zackmdavis

I want to be very clear on the messaging that is coming from OpenAI, and the mendacious nature of it. This is an example of who they really are, and I want to make sure everything [sic] sees it for what it is. Although there is a lot we don’t know about the contract they signed with DoW [shorthand for the Department of Defense] (and that maybe they don’t even know as well — it could be highly unclear), we do know the following:

Sam [Altman]’s description and the DoW description give the strong impression (although we would have to see the actual contract to be certain) that how their contract works is that the model is made available without any legal restrictions (“all lawful use”) but that there is a “safety layer”, which I think amounts to model refusals, that prevents the model from completing certain tasks or engaging in certain applications.

“Safety layer” could also mean something that partners such as Palantir [Anthropic’s business partner for serving U.S. agency customers] tried to offer us during

zackmdavis

Response to Question 1.

(366 words)

Before European colonization, the only humans living in (the land that contemporary American Society calls) San Francisco and the East Bay were various native tribes now collectively known as the Ohlone. (The name "Costanoan", from the Spanish costeño, coast-dweller, is also sometimes used, but Ohlone—which is at least of indigenous etymology—tends to be preferred.) They subsisted in the economic mode of hunter-gatherers: living in thatch villages and killing local wildlife and plants to survive (Gray-Kanatiiosh 8–11). Acorn meal was a specialty.

The Ohlone population in 1770 was estimated at about 10,000 (Gray-Kanatiiosh 24). As of the 2020 U.S. Census (which uses the term Costanoan), there are only 4,000 left (U.S. Census Bureau); other sources claim only 550 left (Gray-Kanatiiosh 28). The decrease of "only" a factor of 2.5 to 18 goes to underscore the difference in scale between pre-argicultural and industrialized Societies: apparently, it's not (just) that

zackmdavis

from itertools import permutations
from collections import defaultdict


class Permutation:
    def __init__(self, data):
        self.data = data

    def __hash__(self):
        return hash(tuple(self.data))

zackmdavis

You've likely got a master's thesis in here! There's a lot of content, and you're clearly passionate about exploring the socioacademic frameworks and lived experiences of what we now recognize as autogynephilia. Spoiler alert, because autogynephilia wasn't coined until 1989, retroactively applying it to a subject who literally could not have identified in that way is inaccurate. That's not terrible news, though, because by recognizing the label as out-of-time-and-place you have a really great opportunity to become more precise. In other words, if we remove the word/label of autogynephilia (not forever, just for now), what terms and cultural, contextual specifics can take its place? If you can start to name those things, you stay within the appropriate era as well as uphold VP and her words as the authority and primary source of herself. Autogynephilia, of course, can always hold as a lens (and-or work as a comparison) through which to view any number of things/people--but your arguments are stronger if the

zackmdavis

Subj: Putnam prep session #2 for eternal mathematical glory ... and donuts, 2 p.m. Fri 4 October

"Hey, Goofusia," said Gallantina. "Did you see this post on the math_majors list? Someone's trying to organize a team for the Putnam competition—here, at SFSU! There's going to be a prep session in Thornton 935 on Friday at 2 p.m. The organizer sounds really desperate—there should be free donuts. Want to come?"

Fraternal twins, the sisters looked so much alike that strangers who didn't know them often asked if they were identical. People who knew them for any length of time never asked.

Goofusia grimaced. "Oh, God, is that that super-hard math competition that guys from MIT win every year, where the median score is zero?"

"Actually, someone not from MIT won as recently as 2018, and last year the median score was nine. But yes."

zackmdavis

Before

Theme: autogynephilic MtF transsexualism

Accompanying narrator: Virginia Prince

Summary: In this piece, Virginia Prince summarizes her origin story, but there's a lot to Prince's story that isn't being told here that can be found in other sources, notably Zagria's biography of Prince in A Gender Variance Who's Who. It's anachronistic that the Stonewall Reader introductory blurb mentions Prince's "journey with her gender identity". The text itself says TVism (transvestism). (I hope I will not be accused of anachronism if I identify the underlying cause as autogynephilia.) Prince's Society for the Second Self was specifically for straight crossdressers and Prince advocated against sex reassignment surgery.

During

zackmdavis

from math import log2
import prediction_markets


def prediction_market_sort(my_list):
    n = len(my_list)
    op_count = 0
    op_budget = n * log2(n)
    is_sorted_market = prediction_markets.create(
        "Is the list sorted?", dynamic_data=my_list

zackmdavis

math_majors@lists.sfsu.edu

Subj: Putnam prep session #2 for eternal mathematical glory ... and donuts, 2 p.m. Fri 4 October

"Hey, Goofusia," said Gallantina. "Did you see this post on the math_majors list? Someone's trying to organize a team for the Putnam competition—here, at SFSU! There's going to be a prep session in Thornton 935 on Friday at 2 p.m. The organizer sounds really desperate—there should be free donuts. Want to come?"

Fraternal twins, the sisters looked so much alike that strangers who didn't know them often asked if they were identical. People who knew them for any length of time never asked.

Goofusia grimaced. "Oh, God, is that that super-hard math competition that guys from MIT win every year, where the median score is zero?"

zackmdavis

in search of gender studies (as contrasted to gender activism)

Online discussion forums for school classes are always depressing, because the ethos of an online forum is antithetical to the ethos of a school class. Anyone who has something real to say can already say it in a real forum on the open internet, and schoolstudents who are being coerced by Society into pursuing a credential (which is called "education") don't have anything real to say in their capacity as schoolstudents: in either case, why post?

Still, the sight of a forum that's empty (or empty of non-classwork, which might be worse) tugs at my heartstrings for the cruel mockery it makes of real forums—and for all that I might remind myself not to be deceived by surface trappings into confusing it for the real thing, I can't shake the illusion that something real might happen here. Since I'm here—since [I have for my own complicated reasons submitted to coercion by Society into finishing up my math degree](http://zackmdavis.net/blog/2024/0

zackmdavis

Differentiability implies continuity. Sorry, that's not very specific. What I really mean to say is, ∀ε ∃δ ∀z ||z − x|| < δ → ||f(z) − f(x) − Df(x)[z − x]|| ≤ ε ||z − x|| implies ∀ε ∃δ ∀z ||z − x|| < δ → ||f(z) − f(x)|| < ε.

The antecedent implies ||f(z) − f(x)|| ≤ ||Df(x)[z − x]|| + ε||z − x|| by the reverse triangle inequality, which is ≤ ||Df(x)||·||z − x|| + ε||z − x|| = (||Df(x)|| + ε)||z − x|| from the property of the operator norm.

So take your arbitrary ε, and I implore you to consider δ := ε / (ε + ||Df(x)|| + 1), so that we have (||Df(x)|| + ε)·(ε / (ε + ||Df(x)|| + 1)) < ε. But that's what I've been trying to tell you this whole time!!