oh yeah i'm done with undergrad

it feels a little anticlimactic


this looks pretty and the content seems good but i'm somehow dubious that the colors and the way everything animates onto the screen reflect its own principles...

Making my software decisions based on dragons:

clang: ✔️ has a dragon
gcc: ❌ no dragon
kde plasma: ✔️ has a dragon
gnome: ❌ no dragon
wireguard: ✔️ has a dragon
openvpn: ❌ no dragon

Trans-inclusive [web / software / information] design: alistapart.com/article/trans-i

[Apologies to whoever I saw this from - I don't remember who to thank]

Oh hey, I should post this here too! I made a CSS-only async chat by using background-image requests to send data and long-running http requests to return it.


the bottom line is though that public things on the internet are still really quite public, and this service is a good reminder of that

i'm bad at threading so in case this gets separated from the original link, the service being discussed is here: humanpredictions.io/

cw: debatable privacy-violating tech (maybe i should have cw'ed this thread somewhere along the way)

i think the big ways this violates your privacy are aggregation (lots of public information about you is gathered into one place) and secondary use (data you made public is being used for a purpose you didn't intend or consent to, and weren't even notified about). these are stolen from A Taxonomy of Privacy §B.1 and §B.4 law.upenn.edu/journals/lawrevi because i'm a nerd who took one policy class once. even then the negative consequences still aren't that obvious but the paper has a solid discussion of both.

(there are the usual concerns about machine learning bias, but it's hard to argue that those would be a *privacy* violation; the effects of a biased model are probably not as bad as they might be other places (::cough:: algorithmic sentencing), and the site at least pays lip service to diversity so they might even do this part right; i don't have any evidence other than that the industry status quo is pretty bad afaik)

what's so interesting about LB is that, although i share the instinctive reaction that my privacy is violated by this thing, it's not all that obvious what the privacy violations are if you think about it — all they're doing is gathering public information on you and running machine learning on it, automating and greatly speeding up a process that i actually wish good recruiters followed (instead of mass-spamming candidates without looking into them at all)!

If you've ever mentioned working in the tech industry online, there's a good chance you're being indexed by this company, "human predictions". humanpredictions.io

Just a head's up. Massive privacy violations abound.

so i end up with a take that's not I-V-vi-IV and if i like the new take more then great! but if i like the old take where everything is root position and no counterpoint rules are followed, it's still there

i probably used to be worried that taking too many music theory classes would overly constrain any music i was trying to compose to follow rigid rules, like "parallel fifths? weird chromatic notes? time to throw this composition away" but instead it's "parallel fifths? weird chromatic notes? ok can we like use the secondary dominant of the neapolitan or whatever to make it work™"

for this lab in our cybersecurity class we get to do web exploitation, which is fun, except the grading scripts use some ancient version of phantomjs so we have to write javascript like it's 2009 or something and uggggggh

more facebook Show more

Show more

cybrespace: the social hub of the information superhighway

jack in to the mastodon fediverse today and surf the dataflow through our cybrepunk, slightly glitchy web portal