• 14 Posts
  • 366 Comments
Joined 3 years ago
cake
Cake day: July 13th, 2023

help-circle
  • This kinda gets philosophy-of-cognative-science-ey, but it kinda smells like “measuring a fish by its tree climbing ability”. Or maybe like goodharts law.

    There are many different kinds of smart, is the kind of smart they’re measuring the one that will be relevant?

    Again, I don’t want to carry too much water for AI, but I also don’t want to propagate arguments that also apply to other tools I like. Consider: externalizing memory. Without the tool you used to capture that memory, you’re completely unable to function. Dependency on it reduces your ability to remember things yourself. By many measures, it makes your dumber and less productive. But I wouldn’t personally consider taking notes to be detrimental.
    So I want to be very careful when we make claims that externalizing some aspect of our cognition makes us worse somehow.


  • And show me where I said people were using AI to read it.

    Because I was describing the class of reading material which is dry and unimaginative, to contrast your description of what reading does to peoples minds.

    I was describing two classes of data: the kind that fit what your describe, and therefore people don’t actually use AI on it, and dry technical data that doesn’t grant the benefits you describe. I did not say if people use AI on the technical material, only that they don’t use it with the enjoyable material.

    So no, I didn’t say that, and if you read it as carefully a you seem to expect everyone else but you too, then you’d have just shut up instead of doubling down embarrassingly.







  • As much as it pains me to carry water for AI, do you think that people are replacing things they liked to do with AI?

    Like… They were reading for pleasure, and then they had the AI read it for them instead?
    That they were talking to their friends they enjoy speaking with, and then they had the AI talk to their friends instead?

    Or are you saying that what you describe is the experience that you get from reading technical manuals for your work? For reading warning labels? From talking with coworkers you dislike?


  • I agree with you for common CLIs that the model will natural have in its training set, but when building your own CLI it seems like the self-documenting nature and progressive disclosure of MCP would have a huge advantage.
    With a cli you don’t have the prefix cost, but your agent also doesn’t know that the tool exists, what it’s for, how to use it, etc. once you give it that information, then you have the same issues with prefix cost, right? Changing tools does break the prefix cache, but changing tools is a thing you can do. You can’t change (model awareness of) CLI, and if you could then it’d also break the prefix cache, or you’d have tool instructions scattered throughout the context without structure.
    Dumping to file is a harness thing, not a MCP thing. A harness will do the same with large terminal outputs.
    Likewise setting up a new CLI can be complicated. I never see an agent try to install an mcp, but I also don’t typically see it try to install a new CLI.

    I don’t see how mcp is any more a fixed set of tools than CLI is. It’s just presented to the agent in a structured way, which is important for discovery and instruction following.

    But yeah ease of chaining is certainly useful wrt CLI. That’s a huge win for sure.






  • I do enjoy tiktok, although the longer it exists (and the longer it’s under American control) the more I feel the negative effects of “the algorithm” in that it’s not just building an echo chamber by accident but it’s intentionally showing me or not showing me things for their own gain.

    I want an algorithm, but I don’t want an Algorithm ™

    I hope piefed takes off because I like it. But still maybe it’s for the best that I use it as an excuse to disconnect completely from social media


  • Automatic personalized feed curation, imo, would make it better, based on tags added by the author, by the reader, and by automatic classification, which is then weighted against your previous interaction with posts with those tags and explicit user preferences regarding those tags.
    In other words, an option for an automatic personalized feed algorithm. But people here (and the devs) are against it because it’s a boogeyman.

    That said piefed at least lets multiple tags be added to a post where Lemmy doesn’t, so that’s the minimum bar for something like what I’m describing.

    But it’s a chicken and egg problem because the tags aren’t helpful for manual or automatic curation until more people actually use them accurately. So tldr I don’t have a solution.