This is exactly the reason why I don’t like F-Droid as a way to get apps. You’ll have to trust an additional party when getting your apps, and updates are often a couple days behind. I prefer to get it straight from the developer’s GitHub or Coderberg or whatever.
You’ll have to trust an additional party when getting your apps, and updates are often a couple days behind.
I know how it works, and in this case, that’s fine with me.
F-Droid has an excellent track record; better than many developers have. And I’m not addicted to having the latest versions of everything on the day they’re released. In fact, not immediately jumping on the latest versions has saved me from nasty bugs more than once.
So it could still be considered less secure than N.
It could be, or it could not be. Depends on the particulars, and on the needs of the individual.
Mind, I’m not going around presuming to tell other people what’s better for them, as one or two others in this thread are doing. I’m just stating what’s a good fit for me.
Depends on the particulars, and on the needs of the individual.
That’s not really how things like security works. It’s either more secure or it’s not. The security of a thing does not depend on needs. Now, does the application of it or does someone need it to be more secure? That’s where risk acceptance and the needs of the individual come into play.
I’m not going around presuming to tell other people what’s better for them, as one or two others in this thread are doing.
Same. I’m not saying “stop doing this.” I’m just trying to educate people and make sure they’re not operating with a misunderstanding. Needs of the individual and all that. I think some people just go crazy for something that’s not big tech, and then quit looking at the particulars.
If that were true, threat modeling wouldn’t exist.
I feel like we’re talking about different things. I’m talking about static concepts, if X is more secure than Y, not individual setups where something is tweaked. Threat modeling is tailoring the security to your needs. It doesn’t bend security of a static object or make the application of something less than what it is. It requires one’s actions to do that by not utilizing it.
Take bullet proof glass, for example. Bullet proof glass is more secure than regular glass. Now, do you need (does your threat model require) bullet proof glass? No? Ok, that doesn’t mean bullet proof is now less secure than regular glass, it’s just unneeded.
Part of what I value in F-Droid is the additional layer in the build/release process, because it makes tampering more likely to be detected.
It’s still nice to know a tool like obtanium exists, though. Thanks for the link.
This is exactly the reason why I don’t like F-Droid as a way to get apps. You’ll have to trust an additional party when getting your apps, and updates are often a couple days behind. I prefer to get it straight from the developer’s GitHub or Coderberg or whatever.
I know how it works, and in this case, that’s fine with me.
F-Droid has an excellent track record; better than many developers have. And I’m not addicted to having the latest versions of everything on the day they’re released. In fact, not immediately jumping on the latest versions has saved me from nasty bugs more than once.
Barely and not really. “F-Droid can’t ensure the apps are safe. You still need to trust the upstream developers. We only do some basic check.” https://forum.f-droid.org/t/is-it-as-safe-as-it-is-from-fdroid-official-repo/15956/2
N + 1 > N
N + X - Y ? N
Except now you’re adding an additional party to trust (the -Y). So it could still be considered less secure than N.
It could be, or it could not be. Depends on the particulars, and on the needs of the individual.
Mind, I’m not going around presuming to tell other people what’s better for them, as one or two others in this thread are doing. I’m just stating what’s a good fit for me.
That’s not really how things like security works. It’s either more secure or it’s not. The security of a thing does not depend on needs. Now, does the application of it or does someone need it to be more secure? That’s where risk acceptance and the needs of the individual come into play.
Same. I’m not saying “stop doing this.” I’m just trying to educate people and make sure they’re not operating with a misunderstanding. Needs of the individual and all that. I think some people just go crazy for something that’s not big tech, and then quit looking at the particulars.
If that were true, threat modeling wouldn’t exist. ;)
I expect that’s probably true. It’s safe to assume I’m not one of them, though. Cheers.
I feel like we’re talking about different things. I’m talking about static concepts, if X is more secure than Y, not individual setups where something is tweaked. Threat modeling is tailoring the security to your needs. It doesn’t bend security of a static object or make the application of something less than what it is. It requires one’s actions to do that by not utilizing it.
Take bullet proof glass, for example. Bullet proof glass is more secure than regular glass. Now, do you need (does your threat model require) bullet proof glass? No? Ok, that doesn’t mean bullet proof is now less secure than regular glass, it’s just unneeded.