Skip Navigation

InitialsDiceBearhttps://github.com/dicebear/dicebearhttps://creativecommons.org/publicdomain/zero/1.0/„Initials” (https://github.com/dicebear/dicebear) by „DiceBear”, licensed under „CC0 1.0” (https://creativecommons.org/publicdomain/zero/1.0/)S
Posts
3
Comments
302
Joined
1 yr. ago

  • I joined a Matrix group, and the UX frustrated one person so much, they just quit. Kinda surprised me some people care about UX that much. I guess I'm used to using software developed by hobbyists, lol.

  • Signal is ok. SimpleX theoretically has better privacy guarantees (metadata privacy, more decentralized). Matrix is ok for communities; I think it exposes a lot of metadata though (who you are talking to, not what you're talking about).

  • It's been a while since I messed with home automation, but ESPHome was amazing to program ESP microcontrollers (i.e. you most likely wouldn't have to write any code). You can use ESPHome devices with both Home Assistant and Openhab (using MQTT, IIRC). The last I checked, it was easier to program your own functionality in OpenHAB than Home Assistant.

  • For servers, I usually choose the distro with a version with the EOL scheduled furthest into the future. Usually that means Ubuntu (Server) LTS.

  • Low self discharge. Good for ultra low power devices like remote controls or lights only used on occasion where a rechargeable battery would self discharge faster than the rate of actual use.

  • I think it's still got a lot of gas. Investors still seem to be hyped about AI (e.g. the game stock downturn after Genie was announced), and that's all that matters. Tesla has held a nonsensical value pretty much for its entire existence. Furthermore a lot of the money going into the market is passive (401ks going straight into very top-heavy index funds), propping the largest companies up.

  • They're firing people living HCOL countries, and hiring in LCOL countries, using AI as cover.

  • Seems they are often ideological capitalists or fascists and get a kick out of shaping the world into their vision.

  • I have not personally worked on large projects using functional languages. I know they are popular in finance/trading.

  • Been a while since I've used Scala, but I remember Scala being much more focused on functional programming than Kotlin.

  • Some of it is a function of media. Media typically doesn't cover when non-white people get murdered. Some of it is the media reflecting the racism of its viewers. Some of it is being caught on video and leaving less room for doubt (and the media deciding to show the videos). There's also been protests and action being taken by normal citizens before these white people got killed (these white people were killed doing it, after all).

  • Every game I've tried works fine. Including resource hungry games like Cyberpunk 2077. It's my understanding that games are typically light on the CPU because they typically also try to target consoles which don't have very good CPUs. It is noticeably slower at some (highly parallelize-able) tasks, but is fine for any game I've tried. The CPU is probably roughly equivalent to the CPU in a Steam Deck.

  • I've got a PC with an i7-4770k, 32GB of RAM, and RTX 3090 that plays games just fine (and does runs local LLMs just fine too).

  • Dunno if that's true or not. Generally, much more compute is used in inference than training, since you only train once, then use that model for millions of queries or whatever. However, some of these AI companies may be training many models constantly to one-up each-other and pump their stock; dunno. The "thinking" model paradigm is also transferring a lot more compute to inference. IIRC OpenAI spent $300k of compute just for inference to complete a single benchmark a few months ago (and found that, like training, exponentially increasing amounts of compute are needed for small gains in performance).

  • It's possible someone would make it usable. A long time ago, I bought a laptop CPU that was soldered onto a board so it would go into a normal desktop socket. Guessing there was a glut of laptop CPUs at the time.

  • In my area, some people put small solar nodes on top of high buildings (office, university, and apartment). The node on my roof can directly communicate with one of these nodes ~20km away. Pretty crazy tor something that can run indefinitely on a 18650 battery and small solar panel. I've heard some people just place "guerilla nodes" to extend coverage.

  • Yeah, I'm guessing it's so if you "hide" the network, it will still connect to it. Anyone can scan these advertisements, then go to wigle.net and likely get a good idea of where you live/work.

  • Modern phones rotate random MAC addresses. For WiFi, capturing SSID probes can be enough to track somebody though (some phones also have some mitigation for that too, like not probing for an SSID after it hasn't been seen for some amount of time). Even when turned off, many phones, including iPhones, turn into BLE beacons similar to AirTags, which can be used to track you.

  • I once saw an old lecture where the guy working on Yahoo spam filters noticed that spammers would create accounts to mark their own spam messages as not spam (in an attempt to trick the spam filters; I guess a kind of a Sybil attack), and because the way the SPAM filtering models were created and used, it made the SPAM filtering more effective. It's possible that wider variety of "poisoned" data can actually help improve models.