gender field

does this need to be a binary? it seems like it's included to make searches on multiple criteria possible in the future (at least, i assume so, otherwise it's pretty useless.) with that in mind, a text field for authors who identify as genderqueer, trans, whatever outside of male/female would be quite useful

In the blog post, Tim wrote:
****And yes, we can discuss the whole radio-buttons-for-gender topic. See here, here. I'm of the opinion that two genders plus maybe "unknown" and "n/a" (for Nyarlathotep?) are the best you can get without consensus-splitting disagreement. You'll note we aren't including other potentially-contentious fields, like sexual orientation or religion.

I think his four-way solution is a good one myself.

What about multiples? PJ Tracey is one example. Ok in this specific example its a pair of females so female still works, but other odd authorshiops exist. I suppose they all become n/a

what, exactly, are the arguments AGAINST adding fill-in-the-blank? it doesn't detract from the utility of the radio buttons, and it's no more open to vandalism than any of the other fields. i think unknown is another useful addition, but it hardly covers all the bases. and n/a sucks. christ, even "other" is better than not applicable

I think "other" should cover it - male/female/other.

> 5

female/male/both/other, to cover #3's point. There are "authors" who are actually a man and a woman writing under one name.

>3 reading_fox:
An example is Jonathan Wylie the name used by Mark and Julia Smith. Also Nicolas Bourbaki is more than one person.

perhaps radio buttons for
single author, Group psuedonym, Organisation.
In the last two cases birth and death deates become meaningless.

I don't see a problem with "n/a" - an organization, for example, isn't "other" gender, it simply doesn't apply. We do have corporate authors and things like that, and I think it's more desirable to give the signal "hey, this isn't even a person" rather than the signal "prefer not to choose male or female."

8: okay, fair enough...but I think an option ought to be given for those authors who DO "prefer not to choose male or female." this is a distinct situation from "unknown" or "not a real person."

Either "n/a" or "other" I think could work for people who prefer not to choose. "N/A", while it's ugly, has the advantage of being vague about why it's not applicable: not applicable because not a natural person? (corporate authorship) not applicable because the binary is inapplicable? (intersexed, trans-identified) not applicable because chooses not to identify? (e.g., Raphael Carter)

"unknown" I agree is distinct from "n/a". It might apply to, for instance, historical writers who used pseudonyms....

... is there some reason I wonder why the touchstone for Raphael Carter is red and not listed above? a server sync issue I wonder?

From the blog, with some additions.

But neither of these apply to the case of, say, Leslie Feinberg, who rather firmly identifies as "none of the above". ... Can you really construct an articulate rationale for being so exclusionary?

No, I shall have to be inarticulate. (That the conversation of this topic starts with a swipe proves my point quite well.)

The problem is that LT CK is designed for facts to which all can agree or reach an easy consensus. Opening the gender field up for free text is very handy when there's a single editor, particularly when it's the person in question. So you get sites which present a dozen or more genders, like Pownce.com, where members can choose to identify as "dude," "lady," "guy," "bloke," etc. That works less well when anyone can have their say. (Is Shakespeare a dude or a gentleman?) What I don't want is identity fights over dead authors and endless terminological fights, whether bisexual is to give way to pansexual to avoid binary connotations, and so forth.

As I discuss at some length in my talks, tags like GLBT and LGBT may seem like synonyms in a cataloging context, but the actual book-sets are rather different. It's fascinating to watch users choose their terms and create spaces of meaning for themselves. Capturing that dance of identity and meaning is what tags do best. Common Knowledge takes these many and varied worlds of meaning and pushes them through a cake nozzle. There's one meaning and someone gets to decide it. If you think it's you, you're probably statistically wrong. I haven't taken a census recently, but I get the distinct impression the genderqueer group here is outnumbered. I suspect that a lot of members of other groups would fight any efforts along these lines.

Much the same is why LibraryThing has no space for religion or sexual identity. These are things we get to decide for ourselves. Letting others do it is a recipe for disaster. I myself have spent four years watching the Wikipedia community tackle the sexuality of Alexander the Great. Can they do it? The results are not encouraging. Most Wikipedians can't understand that ancient concepts are not modern ones, and everything is clouded by thick partisanship, so Wikipedia lurches violently from "gay" to "not gay," with all manner of nastiness generated in the lurches.

FWIW, I think that based on my own experiences and unscientific surveys, by far most of these kinds of labeling disputes in wikipedia come about on biographical entries. Labeling people, in other words, creates a lot more issues than labeling works. (There's some fairly obvious psychology for this.) So it may be quite wise to be slow and careful and cautious and risk-averse and minimalistic on the author pages, while going a little wilder/crazier on the work/edition pages.

Reposting comment by "language" from the blog, in case someone thinks I'm crazy to suggest that opening this up will cause controversy too:

EDIT: Please note. I didn't write this. Don't either thank or criticize me for it.

---

Zir gender is well-known, and zie certainly *has* a gender -- it's just neither male nor female. Can you really construct an articulate rationale for being so exclusionary? It may affect only a small number of authors -- but it's a *very* important issue for those it affects. Please reconsider.

I strongly disagree. I could announce that I wanted all references to me to be accompanied by hopping up and down on one foot, and this was very important to me, but too bad for me, nobody would do it. A tiny group of people do not get to throw monkey wrenches into social software to satisfy their ideological cravings.

And of course Leslie Feinberg "has a gender"; the fact that she doesn't like to be so categorized has no bearing on the issue. From the Wikipedia entry:

Leslie Feinberg is Jewish and was born female and today prefers gender-neutral pronouns "hir" and "ze". Feinberg writes: "I have shaped myself surgically and hormonally twice in my life, and I reserve the right to do it again.

That's all well and good, and she has the right to do whatever she likes with her body and to use whatever pronouns she chooses; she has no right to try to control the usage of the rest of humanity.

Also, words like "zir" are terminally dumb. English has perfectly good ways of being gender-neutral; both "they" and "he or she" work fine.

Tim will now be known as Tim "hopping up and down on one foot" Spalding.

i'm not going to get into that comment by "language." suffice it to say that i think when the way someone else identifies themselves makes you that angry, your response says a hell of a lot more about YOU than about their choice.

tim, what about actually implementing the four-choice selection you mentioned? people have listed several reasonable justifications for providing a "not applicable" choice, and perhaps the fourth could be called "none of the above" rather than "other" (which has off connotations to me, but whatever)?

Editing the above message to make sure people don't think I wrote it. I didn't!

Unfortunately, while the commentator Tim posted has a black-and-white opinion, biology isn't quite so black-and-white. The simple fact that some people are born with chromosomes other than XY or XX, or with chromosome sets that are discordant with presumed physiognomy, means that whatever the popular belief about "male or female", it's not strictly accurate in all cases. ... In other words, even people who discredit the reality and gender identity that transgendered folks claim must still contend with ambiguous genitalia, Turner syndrome, Klinefelter syndrome, Kallmann syndrome, androgen insensitivity syndrome, etc.

... which is why binary doesn't work.

The complex politics and intense personal attachments to and beliefs about gender are why this is a fraught area.

... which is why a limited-option category seems (to me) like the easiest choice for avoiding revert wars.

I keep turning over "no public gender identity" or some such as a way to handle Raphael Carter and Leslie Feinberg. It's not quite right -- both have a public gender identity of "neither male nor female". But there ought to be some way to respect a living author's privacy, where the author chooses not to state.

I would be happiest with a three-value field:

* male
* female
* other, not applicable, or unknown

There's no need to distinguish between the options in the third choice. And we're going to need a third choice to handle authors like 37signals.

Note that there is a difference between "definitely unknown" (as with some historical figures) and "LibraryThing does not have this data yet". So I don't recommend using a blank field as the third option.

There's no need to distinguish between the options in the third choice.

I don't know about that. Don't you think people might be interested in seeing all the individual humans that identify neither as male nor female, unmixed with corporations and groups of people?

Then of course there are authors like Jan Morris - do we need to disambiguate between his early work and her most recent writings?

no touchstone

Written on a plane (and finished, after missing my plane, in a lousy hotel near Dulles).

I see this whole issue as the basic problem of cataloging: someone has to be right. In Dewey, Islam shares space with the Bahai and Sikh faith because that's how Dewey saw it. Someone else might disagree, but someone has to be right. As I said, cataloging takes all the answers that people might think are valid and squeezes them through a cake nozzle.

Now, there are some some good and powerful things about cake nozels. Insisting that we Boston, Massacusetts by its real name and not "Bean-town" allows people to find books by Boston authors.

Ultimately and quite frankly, my opinion on the "topic" pales in comparison with my interest in making things work. That means avoiding fights and it means a system that does useful, interesting things.

Those two instincts are in conflict, I think. I can see--indeed, I predicted--that the gender issue is going to be a source of contention. There's snipiness and worse. The feature isn't a day old and from the first I spoke of the need for discussion, yet I'm being asked "what about actually implementing the four-choice selection you mentioned." We're not rushing this. We're not going to argue about what everyone has to think and believe. We're not getting into the soup.

Instead, let's keep the use of the system before us. Nobody is going to convince anybody of the "truth" of gender. I utterly oppose any attempt to use a classification system to force through something that isn't about the usefulness of the system. Clay Shirky calls this a "voodoo ontology," an attempt to change the world through the ontologies we use. His target is mostly "conservative" (Dewey and his biases); the impulse is unfortunate whatever its perspective.

To avoid conflict, my instict is to come up with a small number of options, such as "male," "female" and "other/not applicable/unknown." I don't think we can come up with much more and get substantial agreement. The more you open these things up the more you're getting into contested space. Gender identity shades from fact to, well, opinion. Cataloging doesn't mesh well with opinion. I am particularly concerned about fights over the gender identity of this or that dead author. We'll never settle it.

But searching is also important. Ultimately, the goal of the system is to enable that. (It's not about arriving on the Mark Twain page and confirming he had testicles.). I want to be able to get a list of the top Portland-based female mystery writers. I want to find that Austrian guy who wrote a poem about the diseases of lumberjacks. In this vein, some people will want to know the top transgender mystery writers. Right?

If searching is the point, you'd want to allow multiple genders. When it comes to searching, more is better, and opening the system up to a free-text soup of terms (eg., genderqueer, transsexual, transgender, bisexual, pansexual, etc.) will make multiple listings particularly useful.

As I said, difficult issues.

I'm not really sure why a writer's gender is important, but I'm sure people will want to search on it. But if you can search on gender then I'm sure someone will want to search on race and/or nationality, which opens up an even bigger can of worms.

I had a front row seat to a Wikipedia squabble on who should be in the category Irish People. Some wanted to be able to add people for whom they could trace a distant Irish ancestor while others insisted that ancestry didn't matter and you had to find a statement by the person that they considered themselves Irish. The dispute is probably still raging.

How about instead of "n/a" or "other" just plain old "-" which says less so is more open?

Another option is leaving the gender field blank.

Another idea: why not have the 3- or 4-field standard set of options and also a free-form text field both under gender? Thee free-form field can have a + so you can put both male and female for someoby who has undergone gender reassignment, or for when when there are irreconcileable differences between what two groups believe belongs in the free-form field. Over time a set of more-or-less standard terms should bubble up from the free-form data.

Leaving it blank just invites people to fill it in.

I agree on the leaving-it-blank issue.

Something I noticed today while doing something else entirely, was that LT may already have a gender problem. I own the Settlement Cookbook, by Mrs Simon Kander, who is now identified in LT primarily as Simon Kander, which is patently not true but seems unfixable. She can't be the only one.

-->29 timspalding:

But you know, Tim, that people feel *compelled* to fill in blank spaces! :-)

--> 30

I settled it. I checked the "female" gender. Now it's the name that is (sort of) wrong, not the gender. The name issue was handled by the "combiners"*.

I even went so far as to add a clarifying link and an "aka" under the disambiguation notice.

P.S. That's the beauty of CK!

Incidentally, re the "leaving it blank" issue, once a button has been checked it cannot be unchecked -- the gender can be switched, but not re-blanked. So if someone through ignorance or malice checks "female" for Leslie Feinberg, there's no way to undo that.

--> 32

It seems as if we should have that option!

male
female
"non-human" (corporation)
"group" (multi-author (multi-gender) group?)
"not stated or unknown"

my wording is "off" because I'm not fully awake, but getting a "non-human" authorship & then a group or multi-gender choice as well as a "not stated/unknown"

I willingly throw these phrases "to the wolves" to clean to the bone my awkward vocabulary.

"Corporate author" for a corporation or other organization?

I don't see that a multi-author group needs a separate category - corporate author would work for that too (if it's good enough for AACR2(R), then it's good enough for anyone... but maybe that's my professional cataloguer bias).

"not stated, other or unknown," perhaps

I like non-human - it opens up lots of future possiblities.

I do not like "Corporate Author" for multi-author. The Felix Francis and Dick Francis book that just came out just happens to have two names on it. It isn't a book put out with "Microsoft" as the author. All those academic books with 10 names are not "Corporate" books.

>35 skittles:

Organization rather than corporation, methinks.
Collaboration? for multi-author.

Not happy with 'not stated or unknown': what about those who do claim transgender or other such appellations?

The "suggestions" I gave were just that... suggestions. I was very sleepy & I didn't completely like what I had offered, but it was all my brain could come up with in its muddled state.

Collaboration, perhaps, but in the case of Felix & Dick Francis, I think one is the primary author & the other secondary... in most records there is a primary author, then secondary... which is what would happen in that case... I think.

I also prefer organization instead of corporate.

I also agree with the "not stated, other or unknown" I realize that it doesn't include all of the gender possibilities, but I was trying to keep it simple with my original suggestion.

39> Sorry - it was late and I was not clear in how I was thinking or expressing myself. "Corporate author" would be for a collective or organization with a name - for example, Better Homes and Gardens or Copper Pig Writers Society or Oysterband

I dislike "non-human" because obviously, humans are part of the authoring organizations. Microsoft does not write books. People who work for Microsoft write books. The only reason I'd accept "non-human" is if books by actual aliens (or plants) who don't have gender start getting added to LT (I wish!).

Re: multi-author: At present (and I realize that this could change any time, the way LT evolves) we have an "author" field and an "other authors" field. The first-named author gets the "Author" field and any other author(s) go in the "other Authors" field. So why would we need a gender for the multiple authors? The "other authors" don't even register in relation to the book. I assume that once LT gets a better author system each author will have his, her or its own page and again they don't need a "multi-author" gender.

Edit: I forgot about single pseudonyms used by multiple authors (e.g. Emma Lathen). But maybe one should recognize the intent of the multiples in that case which is to present a unified front and select the gender applicable to the pseudonym. Or use Not Applicable with a note or tracings for the individual parts of the whole.

That's how I see it, anyway. I'm open to persuasion :)

42> The intent, eh? :) I shall go change James Triptree Jr. to male then. :)

Oh, and the Felix & Dick Francis book: It is indicated that Felix did the research in prior books (the ones written after Dick Francis' primary research, his wife, died) and steps up a little more in this book. Whatever that might mean. :)

Gah, forgot James Tiptree Jr. Come to that, many female authors have used ambiguous names and pseudonyms. E.g. Andre Norton, CL Moore, who also wrote under clearly male pseudonyms, too. I would still count them as female (also Tiptree) but if I came across a book by Andrew North (pseud. Andre Norton) or Lawrence O'Donnell (pseud. CL Moore) and didn't know better, I would assign those as male. Until I knew better and then I'd switch them.

Of course, CL Moore and Henry Kuttner co-wrote under the name "Lewis Padgett" which comes back to the intent to project the male image, I guess. Emma Lathen is less ambiguous because the two authors behind that name are both female, so assigning "female" to Emma Lathen is at least not wrong.

And what about "house" names like Carolyn Keene? Any number of male and female writers have used that pseudonym but the intent of the syndicate that owns the name is that it be taken for female.

Double gah. Maybe assigning gender is more trouble than it is worth?

Ok here is another "Maybe assigning gender is more trouble than it is worth?" one:
Lydia Adamson is either a Pseudonym of Franklin B. King, or Frank King is a pseudonym of Lydia Adamson. One or the other. Now I messed it up by putting Adamson as male, but can't figure out which gender the author is now :)

According to Google Frank King is the real person and Lydia is the pen name...

It can be really confusing.

Leigh Greenwood is a man but since he writes books generally perceived as romances many people aren't aware of that.

37>

How about not stated, other, multiple or unknown?

I agree that, if there must be a gender field, then it should be limited to three alternatives.

Just to throw another wrench in the works: instead of the options in the CK "Gender" category being "male" and "female", could they be something like "masculine" and "feminine" (but not necessarily those words)?

Or, as an alternative, if you want to stick to "male" and "female," how about renaming the field "Sex"? My cat is a female of the species Felis silvestrus catus, but not being Homo sapiens sapiens, she lacks gender altogether. The whole point of the word 'gender' in non-linguistic contexts is to distinguish the social and cultural associations of sex from the biological aspects of sex. Equating 'gender' with 'sex,' though, helps naturalize community-specific ideas about sex-appropriate behavior, which is what the various sex- and gender-rights movements have been working against. (And renaming it "sex" would not prevent the introduction of additional options, either.)

Edited to wonder: why do we even need a sex/gender/whatever field in the first place?

Edited to wonder: why do we even need a sex/gender/whatever field in the first place?

"I'm looking for women fiction writers living in or writing about Africa..."

Can you do combined CK and tag searches? Because CK alone won't help to ID anyone writing fiction about Africa. In fact, unless there're some sophisti-ma-cated data-parsing & query-nesting routines in place – and I haven't seen any evidence of such –, CK won't help to ID all shared more-general locations like "California" or "Africa," unless they are placed on their own line in a CK field. "Los Angeles, California, USA" doesn't show up in searches on "California" or "California, USA" (neither does a search on "California" return entries using "California, USA", nor vice versa).

I think Africa would be an "important location" in a novel about Africa. And from what's been said on other threads, that type of sophistimacated data-parsing is exactly what we're trying to get closer to.

#51

Well yes but what if someone just entered "Cape Town, South Africa" or "Nairobi, Kenya" in the important locations. The knowledge that Kenya is a part of Africa isn't embedded in LT.

Also you have books such as Ivory or Paradise by Mike Resnick. Both of these are about Africa to some extent (both have africa in their tags) yet they are not set in Africa but on other planets.

I think that eventually someone will have to write a more sophisticated search that combines tags and CK. However CK is still in its infancy so we shouldn't be surprised that everything isn't there yet.

Has Tim posted anything anywhere about why he's doing this with a wiki rather than with tags?

There will be more sophisticated search into the CK data. Additionally, CK will have an external API so that anybody can do whatever the like with the data (visualize it, embed it into their application, search it, etc.).

And it's not really a wiki. It's a 100% custom solution that happens to have edit histories (something that people associate with a wiki). We were originally planning to use the actual mediawiki engine by having people create pages in specific formats and just scrape data from it (think of THAT when you are using CK!). In the end, the mediawiki thing just wasn't the right way to go for this information.

We probably could have gone the tags route with machine tags like Flickr does (location:Portland, Maine), but I don't think we would have achieved the 55,000+ entries that we have at this point if we had done it that way. I really feel that the interface and (relative) ease of use are powerful draws for people.

CK allows some structure that tags do not. And a wiki is better for when there *is* just one answer.

> 37 and 55

Tim, I could accept "not stated, other or unknown" as a third option. I'm unhappy with how the current options for this field reinforce binary gender thinking, which has hurt friends of mine (I'm talking assaults/hospitalization, not hurt feelings).

If I may remind you of what you said last year about LibraryThing on whether Pluto is a planet:
"So, it's "sort of" a planet. Or maybe it's a planet, but not a very good example of one. That's a perfect LibraryThing answer. Non-binary, non-authoritative. Pretty good answer though."

http://www.librarything.com/thingology/2006/08/librarything-on-whether-pluto-is....

I'd be happy with "Not stated, other, or unknown" as well. It's not quite as good as some of the alternatives that have been discussed (in particular, including "unknown" in the mix makes it seem like a default, such that people who don't know better would still say "oh, that sounds like a female name, I'll check the female box" when they wouldn't if "Neither/Other" was explicitly stated, but the gap between those two options is FAR smaller than that between the current binary and either of them.

As selkins says, the binary is something that causes real physical harm to people who don't fit into it. Whereas the benefit is...what, exactly? It's not an academic issue like whether or not to spell out US states.

>56 selkins:

I totally agree with you—and with me. That's the problem of regular cataloging, of which CK is a queer spin-off. Someone has to win. For all the tagging that goes on—and more tagging goes on every day than has yet taken place on CK since it started—there's very little controversy. The one who use "homosexual" aren't fighting with the people who say "queer." There are some consequences, but they're not as severe.

To each his own works...

Tim suggested I post a link to my blog entry about Common Knowledge and this topic in particular: http://configures.sarahelkins.org/?p=55 has a long interesting comment from Tim (quite gracious in the face of my snark), partly about this topic, and partly about the more general problem of categorization, who decides, etc. Tim, I wrote a short reply there, but in case you don't see it, even more briefly: you were right about my use of scare quotes (edited), and you're welcome to use the image I created. There's a lot more in your comment I haven't gotten to yet; I'll write a longer reply or another entry later.

Also, for the convenience of anyone from LT who goes there, I've added a LibraryThing category to that and a couple of older LibraryThing-related entries: http://configures.sarahelkins.org/?cat=15 (e.g., one when I added my local library system to what was the Find At ... list in LibraryThing), more specific than the wider "book" category I had already.

#5:

I prefer the option of 'other' also.

I still like "non-human" as a choice... even if it isn't an organization or corporation...

there are some people I've met (& read) that I wouldn't quite classify as "human" ... although, there is a possibility that they might classify that biologically.

I just wanted to bump this in hopes that one of the many options suggested will be implemented. At this point, really, I don't care what it is. I just wish something other than the strict binary was an option -- even if it was changing the radio buttons to checkboxes, so that we could go back to blank if someone marks one where it isn't applicable.

It's been a long time coming, but...
We have added new options to the gender CK field:
'other/contested/unknown'
'n/a'

Hopefully these choices, together with the original 'male' and 'female', will capture nearly all permutations for gender.

Hooray!

>63 conceptDawg:

You sir are the man!

Hooray! Thank you very much.

65: That's too bad. I was going to change my author page to read 'n/a' but now I must keep is as 'male' since I'm the man.

Correction: You zir, are excellent.

>67 conceptDawg: You could just change it daily to keep people on their toes.

Perambulist!