• 0 Posts
  • 6 Comments
Joined 1 year ago
cake
Cake day: June 16th, 2023

help-circle

  • The biggest potential issue is if your local password can be used to login remotely.

    I am definitely coming to the conclusion that in the long run, we’re going to end up using something that looks a fair bit like Webauthn / Passkeys for most things that care about security, with something as additional local authentication.

    There are technical reasons why passwords / passphrases are useful, but there is a lot of research that shows just how horrible they are both from a security perspective and from a usability perspective.

    Biometrics are… Convenient, but only useful in low security applications*, and are almost impossible to use for things like unlocking your phone after it reboots**.

    A separate physical object would work really well in some cases, like a desktop computer, but it wouldn’t work at all for something like your cell phone. Or even a laptop. The object would be stolen along with the device it secures.

    I’m really not sure what the long term answer even looks like, but I do hope that it’s not passwords or the like.

    *: You can’t easily change any of your biometrics, but you can most definitely capture someone’s biometrics, and then duplicate them to gain access to something. It wouldn’t be practical to do this every single day, but just to gain access to something once or twice? Easy enough.

    **: The short version: Your PIN / Passphrase / Password / Pattern get fed into a hashing function of some sort, like PBKDF2, which eventually spits out something that can be used to decrypt the key used to encrypt all the data on the device. But this requires a static value, and biometrics are all about fuzzy matches to other patterns.




  • The problem with this is that Reddit, unlike Netflix, is a social media thing.

    And there are two very different ways of looking at the ‘value’ of social media systems.

    The first, and most common, is simply stating that the value is based on how many people are using it. The more people use it, the more valuable it is, and so first mover advantage is almost impossible to overcome.

    Except… That doesn’t really match reality, social media companies die, or stop being nearly as popular. Even ones that used to be wildly popular.

    The big key is that not all users are equally valuable. You want to be involved in a network with people that you find interesting. Even if you never even post, you want to view media that you find interesting.

    For memes, you want to meme with other people who appreciate what you create, and who create vaguely similar works.

    For conversations, you want to have them with people that have something that you find interesting to say.

    On any metric which is just about ‘how many users’, the loss of third party clients, even if it causes the loss of every single user of those clients, is a very tiny drop in the bucket.

    The problem is that many of these users are very likely to be important users. They are the people who give enough of a damn about their experience to go looking for a ‘better’ interface, and giving a damn sure looks to me like a good indicator of caring enough to contribute in a meaningful way.

    Same deal on moderators, all of Reddit absolutely relies on moderators, unpaid moderators, and those are the people who both really give a damn, and who have been quite outspoken about absolutely needing better tooling than what Reddit natively provides.

    If enough of those ‘high value’ users leave, the value of Reddit to almost everyone else drops through the floor.

    It doesn’t go to zero, but it does make it much more likely that other users, the ones that maybe don’t post a lot, but who do view a lot, will follow them.

    And those users are the ones that view a lot of ads, and thus fund the whole thing.

    You simply don’t get this effect with something like Netflix, because the value of Netflix is what movies and shows they have, not what other users they have.


  • Yeah, this really doesn’t look even remotely surprising to me.

    Absolutely everyone involved should already understand that they are not building something that they know how to build.

    Sure, they have plans, and they can build to exactly those plans… But even then, there is no guarantee at all that they will then achieve net positive fusion energy. Because nobody has done that in a controlled reaction.

    But it’s also not like the rest of the world is sitting still. Other projects exist, and sometimes those projects are going to learn things that will impact the design of ITER.

    For that matter, even if they have the plans, some of the pieces are things that nobody has ever built on that scale before, which means that nobody really knows how to build them until they try.

    This is a really good example of a project that you can not accurately estimate.