• 0 Posts
  • 240 Comments
Joined 2 years ago
cake
Cake day: June 7th, 2023

help-circle





  • Here’s some uses:

    • skin cancer diagnoses with llms has a high success rate with a low cost. This is something that was starting to exist with older ai models, but llms do improve the success rate. source
    • VLC recently unveiled a new feature of using ai to generate subtitles, i haven’t used it but if it delivers then it’s pretty nice
    • for code generation, I agree it’s more harmful than useful for generating full programs or functions, but i find it quite useful as a predictive text generator, it saves a few keystrokes. Not a game changer but nice. It’s also pretty useful at generating test data so long as it’s hard to create but easy (for a human) to validate.


  • In a video someone discussed the average us household income. Someone commented that that number was actually inflated and it would be better to use median. I found the article the OP was referencing and pointed out that it was in fact the median and pointed out a median is a type of average. They argued for far far too long that average exclusively refers to mean, that median “isn’t even an expected value” and that they were right and I was wrong because they are an engineer who works with this all day long. I ended up getting ganged up by several different accounts, I eventually screenshotted the Wikipedia page for average and got them to all delete their posts.


  • A lot of my picks are already mentioned so I’ll pick an odd one:

    Air crash investigator (called Mayday in NA). It’s dramatizations of the reports from air crashes, organized like a murder mystery. Surprisingly compelling.

    I’ve never been nervous about flying but this show really underlined how safe flying is, it’s actually kinda crazy how thorough the reports are and how often they lead to rule changes. I wish the same institutional dedication to safety was practised in other industries (especially cars).

    Episodes that take place in the 80s have you face palming at how stupid the mistakes are, more modern episodes are almost always a combination of many many different small low chance events and minor mistakes from the pilot piling up. I usually skip the terrorist episodes though.






  • My last apartment (~2020) was at less than 30% occupancy and I was the only person living on my floor. Yay housing crisis.

    My apartment before that was in an older Orthodox Jewish Ukrainian community that was very hostile to outsiders, and particularly my interracial relationship. The head door guard was especially racist. There was one other guy with a dog who was super nice and friendly we’d talk a bunch especially while walking. He was feeling as unwelcome as we were and they drove him out a few months before we got driven out.

    Now, as a homeowner I’ve exchanged misdelivered packages with my neighbours and done trick or treating. I don’t really know their names but i recognize their faces. There was one guy that was a creep to the femmes of our building but he’s moved out thankfully, but it kinda spoilt the first impressions a bit.


  • saigot@lemmy.catoAsklemmy@lemmy.mlWhat do you donate to?
    link
    fedilink
    arrow-up
    3
    ·
    edit-2
    2 months ago

    I have this fun strategy, I’ll save my charity money until there is a public call for action, then I dump a larger sum all at once.

    For instance my work during covid set up a (money only) drive for a food bank that corporate would match. My bosses donated 200 a piece, until they noticed that I donated 1000, then they all found it in their hearts to donate 1000 too. Turned that 1000 bucks into about 8000.

    I do have some reoccurring and misc donations.




  • Does worst mean:

    • least able to achieve their stated agenda, ie worst at their job. (Trump)
    • worst vision for America, ie most evil (Reagan)
    • worst overall impact to America, ie one you’d kill with a time machine (Bush Jr, but Trump might catch up in term 2)
    • Worst for the world, ie the one I’d kill with a time machine (Washington)

    Although I’m not American and don’t know your history that well.


  • It would be interesting to break down exactly why they have higher rates. It could be:

    • over reliance on it’s automated systems (but FSD is only enabled on a small percentage of cars)
    • safety design flaws (but it has a high safety rating by third parties)
    • Unconventional UX decisions (steering yoke, giant displays with no physical buttons etc)
    • novel behaviour of electric vehicles such as the fact users less familar with it’s acceleration behaviour and lack of gear shifts, the increased weight or one pedal driving. (but why aren’t other EVs up there)
    • The politicized nature of Tesla attracting more risky customers. (but is that really worse than the type of people buying ferraris and other performance cars?)

    The article points to the first point, and that certainly seems plausible, but they don’t really provide any evidence to support that.