stuaxo

10 ms·

1. 271.▲
   by stuaxo 7mo ago

   The linked chat log is unreadable because of the verbose AI slop. I say this as someone that has been building systems on LLMs and uses LLMs in dev work.
2. 272.▲
   by stuaxo 7mo ago

   Is it to curtail shoplifting or is the main benefit selling the data ? If they wanted to curtail shoplifting they could employ more staff.
3. 273.▲
   by stuaxo 7mo ago

   I like the analogy of data as oil: polluting when it gets out. I'd like to severely limit the amount of PII on the system.
4. 274.▲
   by stuaxo 7mo ago

   Does this work on the open source amdgpu drivers ? I've been a bit too busy to turn mine on for a while.
5. 275.▲
   by stuaxo 7mo ago

   These aresholes need to pay up.
6. 276.▲
   by stuaxo 7mo ago

   There is much work to be done in society that needs doing. Some of it can be new jobs, and some of it can be done by making the existing jobs have less hours freeing people up to do more meaningful things.
7. 277.▲
   by stuaxo 7mo ago

   Exactly - when UBI is tried people tend to be busy. The super rich are often idle and project that onto everyone else we should not take the idle richs word as gospel.
8. 278.▲
   by stuaxo 7mo ago

   This is a shame - while I never used it, it always looked like the way I would have wanted to remotely collaborate.
9. 279.▲
   by stuaxo 7mo ago

   Seems unlikely. I've seen teams at GDS + teams based on the GDS way of working in other UK gov departments solve some really knarly problems.
10. 280.▲
    by stuaxo 7mo ago

    I'm not sure I get the link between being against screenshots and working from home during COVID ?
11. 281.▲
    by stuaxo 7mo ago

    And of course the random choice of things displayed on the infinite scroll - deliberately emulating a slot machine effect.
12. 282.▲
    by stuaxo 7mo ago

    You could do this, GIMP is scriptable which is half the battle - you may not even need to train a model.
13. 283.▲
    by stuaxo 7mo ago

    This is good and we need more of this sort of thing. I don't want things to be silo'd just because I run them in the GUI Vs the terminal.
14. 284.▲
    by stuaxo 7mo ago

    Are Google one of the last holdouts ? I've been on cheap Android phones and just moved from Samsung to Motorola and both have headphones sockets.
15. 285.▲
    by stuaxo 8mo ago

    You can use it with Ollama rubbing a local model.
16. 286.▲
    by stuaxo 8mo ago

    Does anyone else find this kind of LLM written stuff tiring to read. Admittedly I only read the titles.
17. 287.▲
    by stuaxo 8mo ago

    This whole thing of private equity + companies getting massively inflated - only ends one way, it might not be this this buyer but one down the line, but there is something deeply wrong with the whole model, the one that starts with startup
18. 288.▲
    by stuaxo 8mo ago

    If I can use this with a local LLM it could be useful.
19. 289.▲
    by stuaxo 8mo ago

    We need LLMs that have a certificate of origin. For instance a GPL LLM trained only on GPL code where the source data is all known, and the output is all GPL. It could be done with a distributed effort.
20. 290.▲
    by stuaxo 8mo ago

    Thanks, I should add PyPy to the list of projects I send a little to ... PyPy should be better supported by organisations + not need individual contributions, but things are where they are I guess.
21. 291.▲
    by stuaxo 8mo ago

    It's just slightly spicey ketchup, pretty sweet. There are definitely better things out there.
22. 292.▲
    by stuaxo 8mo ago

    The website should have a prominent Donate section, maybe have some tiers of donation like the Ladybird browser does. I wanted to put a little £ towards the project but couldn't see a place to do it.
23. 293.▲
    by stuaxo 8mo ago

    It's always been about that many versions behind. There is more churn in those versions than you'd think.
24. 294.▲
    by stuaxo 8mo ago

    ah bloody hell, it's 10 past 9 on a saturday, my fault for reading the internet. All the people associated with Epstein should be in the Hague in front of a judge.
25. 295.▲
    by stuaxo 8mo ago

    Not everyone is as capable as using AI. I worked with someone that was trying to push code with all the tells that they didn't understand the code and that it had been through multiple LLM iterations, it burned a lot of the productivit
26. 296.▲
    by stuaxo 8mo ago

    A great thing you can do with LLMs: "in (language I'm familiar with) I use (some pattern or whatever) what's the equivalent in (other language)?" It's really great for doing bits and then get it to explain or you lo
27. 297.▲
    by stuaxo 8mo ago

    Plants grow faster (but not better) with more Co2 I wonder if this could be related to global obesity ?
28. 298.▲
    by stuaxo 8mo ago

    I've been thinking this for a while. Attribution is sorely needed, I believe there was a recent paper on this so maybe it will happen. In the meantime the only way to really sort this is to have models that have only been trained on
29. 299.▲
    by stuaxo 8mo ago

    Have worked on a help desk. Dialog boxes are close to this UI and nobody read them, come to think of it power users don't tend to either.
30. 300.▲
    by stuaxo 8mo ago

    I don't see how (with current LLMs that have been trained on mixed licensed data) you can use the LLM to rewrite to a less restrictive license. You could probably use it to output code that is GPL'd though.