• 0 Posts
  • 18 Comments
Joined 1 year ago
cake
Cake day: June 18th, 2023

help-circle
rss
  • I get the same sort of feeling. I’ve been fortunate enough to go to a full size track as well, though it’s a different experience.

    There’s something about how open a gokart is. You feel every little bump in the track, and you can see EXACTLY how close your wheels are to the edge. Even leaning for weight transfer makes a difference in a kart. I do miss having power steering though. My hands and arms are always dead after a long karting session.



  • I hadn’t heard of this, and it makes me quite angry. Sooo many phones have Qualcomm chips in them, including every phone I’ve ever owned.

    The amount of data they’re collecting is unreasonable for what’s actually required for A-GPS (the only actual feature this enables). And it’s all completely invisible to users because they don’t even include the Privacy Policy with the phones.

    If I want to stop Qualcomm sending out a bunch of my data unencrypted over the web, I’ve got very few options, all of which are: Buy a new Phone…

    Edit: After more research on this, it seems this A-GPS request is still happening from the OS, which controls Wifi. /e/OS just didn’t reconfigure the Qualcomm driver like GrapheneOS. This isn’t a hardware or firmware backdoor or something like I thought initially. The article seems like an ad for NitroKey / NitroPhone, which is just a modified Pixel with GrapheneOS on it. I might look at GrapheneOS for my Samsung phone.




  • With some of the cats I’ve interacted with, letting them outside would probably be cruel itself. One is scared of the weirdest things, like string, and complete fails at even hunting bugs in the house. Not to mention the crazy heat waves we’ve been having that would definitely be a problem for an indoor cat with thick fur.


  • I pretty much agrre with you and everyone else here. AI is not as useful as a lot of people are pushing it to be.

    I used GitHub CoPilot for several months, and some of the advanced autocompleting it can do during refactors is amazing. But I wish I could use just that as a dedicated feature (like an AI context aware find/replace).
    I found most of the time CoPilot was more distracting than helpful. I’ll have 90% of a solution done in my head, and then as I’m writing it out, CoPilot will recommend something that’s almost, but not quite, what I want and completely interrupt my train of thought.

    When I code with AI, it seems to just move all my time from coding to debugging and reading the AI’s code. For me, I end up with a worse result in the end than if I just wrote it all manually, and I don’t end up internalizing the structure as well.

    I use ChatGPT a little differently now that I realized this. Much more like Rubber Duck Programming, where I’ll set up the conversation so ChatGPT asks ME questions. For writing documentation, I’ve found this to produce far better and more accurate results. You can even ask it for a summary at the end.


  • I don’t know any details on how data was leaked (Apparently “the fappening” was 8 years ago now, so I’m sure those issues are fixed). I’m probably a little biased against Apple, but I think this applies to any cloud service, and I’m just using iCloud as an example:
    It may have been social engineering , but I don’t think the specific attack matters much. There are millions of iCloud uses (including a lot of high profile individuals), and that makes the whole system a target. Only the most targeted of attacks would bother trying to attack someone’s home NAS.


  • Apple had some pretty major iCloud leaks a few years ago. I don’t trust Apple any more than I trust anyone else. Every cloud service has the same problem across Apple and Android.

    With Apple being one of the top names in anti-right-to-repair, I can’t see myself buying another one of their products.

    At least with an Android phone it’s a lot easier to run custom apps or even OS firmware to control your data better. Having everything on my phone backed up to my own NAS is great.



  • As far as I know, I’ve never dropped incoming emails, but I have no way of knowing due to insufficient monitoring. My mail server is in a datacenter, but I don’t have any redundancy or failover. It’s not worth my time to set up vs paying someone to manage email for me. Google’s spam filtering and integrations are also better than I’ll ever be able to achieve for $6/month Google Workplace Gmail.


  • It’s not better than a full size desktop or laptop keyboard, but I find touchscreens are absolutely better than the tiny plastic keys on say an old Blackberry. If you’ve ever typed a password with symbols, unicode, or emoji characters, then you’ll appreciate how bad a lot of those keyboards were. The worst of them didn’t even have full QWERTY layouts.


  • Yeah, as someone who hosts a private email server, don’t do it. I don’t use my mail server for anything remotely important, because I don’t have enough monitoring in place to be sure it’s working 100% of the time. Silently dropping emails is a huge deal, especially if your monitoring is email-based… It’s 100% worth it paying for email hosting if you want to set up custom domains and mailboxes.


  • You seem to have a different remembered version of history. They very first iPhone had Wifi, and could do loads of things other devices couldn’t, like play video, browse the web using a real full browser, and IMO way better typing than any physical keyboard.

    You’re acting like iPhone’s lack of an App store on day one put it at a disadvantage, but there weren’t exactly a lot of other options. The app store was released only a year later, and you could do loads from the browser before that.


  • I’m not OP, but if transcoding is happening on user CPUs, it’s theoretically possible to modify or inject stuff into the transcoded video. There’d need to be some way of validating a transcode matches the original, which is non-trivial.
    A consensus algorithm could work, but that would massively increase the required compute. I’m not even sure things like NVENC vs CPU ffmpeg are deterministic in how tbey compress video. Different encoders could very likely end up with visually identical transcodes, but the hashes wouldn’t always match.
    Maybe someone else has a better idea for validating transcodes?


  • “A few double spends” is underestimating the impact. When this has happened in the past, the whole network gets fragmented, and at some point everyone needs to decide which version of history to throw out, allowing potentially anyone to double-spend in that time frame. A bad actor with enough compute could cause a network split and put whatever they want in the ledger. Getting caught isn’t really a concern if it’s all anonymous wallets, and it only takes 1 unnoticed transaction to move millions.

    The entire basis for trust in Bitcoin (and any proof of work blockchain) is that the network is so big, no single actor has the resources to become a majority and influence the ledger.


  • I think you’re missing a critical part of how blockchains function: If Bitcoin was running on only 100 Mac Minis, there is nothing stopping someone buying 101 more Mac Minis, becoming dominant in the network and suddenly they can decide to just print their own bitcoins for themself.

    The profitability of running Bitcoin miners is proportional to the market cap and the value of Bitcoin itself. For Bitcoin to remain stable, the total value must remain less than the cost of hardware to dominate the consensus algorithm.