@iliana just now hitting me that for some large group of people now fez is probably... like... a treasured childhood experience and not just a thing that they did one weekend a while ago
there are apparently bugs in one or more of (a) numpy (b) openblas (c) mkl (d) anaconda that make it so some of my students get errors from calculating cosine distance between word vectors (I think the norm() function is returning negative numbers for no good reason sometimes) and the process of figuring out the problem is very annoying. like it's a dot product and two norms? how can you mess up those two extremely simple functions
open source licenses, as texts that advocate for their own propagation ("The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software"), is a kind of chain letter and maybe by extension could be considered a kind of himmelsbrief, a "letter from heaven," granting protection to those who possess them (and, per the upthread link, "divine punishment for disbelief of their claims")
uhhhhhh is there a guide on how to use lists on mastodon (in particular the web app)? it took me like ten minutes of poking around to figure out how to even make one (there isn't just an option on a user profile to "add to list," you have to go through the "getting started" tab thing). mostly I just want to know (a) if my lists and their content are visible to the public and (b) if other users are notified and/or can find out that I added them to a list
I teach a class at NYU called Reading and Writing Electronic Text that is half introduction to Python, half introduction to computational poetry. the syllabus is here: http://rwet.decontextualize.com/ and the tutorial/lecture notes are here: https://github.com/aparrish/rwet I've taught the class many times before, but this year I'm revising all the notes and porting everything to Python 3/Jupyter Notebook. I've also made entirely new tutorials on dictionaries, RNNs and keyword extraction. just wanted to share!
my task for today was to write a description for an interactive typography course and instead i spent the better part of the afternoon reading (or trying to read) a belle époque french experimental phonology book whoops
@emsenn and in any case "recommending we revert to the Internet as it was before Facebook" is a strawman—I don't know of anyone who is advocating for that? in any case, before facebook there were plenty of easy-to-use centralized platforms that didn't share all of facebook's problems (livejournal, friendster, myspace). imo recommending people use something other than facebook is more like advocating for public transit instead of cars than it is like advocating for restricting suffrage.
@emsenn facebook is a system for taking advantage of the technically illiterate, not empowering them. using facebook is also *tremendously* technically sophisticated—understanding it and gaining mastery over it is not easy (there are whole books and classes for learning how to use it)—and I would argue that "accessibility" per se is not among facebook's goals for their product.
@Gargron imo the entire star trek franchise exists only as background to the most important narrative of the series, which is data having a cat that he loves
among the miscellaneous hazards of scientific research in the belle époque (from Alvarado, Carlos S. “Psychic Phenomena and the Brain Hemispheres: Some Nineteenth-Century Publications.” Journal of Scientific Exploration, vol. 30, no. 4, Dec. 2016, pp. 559–85.)
this paper in which b.f. skinner argues that gertrude stein wrote "tender buttons" in a trance state via automatic writing (because of a paper she wrote when she was a student of william james) is such a perfect confluence of historical figures and methodologies that it's hard to believe it even exists, it's like someone's historical fiction come true