XCSme

37 ms·

1. 871.▲
   by XCSme 2y ago

   I think that's a good point, and it does make me doubt my initial comment. One issue now, on a world without ads, is what incentive would also those actors have to provide you with that information? Search engine, review websites, YouT
2. 872.▲
   by XCSme 2y ago

   How else do you find out about new stuff, or recommendations, or things you never considered before? How do you find out about new movies, concerts, etc.? Most customers don't know themselves that a solution to their problem (which the
3. 873.▲
   by XCSme 2y ago

   Yes! Every small thing you do is a leaning experience or it refines your skills. Sometimes, if I have to do something boring, I make it challenging by either trying to automate it, or do it in a more efficient way.
4. 874.▲
   by XCSme 2y ago

   For me it helps if the goal is simply to "run". Without setting any end-goal. Just to start on it, and only do it ad long as I like doing it. It also helps if instead of "run for 5km" I think of the destination "run
5. 875.▲
   by XCSme 2y ago

   > I just know how much work is involved to get them going again, and that amount of work is daunting This is usually not true. Whenever I get back to something, I realise it was just 1h of work to implement that feature I've been pr
6. 876.▲
   by XCSme 2y ago

   I can also waste energy by leaving my light on, I don't think that stores any value...
7. 877.▲
   by XCSme 2y ago

   Both are pretty, though.
8. 878.▲
   by XCSme 2y ago

   How exactly does BTC store any more value than an NFT?
9. 879.▲
   by XCSme 2y ago

   That looks exactly like Plausible
10. 880.▲
    by XCSme 2y ago

    Congrats on building Bearlytics! I agree with you, it's weird most cloud analytics offer self-hosted versions, when that goes directly against their business model. In the best case, the self-hosted version can act as a lead magnet, at
11. 881.▲
    by XCSme 2y ago

    I am trying this with https://uxwizz.com but it's not easy. I do like this model, but there are some challenges: * With one-time the initial payment is higher ($299 one-time vs $29/mo), many people don't think too
12. 882.▲
    by XCSme 2y ago

    I am also building a piano-learning game. I remember using WebMIDI for the interface, TonalJS for chord/theory. From another comment, https://github.com/0xfe/vexflow looks cool, I could use it together with PIXI.j
13. 883.▲
    by XCSme 2y ago

    I did build my own: https://uxwizz.com It has a simple PHP/MySQL backend, so it's easy to query the data directly or integrate it with other data visualization tools and workflows. Recently I also added text-to-SQL and
14. 884.▲
    by XCSme 2y ago

    > own senses that matter does indeed exist What about people with schizofrenia? They can also use their senses and say something exists, when it actually doesn't.
15. 885.▲
    by XCSme 2y ago

    That could be the poor's man, on-demand, lan party. If not for the PCs, you would still need some devices to run the games.
16. 886.▲
    by XCSme 2y ago

    Random thought: would a gaming streaming service like GeForce Now achieve a similar result for a lan party? Assuming you have the network bandwidth, I am curious what the difference in input lag/quality would be, and if, when doing a b
17. 887.▲
    by XCSme 2y ago

    There is no evidence that anything exists...
18. 888.▲
    by XCSme 2y ago

    I built a self-hosted alternative to Hotjar/GA, that also has some unique features: https://www.uxwizz.com You can enable/disable specific tracking and privacy settings as you wish. To track engagement you can use the
19. 889.▲
    by XCSme 2y ago

    I had the same experience with LLM text-to-sql, 3.5 instruct felt a lot more robust than 4o
20. 890.▲
    by XCSme 2y ago

    Probably even at lvl 2 out of 9 it would lose all the games.
21. 891.▲
    by XCSme 2y ago

    I think it's impossible to be 100% legal. Many times, the user IP, which is considered PII, is stored in various servers/routers log that you have no access to...
22. 892.▲
    by XCSme 2y ago

    Serverless, as in using AWS servers for code execution and AWS S3 servers for storage? Or serverless, as in... no server?
23. 893.▲
    by XCSme 2y ago

    I am not sure it worked for me. After the calibration, all I saw was the same annotations around my eyes, no predicted gaze.
24. 894.▲
    by XCSme 2y ago

    Hey, This looks cool. I tried the demo, but I was expecting to somehow control a cursor with my eyes. All it does is it seems that it detects my eyes position? Or did I miss some UI to enable attention tracking on the page?
25. 895.▲
    by XCSme 2y ago

    I am building website analytics. I wouldn't call it "SaaS", because it's not a service, but a product: - UXWizz ( https://www.uxwizz.com ) - Self-hosted analytics with heatmaps, session recordings and A/B
26. 896.▲
    LLM Text-to-Chart Demo for MySQL(twitter.com)

    3 points by XCSme 2y ago|1 comments
27. 897.▲
    by XCSme 2y ago

    In this Twitter thread I also describe how it works, here's it copy-pasted: How it works: - The system prompt describes the database structure and in which format to respond - You ask a question - The AI generates a MySQL
28. 898.▲
    by XCSme 2y ago

    To complete the other comment: the MuscleMemory is updated through learning, so a more complete example would be: function muscleAction(Vec3d target, Vec3d environment, MuscleMemory memory) -> {actions: MuscleActivation[], result:
29. 899.▲
    by XCSme 2y ago

    I think it's actually a good name. The "memory" is stored as the parameters of a function. So, when you practice, you actually update this memory/parameters. This is why you can use the same "memory" and achiev
30. 900.▲
    by XCSme 2y ago

    > What are even the tasks where thinking makes humans worse? Not really related, but athletes perform A LOT worse when they are thinking about their movements/strategies/tactics. A top performing athlete does best when they are