.

>11 parlerodermime:

They edited Rachel Joy Scott's cause of death from "school shooting (Columbine High School Massacre)" which was built to index with other school shootings, in case that became relevant, to "shot to death at Collumbine High School" which wouldn't index to anything since it's so specific, and, again, is mispelled. They also removed the secondary data field which listed gunshot and where she was shot in parentheses.


The correct cause of death in that case is "gunshot". "School shooting" is not a "cause of death".

Perhaps I've read too many autopsy reports, but I see a distinction between cause of death and manner of death.

Examples:

John Jones shoots himself intentionally. Cause of death is gunshot, manner of death is suicide.
John Doe shoots John Jones intentionally. Cause of death is also gunshot, but manner of death is homicide.
John Jones is cleaning his gun and it goes off, killing him. Cause of death is again a gunshot, but manner of death is accidental.

I've started a thread for cause of death best practices. Should we have one for gender, too?

https://www.librarything.com/topic/324608

>12 al.vick: Yeah. I was just looking for good examples to show Tim, and those were the ones that came to mind. I figured I'd edit them later.

>13 lilithcat: That's fair. We haven't standardized how to fill out that field. To be fair, my original population of the field had two neat separate index-able entries, one for school shooting, and another for gunshot (with wound placement in parentheticals), so, I had originally listed what you liked. I just also had a field for school shooting as a special circumstance.... Anyway, we can continue this on the "cause of death" best practices thread.

>11 parlerodermime: I'm not going to engage on the more general topic of C's edits here. There are other topics for it. It is definitely an ongoing topic of concern among members and staff. I just wanted to ask about gender--because I'm not seeing problems there for this user.

Wait, the individual in question is also deliberately misgendering people? That's a much bigger problem than just quibbling about level of detail for cause of death.

>18 lorax: They're not misgendering people; they're deleting the information in the gender field when there doesn't appear to be a reason to do so. See their edits on Ron Gorchov and Al Kasha, for example.

>19 amanda4242: Yes, this. Thank you for digging some better examples up, amanda4242.

>16 timspalding: and it wasn't C's edits, so no need to smirk, >17 Collectorator: you even started the "bad edits" thread in question.
Point taken that I should have just taken gilroy's advice and emailed instead of making this talk post though.

Well that's weird. Not as blatantly offensive as I feared, but just weird.

>21 Collectorator: But I wasn't blaming you, and when you make these accusations, it distracts from the actual conversation at hand. I was blaming someone / an organization you personally have described as making "hundreds of bad edits" which constitute either "intentional wreckage or sloppy editing."

Alright, >16 timspalding:
I went looking for some of the gender removals I'd seen earlier: In addition to those listed by amanda4242 in >19 amanda4242: this user has removed the gender of: J. O. Walker, Joseph Bartscherer, Eric Wald (they just removed all ck here, rather than specifically targeting gender), R.B. Cundall, .........

The original point I was trying to make in >11 parlerodermime: is that they get away with it because of (1) their editing volume making it hard to review everything they do, and (2) cases like this where they are indeed removing gender from the disambiguation page: Heather Chapman, so just looking through their edits means weeding through a lot.

Personally, I find the edits sloppy/malicious enough that I don't trust the data this user inputs. For instance, they added a disambiguation notice to Stephen F. Williams's page that it's not the same Stephen F. Williams who died of COVID-19 this August, and took that opportunity to delete ALL ck data on that page, even though, it's pretty obvious with even a short google search that at very least, "occupation" was correct and shouldn't have been removed. And he's gendered in that book review as he, so unless he was misgendered there, it shouldn't have been removed either. And wow, look here, the book titles on the page are further supported by the wikipedia entry which also says he died in August of COVID-19, and here's a news article that makes clear it was all correct data.

Yes, some of the edits fall under the category of removing data from "splits that no longer exist," except I'm guessing they're also the one doing the split? And unless I'm misunderstanding the process, I don't think these are legitimate splits.

See: Leroy "Nicky" Barnes vs. Nicky Barnes. Given the activity level on both pages on September 8, it seems they split off a few distinctive names that probably ought to be still grouped with the Nicky Barnes page (or else folks who have categorized the books under those name variations in their libraries now have a book where the author doesn't point to the author stats). Given that these are stranded names with now "0" books, that seems likely. They migrated some information, but deleted it all from the stranded author page, and didn't migrate gender.

This, and the similar case of Trần Ngọc Châu vs. Tran Ngoc Chau on September 6. Again, they migrated some information, but not gender.

I would attribute failure to migrate all relevant data when splitting to sloppy editing.... but it's still a loss of data. And I don't understand the need to make these splits.

Sure, maybe it's hard to make the case that they're specifically targeting the gender field, but there are certainly some examples where removal of the gender is the only edit made to a page.

Wondering if some of this conversation would be better over in the "bad edits" thread.

One further problem with that account: They have a pattern of editing the input of specific LT accounts. I thought it was just me, but so far I've found seven accounts they regularly follow and "correct" -- and that's looking at just one of their recent editing sessions.

This field is completely unnecessary. Instead of 5 options or free text, it could be eliminated altogether.

>26 Petra.Elster: Why don't we eliminate the name field too while we're at it?

>26 Petra.Elster: None of the CK is necessary, but many people find it interesting information. Also, Tim pointed out on the RSI thread that the gender field plays a small role in recommendations.

https://www.librarything.com/topic/279371#7173443

>26 Petra.Elster:
"completely unnecessary", for you maybe. I find it useful.

>26 Petra.Elster: Oh come on, it sparks so much debate! Surely you wouldn't want people to do something productive instead?
I understand and respect the sentiment, though.

>6 timspalding:

I also have gotten a good bunch of authors switched to ‘not set’, which have their gender set. I tried to remove and set gender again, but that doesnt help.

Do you need a list of the authors because it is fairly long (around 35) but an example is Helga Flatland

Thanks

Should I set James Herriot (the veterinarian) to male? I think that is a pretty safe bet.

>32 krazy4katz: Yes. There are combination issues on the James Herriot disambiguation page, so it looks like whoever did the split didn't bother to transfer the CK to James Herriot (1).

>33 amanda4242: >34 MarthaJeanne:
Thank you! Yes, there are combination issues. I went to the disambiguation page and place all the "other" works that I was sure belonged to him under his name. However trying to combine was tough. There were abridged audio works that could not be combined because the work-to-work relationship wasn't done etc. I don't think I am ready for that level of effort at the moment.

I moved a few more unknowns over to James Herriot (1). I think they are probably all him; with some it's never going to be possible to be 100% sure, e.g. the Reader's Digest Condensed Books, but it seems unlikely it's anyone else. Intriguingly James Herriot (3) seems to definitely be a different person - the book in question was published knocking on 20 years after the vet died.

James Herriot

>36 mart1n:

It also doesn't seem to be the sort of book Herriot (1) would write!

I found more "unknowns" that belong to Herriot(1), thanks to Google translate.

>36 mart1n: >37 lilithcat: Thank you! I was also unsure about the Reader's Digest Books, so I left them, but at least he must be a co-author, so probably it's OK.

I reckon all of the remaining ones must be him; he's such a huge and (almost!) unique name. Would it be wrong to bung them all over?

What I see there are

a number of RD Condensed books in various languages. 99.9% JH1

a number of collections of JH novels. again 99.9% JH1

A Cricket (magazine) story anthology. includes Tosh / James Herriot -- I'll look further, but I would say at least 80% JH1

I agree they look as though he is at least a contributor to the RD Condensed books. I don't know how to get them into the section that says "Also by James Herriot on his page, which is where the other RD books are located. If you go to each of those books and list him as a contributor will they end up in that section?

>41 krazy4katz: What you want is for the authors to look like this one: https://www.librarything.com/work/9454552

If he is listed as primary author, the book will stay in the upper part.

Thank you! Sorry I digressed from the purpose of this thread.

Okay, there's clearly some sort of combination wiggle-waggle going on for a small but persistent group of authors. I'll dig into it soon, when I get a chance.

>36 mart1n: If it helps, I have seen contributions from James Herriot (1) appear in RD Condensed books.

>45 -pilgrim-: Thank you. I think we have moved all of them that mention him over there now.

I still have a problem with several, 40+ authors, who refuses to move away the not set category even when they do have their gender set to female or male. Are there still problems with New feature or am missing something that I should do to Update them?

>47 Alendor: When the recent change was made, some authors seem to have lost their connections. Just go into each one, click on the edit pencil, then save. I've done a few for you, and they have changed categories.

>48 MarthaJeanne:

That is very weird. I can see that number has fallen so thank you, but when I click on edit and save... nothing happens. I tried a couple of different authors and tried to wait several days in case it needed some time.

>49 Alendor: Which site are you using? It may be that the English site works better than the international sites.

>49 Alendor: I have now resaved all those for whom there was already a gender entered.

>50 MarthaJeanne:

Thank you very much. That was the solution - I have to switch from the Danish to the English site for it to work.

>50 MarthaJeanne:
Thanks, that was the solution.

To be honest, I did try to edit and resave the gender data back in September, but it wouldn't stick then. Not sure what's changed behind the scenes.