3 ms·

It's more that in-browser JS changes all the time and is basically never audited, nor can it be pinned and prevented from changing. It'd be downright trivial a

by Groxx 9y ago

It's more that in-browser JS changes all the time and is basically never audited, nor can it be pinned and prevented from changing. It'd be downright trivial and unnoticeable to change it to capture your password rather than to behave as advertised. Compare that with the app. Sure it has an updater, but you can use it offline. Don't trust it in day-to-day affairs? Block network access. You can reliably not trust it, and trust that it hasn't exposed your password behind your back (minus on-disk, but that's a risk either way, and it's more audit-able / third parties can build against the format to verify it independently).

yladiz 9y ago
Playing devil's advocate: if you can trust that 1Password is doing everything they can to protect you, the user (using HTTPS, resource integrity) while using the browser app, then are you worried that 1Password may act maliciously? I see this argument all the time but I don't buy it because why on Earth would 1Password do such a thing, if their entire model is based on the customer trusting them handling their data?
- mvid 9y ago
  They can be compelled by an outside force to do so. Or their business model may change
  - Groxx 9y ago
    "Compelled by an outside force" is the main fear for many people. Because it happens all the time, and some of those instances also have an NSL / gag order so they're unable to talk about it until years after the fact (if ever). Or they just threaten violence. Threat models aren't the same person-to-person - this probably won't happen to you (the grandparent), but embedded journalists / people trying to overthrow a corrupt regime depend on this stuff to literally keep them alive. Another fairly common possibility, and one that affects damn near everybody: they can get hacked and have their source code modified. This happens with some regularity, and it can affect apps too: https://www.macrumors.com/2017/05/07/handbrake-app-security-warning-servers-hacked/ https://www.macrumors.com/2017/05/07/handbrake-app-security-... but in a browser this happens silently and unpreventably. Apps don't (usually) update invisibly just because you launched them.

by Groxx 9y ago

yladiz 9y ago

Playing devil's advocate: if you can trust that 1Password is doing everything they can to protect you, the user (using HTTPS, resource integrity) while using the browser app, then are you worried that 1Password may act maliciously? I see this argument all the time but I don't buy it because why on Earth would 1Password do such a thing, if their entire model is based on the customer trusting them handling their data?

mvid 9y ago
They can be compelled by an outside force to do so. Or their business model may change
- Groxx 9y ago
  "Compelled by an outside force" is the main fear for many people. Because it happens all the time, and some of those instances also have an NSL / gag order so they're unable to talk about it until years after the fact (if ever). Or they just threaten violence. Threat models aren't the same person-to-person - this probably won't happen to you (the grandparent), but embedded journalists / people trying to overthrow a corrupt regime depend on this stuff to literally keep them alive. Another fairly common possibility, and one that affects damn near everybody: they can get hacked and have their source code modified. This happens with some regularity, and it can affect apps too: https://www.macrumors.com/2017/05/07/handbrake-app-security-warning-servers-hacked/ https://www.macrumors.com/2017/05/07/handbrake-app-security-... but in a browser this happens silently and unpreventably. Apps don't (usually) update invisibly just because you launched them.