Skip Navigation

Posts
9
Comments
793
Joined
1 yr. ago

"If man chooses oblivion, he can go right on leaving his fate to his political leaders. If he chooses Utopia, he must initiate an enormous education program - immediately, if not sooner."-R B Fuller

  • The movie completely skipped over 2112. Total disservice to the young people who could use these tools from the Ancient Race to fight back against Earthly authoritarian governments.

  • As a young person, I loved A Wrinkle in Time and the entire set of sequels.

    And so I say a heartfelt FUCK YOU to Ava DuVernay, who crappified this masterpiece into vapid pap film that felt like it was written by Kellogg's marketing team in conjunction with the Ad Council of America.

    Apparently the film lost $130 million. Thank God there is little chance of a sequel.

  • I mean in MB/GB

  • How big is just the repo?

  • They can’t simply go back to correct their beliefs

    Backtracking is a common and integral part of the reasoning learned by what used to be called Large Reasoning Models but today is part of almost any modern model design, much like Mixture-of-Experts

  • tech bros turning the tech into a large scale game of blackjack

    Agreed 100%

    As a person who was bookish and deeply computer-focused from back when that meant only that you had no friends outside D&D -- before it also meant, "probably has a challenging career and makes good money".

    That "tech" now prefixes the socializing extrovert bros who didn't spend their entire adolescence learning assembly language and soldering alone in their room... is a complete, 180-degree U-turn subversion from the original "techy nerd" meaning

  • look like human reasoning

    You should really listen to the podcast. Not going to summarize it here, but it looks like human reasoning including semantic activation patterns, use of productions, and dozens of other things that cognitive psychologists have known about for decades but for which they have not been able to come up with a mechanism.

  • replace all the ‘reasoning’ steps from an LLMs output with “please wait” and it will still supply the same final answer.

    If you listened to the podcast (or were familiar with the literature) you'd know that's simply Not True.

    Sometimes, by listening to bona fide experts, you might learn new things instead of already knowing everything.

  • Would you believe, the evidence is rather more compelling than that. Similarities in non-obvious failure modes, for example.

  • Bookmarking this. I am low-key low-priority background-task setting up a Jellyfin with yt-dlp integration somehow and I love this simple brilliant idea, playlist with a cronjob. Same basic principle as Invidious.

  • Naomi Brockwell? Yeah, I'll wait for a vid from someone else who was there

  • Same things that happened to the Internet after pets.com died

  • Remember the good old 1980s? When things were so uncomplicated? I wish I could go back there again, and everything would be the same.

    But I've got a ticket to the moon...

  • model & tooling improvements, giving more weight to higher quality reports and MRs, and more respect for project rules and processes. A mix of soft and hard gates, CLAs, improved early automated bug report & patch reviews and other CI gates

    This is exactly what I see happening inside the handful of large software companies into which I have visibility.

    We have to deal with the fact that slop or not we all have 100x more security bugs than usual thanks to Mythos and friends, and some of them are real ones that will bit us in the ass if we don't fix like, now

  • Everything -- and I mean everything -- is going to depend on what SKILL.md file(s) they're using.

    And if they're using none, then this is not a best practice use of the technology.

  • Depends on how things play out. A combination of prices, regulations, and public sentiment leads a lot of petro-companies to offer energy-saving advice (even if it hardly offsets the rest of the negatives of the business, they do wind up doing this)

  • 10x token usage reduction is often achievable.

    Not doing stupid shit like asking an LLM to scan a million line file when you could grep, breaking skills files into progressively-discoverable chunks, running normal algorithms where a full LLM isn't necessary.

    And then using an open source model to really reduce the cost

  • My wife was a member of a Facebook group related to farming (as in growing food)

    IIRC a mod shared a post that had been auto-banned: one person complained about weeds; the triggering reply was

    "Hoe, do the work"