• 0 Posts
  • 33 Comments
Joined 1 year ago
cake
Cake day: September 20th, 2023

help-circle

  • aleq@lemmy.worldtoTechnology@lemmy.world*Permanently Deleted*
    link
    fedilink
    English
    arrow-up
    11
    arrow-down
    22
    ·
    21 days ago

    Seems this is a common thing. Found this and also Claude (the LLM) tells me it goes very far back and kinda got very big around Reagan’s time. Allegedly Obama had some restrictions on his first term, but these were lifted on second term, but I haven’t bothered to verify that.

    However, the answer is yes. Just that it seems to be nothing new.



  • They didn’t make a video about it because they thought it was a problem for creators, not a problem for consumers.

    Which is true. Influencers are great at making their thing your thing, because that’s kind of their job, and we’ve seen it many times before. Just look at all the outrage about the YouTube algorithm and such, it doesn’t matter to anyone except influencers but somehow it’s made to be everybody’s business.

    This feels very similar. Scummy business practice, good on them for suing, but to the rest of us it should only be a curiosity.


  • I think you’re spot on with LLMs being mostly trained on these kinds of tasks. Can’t say I’m an expert in how to build a training set, but I imagine it’s quite easy to do with these kinds of problems because it’s easy to classify a solution as correct or incorrect. This is in contrast to larger problems which are less guided by algorithmic efficiency and more by sound design/architecture.

    Still, I think it’s quite impressive. You don’t have to go very far back in time to have top of the line LLMs unable to solve these kinds of problems.

    Also there is no big consequence if they don’t and it’s probably possible to bruteforce (which is how many programming tasks have been solved).

    Usually with AoC part 1 is brute-forceable, but part 2 is not. Very often part 1 is to find the 100th number, and part 2 is to find the 1 000 000 000 000th number or something. Last year, out of curiosity, I had a brute-force solution for one problem that successfully completed on ~90% of the input. Solution was multi-threaded and running on a 16 core CPU for about 20 days before I gave up. But the LLMs this year (not sure if this was a problem last year) are in the top list of fastest users to solve the problems.


  • I hardly see it changed to be honest. I work in the field too and I can imagine LLMs being good at producing decent boilerplate straight out of documentation, but nothing more complex than that.

    I think one of the top lists on advent of code this year is a cheater that fully automated the solutions using LLMs. Not sure which LLM though, I use LLMs quite a bit and ChatGPT 4o frequently tells me nonsense like “perhaps subtracting by zero is affecting your results” (issues I thought were already gone in GPT 4, but I guess not, Sonnet 3.5 does a bit better in this regard).





  • aleq@lemmy.worldtoLinux@lemmy.mlLinux middle ground?
    link
    fedilink
    arrow-up
    15
    arrow-down
    7
    ·
    4 months ago

    For private use? Hot take, but Arch. It’s easy to maintain and not easy to break at all. I think I spend zero time on maintenance other than running package updates. I only reinstall when I get a new computer.

    (I say for private use only because you’ll be getting weird looks from people if you use arch on a server in a professional setting, and it might break if you try to update it after five years of not doing it since there aren’t any “releases” to group big changes - in practice I run arch on my home server too with no issues)






  • I think Twitter is going down, may or may not go bankrupt but I think it will lose relevance. Wonder if it will be replaced. Lots of people (myself included) kinda assume that bluesky, mastodon or some other twitter-like service will take over. But Twitter is not really necessary, so I don’t think it’s a given that something will take its place.

    As a time sink, more multimedia-oriented platforms like Reddit/Lemmy, Instagram, Tiktok or Youtube, seem more attractive.



  • Right before the Mozilla buyout, Fakespot added a clause to their TOS giving them the right to give user data to Mozilla.

    There’s one possible interpretation of that, which would be my guess, that this was somehow necessary as part of the purchase. Before purchasing a company the company being purchased has to show the buyer what their assets are and give them a fair and accurate representation of what the company is. It’s possible that this clause was necessary in order to enable this.




  • I’m on arch, which I consider one of the larger distros, where most such configuration is very simple. Not sure what rolling mesa is. I probably wouldn’t recommend Ubuntu to anyone who is against using Snap, but there are many distros to choose from if you want KDE as well? It’s more a question of why people would go for Hannah Montana Linux (figuratively speaking, some very niche distro).

    But to respond to your core point, sure. If you do have a lot of customization needs for whatever reason, then by all means. (I still don’t get it)