turnsout

35 ms·

1. 661.▲
   by turnsout 2y ago

   Heads up: depends on an account at Cloudinary
2. 662.▲
   by turnsout 2y ago

   Ha, comments like this are why I love hn!
3. 663.▲
   by turnsout 2y ago

   Same—I just wish they'd either drop the "(Legacy)" and admit that they can never remove it, or add those same features to the Export dialog!
4. 664.▲
   by turnsout 2y ago

   In Photoshop there are at least three completely different dialog boxes [0] for saving an image as a JPEG, each with totally different UI widgets and functionality. They simply refuse to revise anything in the interface—they just keep addin
5. 665.▲
   by turnsout 2y ago

   There are definitely cases where you want the frequency domain power spectrum to be as close to perfect as possible, and this paper (GBN) is a step above the previous SoTA (Blue Noise Through Optimal Transport). However I agree about the vi
6. 666.▲
   by turnsout 2y ago

   Indeed, and they’re at 2.87T today… Built largely on differentiated high-margin products, which is not how I would describe OpenAI. I should clarify that I’m a fan of both companies, but the reality is that OpenAI’s business model depends o
7. 667.▲
   by turnsout 2y ago

   So awesome! I had a similar reaction to Shake—I first created an ANSI C command-line based compositor that read in and interpreted scripts. A year or so later I abandoned it in favor of a Mac app with a node-based GUI that compiled down to
8. 668.▲
   by turnsout 2y ago

   This looks really nice, but if you put information into a closed-source unmonetized side project, you better hope you never need to get it back out
9. 669.▲
   by turnsout 2y ago

   markdown
10. 670.▲
    by turnsout 3y ago

    Interesting blog post, but it just underscores how truly imaginary and disconnected most user personas are from any real people.
11. 671.▲
    by turnsout 3y ago

    Whoa, really? Amazon’s employee value prop just shot way up.
12. 672.▲
    by turnsout 3y ago

    In our hypothetical example, the compressed 4k data or the compressed 480p data? You would enhance the compressed 480p—that's what the example is. You would probably not enhance the 4K, because there's very little benefit to incre
13. 673.▲
    by turnsout 3y ago

    TVs already do this… and it's basically a bad thing
14. 674.▲
    by turnsout 3y ago

    I think you're missing the point of this paper—the precise thing it's showing is upscaling previously downscaled video with minimal perceptual differences from ground truth. So you could downscale, then compress as usual, and then
15. 675.▲
    by turnsout 3y ago

    So streaming services can save money on bandwidth
16. 676.▲
    by turnsout 3y ago

    I agree that it's not perfect, though it does appear to be SoTA. Eventually something like this will just be part of every video codec. You stream a 480p version and let the TV create the 4K detail.
17. 677.▲
    by turnsout 3y ago

    Any chance of this passing the Senate?
18. 678.▲
    by turnsout 3y ago

    Absolutely! I work in design/innovation consulting, and we use JTBD quite a bit. It's a nice way to interpret user needs and avoid the pitfall of taking user feedback too literally. I view JTBD as a flavor of Human-Centered Design
19. 679.▲
    by turnsout 3y ago

    This framework is attempting to make PMF seem so hopelessly complicated that you'll need their magic system (pricing unavailable, apply here). Trust me, you don't need a system. It's so much simpler than this. First, start wi
20. 680.▲
    by turnsout 3y ago

    Yes, and it's the same for the M3 MBA. If the M4 remains capped at 24, I'll still consider it, but it'll be less cut and dry. The speed boost would be nice, but literally the only time that I feel CPU-bound is when working wi
21. 681.▲
    by turnsout 3y ago

    Yeah, exactly this. I'm still on a nearly 4-year-old 16GB M1 MacBook Air. I'm coding SwiftUI, cutting 4K video, and using local 7B models. If I could spec an M4 MacBook Air up to 32GB of RAM, and it was better with local models, I
22. 682.▲
    by turnsout 3y ago

    So in answer to my original question, the tl;dr is that I can continue to pretend that IPv6 doesn’t exist. I’ve been hearing that I need to prepare for IPv6 for the past 20 years, and the benefits are always exceptionally vague or don’t app
23. 683.▲
    by turnsout 3y ago

    Wasn’t aware that T-Mobile is charging for access to IPv4… is there a link to more info about this surprising turn of events?
24. 684.▲
    by turnsout 3y ago

    I stopped using inheritance when Swift protocols and TypeScript interfaces came along. Although it sometimes means you need to do nasty things with generics, it beats the fragility of inheritance. With ObjC, I eventually came to hate updati
25. 685.▲
    by turnsout 3y ago

    Sounds like an interesting business model—I guess we’ll see if they survive.
26. 686.▲
    by turnsout 3y ago

    If you can’t access IPv4, I don’t think you’re going to make it as an ISP
27. 687.▲
    by turnsout 3y ago

    I run a server and have yet to adopt IPv6. My logic is: people can type in a URL and get to my site. Am I missing out on anything by continuing to pretend that IPv6 doesn't exist?
28. 688.▲
    by turnsout 3y ago

    Wait, is there a Suno API? I've used the site, but it's manual
29. 689.▲
    by turnsout 3y ago

    I don’t have an opinion, just wondering why they didn’t choose other another option such as Sentence Embeddings, OpenAI embeddings, etc.
30. 690.▲
    by turnsout 3y ago

    Curious about the choice of FAISS. It's a bit older now, and there are many more options for creating and selecting embeddings. Does FAISS still offer some advantages?