+-------
+
+Somewhat awkwardly, I actually had a date scheduled with "Noreen" that evening. The way that happened was, elsewhere on Facebook, earlier, on 7 February, Brent Dill had said that he didn't see the value in the community matchmaking site _reciprocity.io_, and I disagreed, saying that the hang-out matching had been valuable to me, even if the romantic matching was useless for insufficiently high-status males.
+
+"Noreen" had complained: "again with pretending only guys can ever have difficulties getting dates (sorry for this reaction, I just find this incredibly annoying)". I had said that she shouldn't apologize; I usually didn't make that genre of comment, but it seemed thematically appropriate while replying to Brent (who often espoused cynical views about status and social reality).
+
+Incidentally, I added, I was thinking of seeing seeing that new _Hidden Figures_ movie if I could find someone to go with? It turned out that she had already seen it, but we made plans to see _West Side Story_ at the [Castro Theatre](https://www.castrotheatre.com/) instead.
+
+The date was pretty terrible. (Or, maybe I was the only one who categorized it as a "date"? Maybe in her ontology, we were just seeing a movie.) We walked around the Castro for a bit continuing to debate the gender thing, then saw the movie. I was very distracted and couldn't pay attention to the movie at all.
+
+------
+
+I continued to be very distracted the next day, Monday 13 February 2017. I went to my office, but definitely didn't get any dayjob work done.
+
+I made another seven Facebook posts. I'm proud of this one:
+
+> So, unfortunately, I never got very far in the _Daphne Koller and the Methods of Rationality_ book (yet! growth m—splat, AUGH), but one thing I do remember is that many different Bayesian networks can represent the same probability distribution. And the reason I've been running around yelling at everyone for nine months is that I've been talking to people, and we _agree_ on the observations that need to be explained, and yet we explain them in completely different ways. And I'm like, "My network has SO MANY FEWER ARROWS than your network!" And they're like, "Huh? What's wrong with you? Your network isn't any better than the standard-issue network. Why do you care so much about this completely arbitrary property 'number of arrows'? Categories were made for the man, not man for the categories!" And I'm like, "Look, I didn't get far enough in the _Daphne Koller and the Methods of Rationality_ book to understand why, but I'm PRETTY GODDAMNED SURE that HAVING FEWER ARROWS MAKES YOU MORE POWERFUL. YOU DELUSIONAL BASTARDS! HOW CAN YOU POSSIBLY GET THIS WRONG please don't hurt me Oh God please don't hurt me I'm sorry I'm sorry."
+
+That is, when factorizing a joint probability distribution into a Bayesian network, you can do it with respect to any variable ordering you want: a graph with a "wet-streets → rain" edge can represent a set of static observations just as well as a graph with a "rain → wet-streets" edge,[^koller-and-friedman-i] but "unnatural" variable orderings generate a more complicated graph that will give crazy predictions if you interpret it as a _causal_ Bayesian network and use it to predict the results of interventions. Algorithms for learning a network from data prefer graphs with fewer edges as a consequence of Occamian [minimum-message-length epistemology](https://www.lesswrong.com/posts/mB95aqTSJLNR9YyjH/message-length):[^koller-and-friedman-ii] every edge is a [burdensome detail](https://www.lesswrong.com/posts/Yq6aA4M3JKWaQepPJ/burdensome-details) that requires a corresponding [amount of evidence](https://www.lesswrong.com/posts/nj8JKFoLSMEmD3RGp/how-much-evidence-does-it-take) just to [locate it in the space of possibilities](https://www.lesswrong.com/posts/X2AD2LgtKgkRNPj2a/privileging-the-hypothesis).
+
+[^koller-and-friedman-i]: Daphne Koller and Nir Friedman, _Probabilistic Graphical Models: Principles and Techniques_, §3.4.1, "Minimal I-Maps".
+
+[^koller-and-friedman-ii]: Daphne Koller and Nir Friedman, _Probabilistic Graphical Models: Principles and Techniques_, §18.3.5: "Understanding the Bayesian Score".
+
+I thought this shed some light on my recent frustrations. People are pretty observant about what other people are like. If prompted appropriately, they know how to anticipate the ways in which trans women are different from cis women. The part of them that talked just didn't see the problem with trying to represent this knowledge (about physiological males with male-typical interests and personalities whose female gender identities seem closely intertwined with their gynephilic sexuality) using a variable ordering that put "biological sex" closer to last than first. And I just didn't think that was what the causal graph looked like.
+
+-----
+
+In another post, I acknowledged my problematic tone:
+
+> I know the arrogance is off-putting! But the arrogance is a really fun part of the æsthetic that I'm really enjoying! Can I get away with it if I mark it as a form of performance art? Like, be really arrogant while exploring ideas, and then later go back and write up the sober serious non-arrogant version?
+
+Someone came to my defense: it was common to have mental blocks about criticizing trans ideology because of the fear that saying anything would hurt one's trans friends and make one an outcast. One way to overcome that block was to get _really angry_ and _visibly having an outburst_, because then people ascribe less agency and culpability to you; it would be clear that you'd cooped up these feelings for a long time because you do understand that they're taboo and unpopular.
+
+The person also said it was hard because it seemed like there were no moderate centrists on gender: you could either be on Team "if you _ever_ want to know what genitals someone has for _any reason_, you are an _evil transphobe_ who should _die_", or Team "trans women are disgusting blokes in dresses who are _invading_ my female spaces for _nefarious purposes_ and we should burn them all".
+
+I added that the worst part is that "trans women are disgusting blokes in dresses who are invading my female spaces for nefarious purposes" view was basically _correct_. It was _phrased_ in a really dismissive manner. But words don't matter! Only predictions matter!
+
+-----
+
+The thread on the "Totally Excellent Rationalist Friends" post continued. Someone who I'll call "Kevin" (whom I had never interacted with before or since; my post visibility settings were set to Public) said that the concept of modeling someone based on their gender seemed weird. Correlations with gender were weak enough to be irrelevant after talking with someone for half an hour.
+
+I replied, but this was circular, right?—that the concept of modeling someone based on their gender seemed weird. If gender didn't have any (probabilistic!) implications, why did getting gendered correctly matter so much to people?
+
+"Kevin" said that the distinction was between modeling someone as their gender, and addressing people in a way that respects their agency and identity, and it seemed reasonable to care much more about the second thing.
+
+I said I didn't know what the second thing meant. I liked the words "agency" and "identity", too! But the reason I liked the words, is because they were associated with agentic and identificatory things that people do in the world, that my brain could make predictions about. Regarding the predictive value of gender, human psychology was a very high-dimensional vector space! If you'd bought into an ideology that says everyone is equal and that sex differences must therefore be small-to-nonexistent, then you can choose to selectively ignore the dimensions along which sex differences are relatively large, and when you're locked into that worldview, it does indeed genuinely look to you like individual personality differences swamp sex differences! And when you're locked into that worldview, looking at the dimensions along which the differences are relatively large is genuinely painful! Once you notice this, maybe you can think of clever strategies to better serve the moral ideal that makes psychological-sex-differences denialism so appealing, while making use of the additional power you gain by letting yourself look at the whole configuration space!
+
+"Kevin" asked for some examples where gender-category membership was really important. He wasn't saying that sex differences didn't exist (for example, when doing statistical research), just that they were irrelevant in direct interpersonal situations.
+
+I replied, "Really important" was part of the map, not the territory! From the standpoint of someone who had never bought into the everyone-is-equal ideology in the first place, my desperate search for clever strategies to serve the androgyny-as-moral-ideal religion probably looked crazy and immoral. If my ancestors could see me, they'd probably be like, "Why are you making so many goddamned paperclips?! This wasn't supposed to be about paperclips!" And I was like, "But I want _moar paperclips._"
+
+After one more back-and-forth between me and "Kevin", "Noreen" expressed frustration with some apparent inconsistencies in my excited presentation. I saw what she was getting at, and expressed my sympathies, tagging Michael Vassar (who was then using "Arc" as a married name):
+
+> I'm sorry that I'm being confusing! I know I'm being confusing and it must be really frustrating to understand what I'm trying to say because I'm trying to explore this conceptspace that we don't already have standard language for! You probably want to slap me and say, "What the hell is wrong with you? Talk like a goddamned normal person!" But I forgot hoooooooow!
+>
+> **Michael Arc** is this how you feel all the time??
+>
+> help
+
+-----
+
+In another post, I collected links to Bailey, Lawrence, Vitale, and Brown's separate explanations of the two-type taxonomy:
+
+> The truthful and mean version: _The Man Who Would Be Queen_, Ch. 9
+> The truthful and nice version: "Becoming What We Love" [http://annelawrence.com/becoming_what_we_love.pdf](http://annelawrence.com/becoming_what_we_love.pdf)
+> The technically-not-lying version: [http://www.avitale.com/developmentalreview.htm](http://www.avitale.com/developmentalreview.htm)
+> The long version: [https://sillyolme.wordpress.com/](https://sillyolme.wordpress.com/)
+
+I got some nice emails from Michael Vassar. "I think that you are doing VERY good work right now!!!" he wrote. "The sort that shifts history! Only the personal is political" (Subject: "Talk like a normal person").
+
+I aptly summed up my mental state with a post that evening:
+
+> She had a delusional mental breakdown; you're a little bit manic; I'm in the Avatar state.[^avatar-state]
+
+[^avatar-state]: A reference to _Avatar: The Last Airbender_/_The Legend of Korra_, in which our hero can enter the ["Avatar state"](https://avatar.fandom.com/wiki/Avatar#Avatar_State) to become much more powerful—and also much more vulnerable (not being reincarnated if killed in the Avatar state).