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/)FR
frezik @midwest.social
Posts 5
Comments 3.3K
Who would win?
  • It also depends on the parameters of the shot. DS2 was shooting at partial charge to take out ships so there's less recharge time between shots. If the DS commander did the same to the Borg, they might blow a hole in the Cube without destroying it, and the Borg don't give a fuck.

    They beam in some drones and it's over.

  • Telegram says it has 'about 30 engineers'; security experts say that's a red flag
  • Headline is terrible. The big red flags are that they don't do end-to-end encryption by default, the servers are in Dubai, and use a proprietary algorithm.

    Last part should be clarified further. They didn't reinvent AES or anything. It's more like a protocol that puts together existing algorithms. It means they can use transport layers without TLS or anything else that wraps your messages in crypto otherwise.

    https://core.telegram.org/mtproto

    I'd still say this is a red flag. How you wrap encryption around your messages has several pits you can fall into. It's not as bad as reinventing AES, though.

  • On the Internet, what is a dead giveaway that someone is actually a kid?
  • TLS already has algorithms hardened against QC. The effects of QC against encryption are greatly exaggerated, anyway. The number of qubits that would be needed to break encryption may be too large to ever be feasible.

    Get IPv6 going and stuff like SNI becomes unnecessary.

  • Progress happens with every death
  • I'd like to take this opportunity to highlight a recent discovery that I think should be shouted from every major news outlet. The implications are big, but they're rather technical and non obvious.

    https://www.youtube.com/watch?v=B1PbNTYU0GQ

    In short, it turns out water evaporates much faster from to light than heat. Green light with a certain polarization hitting the water surface at a 45 degree angle seems to do best. From the research slides, the effects of polarization and angle might be small. That means green LEDs (which are cheap and very efficient, but wouldn't be polarized on their own) can evaporate lots of water. Something like 4 times the amount we would get from using the same amount of energy to heat it up. This is being called the photomolecular effect.

    This fills in a big gap in our climate models. There have been measurements done on clouds that show water was evaporating much faster than theory would predict. I'm not clear on if it would make the results more pessimistic or not. My guess is that more clouds in the model increase the albedo of the Earth, thus reflecting more light back into space, and the resulting temperature should be lower. But I'll hold off on strong opinions until the models get updated.

    The other big thing is desalination. Most desalination plants don't use thermal evaporation because it's too energy intensive. They use reverse osmosis. The photomolecular effect brings up the possibility of an even more efficient solution to drinking water problems.

    I haven't seen academic research into this yet, but I also wonder about the implications for lithium extraction from sea water (and pretty much any other sources, really). Lithium is basically one of the salts you remove during the desalination process, so the photomolecular effect potentially makes sea water extraction cheaper. Lithium from sea water is an indefinite resource--there's more there than we would know what to do with.

    Edit: actually, scratch the desalination aspects.

    So thermal distillation is almost an order of magnitude behind, and the 4 fold improvement doesn't fully close that gap. In fact, it's worse than that. The multi-stage plant works by recovering heat when the distilled water is recondensed. Merely heating water to do this would take 626 kWh per m3. That's more than two orders of magnitude, and since we can't benefit from a multistage setup to recover heat when using the photomolecular effect, it's going to be a 4 fold improvement over that very high number.

    Still, very big news for improving our climate models.

  • World's largest sodium-ion battery goes into operation - Energy Storage
  • Lithium batteries are often -30 to 80C, but that's just saying what's possible to squeeze some kind of voltage out of them. Basic principle is that the colder it is, the harder it is for chemical reactions to happen, and thus this will affect all chemical batteries to some degree.

  • Shit libertarians say.
  • Private forums can regulate whatever they want. If you don't like it, find another one.

    However, I do think this breaks down when a few outlets dominate. YouTube, Facebook, and Xhitter are all examples. Federated platforms, like Lemmy, are the solution.

  • On the Internet, what is a dead giveaway that someone is actually a kid?
  • Encryption everywhere isn't about the individual content. By making it ubiquitous, it's harder for bad actors to separate the encrypted data they want from the one's they don't. If only special content is encrypted, then just the fact that it's encrypted is a flag for them. It also makes it much harder to ban. It's pretty much impossible to ban the algorithms in TLS at this point. Too much depends on it.

  • Link broken in app

    Not 100% sure if this is a Summit issue or something in Lemmy more generally. Here's the post in question:

    https://midwest.social/post/10123989

    The link to the blog works on my instance for the desktop. Several other users were reporting the link being broken, and it does break for me on Summit, as well.

    When I hit the link on Summit, the requests on the server are GET /api/v3/post?id=2024 and GET /api/v3/comment/list?max_depth=6&post_id=2024&sort=Top&type_=All. It looks like it parsed out the "2024" from the original link and tried to use that in a Lemmy API call.

    5
    Lemmy Support @lemmy.ml frezik @midwest.social

    Post link sometimes goes to the wrong place

    Here's the post in question: https://midwest.social/post/10123989

    Which linked to my blog here: https://wumpus-cave.net/post/2024/03/2024-03-20-moores-law-is-dead/index.html

    On my instance (midwest.social), this works fine. However, some other users were reporting a broken link, and I also see a broken link when using my mobile app (Summit). When it breaks, I see these calls in the server logs:

    • GET /api/v3/post?id=2024
    • GET /api/v3/comment/list?max_depth=6&post_id=2024&sort=Top&type_=All

    Which appear to be Lemmy API calls with some of the actual link data built in.

    2