• 4 Posts
  • 20 Comments
Joined 1 year ago
cake
Cake day: June 11th, 2023

help-circle







  • bh11235@infosec.pubtoProgrammer Humor@lemmy.mlNames
    link
    fedilink
    arrow-up
    58
    ·
    10 months ago

    Now of course one could make some damning argument about the state of the tech industry in practice, resulting in one of those bell curve memes with “using SQLalchemy is a sin” on both far sides and “noooo it’s just a name it’s fine there’s no fraud involved” in the middle




  • Reading this comment section is so strange. Skepticism about generative AI seems to have become some kind of professional sport on the internet.

    Consensus in our group is that generative AI is a great tool. Maybe not perfect, but the comparison to the metaverse is absurd: no one asked for the metaverse or needed it for anything, as opposed to several cases where GPT has literally bailed us out of a difficult situation. e.g. some proof of concept needed to be written in a programming language that no one in the group had enough experience with. With no GPT, this could have easily cost someone a week. With GPT assistance – proof of concept ready in less than a day.

    Generative AI does suffer from a host of problems. Hallucinations, jailbreaks, injections, reality 101 failures, believe me I’ve encountered all these intimately as I’ve had to utilize GPT for some of my day job tasks, often against its own better judgment and despite its own woefully lacking capacity to deal with the task. What I think is interesting is a candid discussion: why do these issues persist? What have we tried? What techniques can we try next? Are these issues intractable in some profound sense, and constitute a hard ceiling for where generative AI can go? Is there an “impossibility theorem for putting AI on autopilot”? Or are these limitations just artifacts we can engineer away and route around?

    It seems like instead of having this discussion, it’s become in vogue to wave around the issues triumphantly and implicitly declare the field successfully dunked on, and the discussion over. That’s, to be blunt, reductive. Smartphones had issues, the early internet had issues. Sure, “they also laughed at Bozo the clown” and all that, but without a serious discussion of the landscape right now, of how far away we are from mitigating these issues and why, a lot of this “ha ha suck it AI” discourse strikes me as deeply performative. Like, suppose a year from now OpenAI solves hallucinations. The issue is just gone. Do all the cool kids who sneered at the invented legal precedents, crafted their image as knowing better than the OpenAI dweebs, elegantly implied how hallucinations are a cornerstone in how the entire field is a stupid useless dead end – do they lose any face? I think they don’t. I think this is why this sneering has become such a lucrative online professional sport.





  • Two govt spooks are hunting a dangerous fugitive who is also a humanities graduate. He escapes into a sprawling maze of tunnels. “It’s hopeless,” one of the spooks says. But the other simply says, “Watch.” then proclaims loudly, “studying linear algebra is important because of its use in stochastic processes and image manipulation.” Before he finishes the sentence, the fugitive emerges back out the tunnel and shouts, “but what’s even more important --” and is immediately knocked unconscious and taken for questioning


  • To me vim’s main strengths are

    • It delivers the same OK-ish experience no matter what file type or language I’m dealing with. Yeah it’ll never be as good as a dedicated Python IDE for writing Python, but I’d rather know vim than 5 different IDEs for Python, YAML, Dockerfiles, Rust, Latex, whatever I need to deal with today.
    • It just edits files and doesn’t hide internal state, intermediate files, etc to make my life ‘simpler’ (notepad is the same, so I guess this is more of a strength vs IDEs). When an IDE fails to align all of its internal moving parts just right to compile a project I know I’m in for an hour of figuring out which checkbox needs to be unticked in what sub-sub-sub-sub-submenu, I like it much better to have a “flat” experience of invoking a command line and getting an error message directly from the tool I am invoking.
    • 20dd to delete 20 lines, that’s very neat.




  • bh11235@infosec.pubtoProgrammer Humor@lemmy.mlThe Ol' Two Year Shuffle
    link
    fedilink
    English
    arrow-up
    3
    arrow-down
    1
    ·
    1 year ago

    Your comment makes me personally angry. After considering several ways of explaining why it makes me personally angry, I’ve settled on telling you about this person I worked with a long, long time ago – let’s call her Anya.

    Anya is the model employee per your value system. A risk-taker, a people person, full of gumption and ambition to get ahead. All her life she’s used these skills to project the image of someone knowledgeable, dependable, who is on top of things. So far so good. Unfortunately at one point she realized that she is much more capable at this, by many orders of magnitude, than at actually becoming knowledgeable or on top of anything. To her, learning and understanding the details of a system is a hassle; so why go through the hassle when it’s so much easier to just navigate every conversation about the system, and appear knowledgeable? Why make the effort to improve at the actual job when it’s so easy to judo-deflect every negative incident as actually a positive, or someone else’s fault? She has a gift; being a human, and not a saint, she is compelled to make use of that gift.

    Anya is not a bad person. She just takes the path of least resistance – let he among us who is without that particular sin cast the first stone. Maybe she even has the natural capacity to match and exceed the skill level of her colleagues; it’s just that she never will, because what’s in it for her. One way or the other, navigating any problem with Anya on your team is an ordeal. Every step forward involves defusing some part of whatever elaborate web of obfuscation she’d weaved to maintain her image. To be blunt, the thought of people like her being actually in charge of some truly technical system, something that can’t be reasoned with or bullshitted, that will cause damage and cost lives if not handled properly – that thought puts the fear of god in my heart.

    So in conclusion, being familiar with Anya, I don’t buy your Randian Dr. House fantasy, this dichotomy of skillful extroverted pushy go-getters who know the job and don’t take no bullshit and ‘tell it like it is’ vs risk averse socially inept introvert moochers . Given the choice between working with Anya on a project or instead working with Anya’s risk averse socially inept introvert colleague who is actually physically capable of articulating the words “I don’t know, I’ll go check” – give me the colleague any day of the week.





  • These are some games I’ve played that do not require reacting in real time.

    Cosmic Star Heroine – described as “remember your favorite 1993 JRPG? this plays like your fond memory, as opposed to how that game actually played”. Ultra-polished gameplay, blunt, thin plot, no dialogue trees and no sob stories. OK, one sob story.

    The Witness – also known as “fuck yeah science, the puzzle game”. Unleash your inner Feynman and make sense of the unspoken 'puzzle rules` by reasoning about examples.

    Superliminal – The inverse puzzle game to the above, where you can only get ahead by aggressively keeping your inner Feynman in check, and thinking laterally. Has similar snarky, meta vibes to Portal and Stanley parable.

    Inscryption – people say this is a “deckbuilder roguelike” so the immediate response is “oh you mean like Slay the Spire?” and the answer is no. While there is some deckbuilding and some roguelike-ing, this game stays away from the pure mechanical polish of those genres, and you could even say the entire game is a commentary of the sterility of pure mechanical polish in game design.

    Wasteland 3 – an irreverent turn-based RPG set in a post-apocalyptic future, and honestly a breath of fresh air in the landscape of generic high fantasy RPGs where you spend 30% of the game running an errand, taking the wrong turn, stumbling on a magic fire-breathing weasel 15 levels higher than you, getting wasted and loading your save.

    9 People, 9 Hours, 9 Doors – a classic of the visual novel genre. A bunch of strangers trying to survive a sadistic game – so, inspired by Jigsaw, and in turn a part of the genre that ended up inspiring Squid Game, if that helps set your expectations. Has some of the infamous excesses of Japanese media but concludes in what is IMO a contender for the best twist ending of all time. For most platforms this game can be found bundled with its sequel, under the name “The Nonary Games”.