this post was submitted on 21 Feb 2024
248 points (96.3% liked)
Firefox
17952 readers
342 users here now
A place to discuss the news and latest developments on the open-source browser Firefox
founded 4 years ago
MODERATORS
you are viewing a single comment's thread
view the rest of the comments
view the rest of the comments
How is a piece of software that runs in the browser instead of directly in the os, uses a million little libraries and became popular as a way to avoid scrutiny on the distribution platform less secure than a website?
Let’s assume you have great answers for all that and I’m made to look like a fool: when someone goes to a website, their guard is up. When they click on an app their guard is down.
If nothing else pwas bypass user distrust of weird crap on the internet and that’s a bad thing
@bloodfart @kilgore_trout
"when someone goes to a website, their guard is up. When they click on an app their guard is down."
???
I mean, you can come up with attacks that exploit a users behavior when they think they’re being careful but just consider the propensity to allow camera or location access in browser versus in app: when it’s in browser the phone request says “this website wants to use your location” and gives you the option. When it’s in app the phone (at least used to, I don’t think I have any pwas anymore) says “app_name wants to use your location”.
Everyone trusts the app more. We all see that some website wants to track us and think “yeah right, and then bundle that and sell it!” But for better or worse we trust the applications more. It’s not reflective of actual secure procedures but it’s how people act.
You made another comment about how specifically the webkit jail is very secure. It’s pretty good. That security is exactly why apples trying to tighten up the leash on pwas. One of the only reasons they’ve been able to keep em in and still say “oh we’re so secure” is because they know what the os will allow and what the browser will allow and what’s allowed to be on the os.
This is all happening right after the eu said they gotta allow other browser engines so that’s one of the three legs of that security structure. I think a lot of what we’re seeing is in preparation for pwas to try and start bundling browser engines or targeting the behavior of non webkit engines (not even in like a security targeting way, like build targeting). Once that happens it doesn’t matter how perfectly the security structures of ios and webkit link up, the leaks between ios and gecko or chromium are the new top priority.
Another concerning aspect about having pwas on other engines is how deeply security practices are integrated into ios. It’s got a bunch of little screens and settings and doohickies and gewgaws meant to make otherwise hard to comprehend security ideas not just easy to understand but easy to address.
How can those user facing controls and whatnot be kept up to date in the face of more browser engines far outside the control of the developers making them?
We do everything over the web nowadays and I’m not so sure the second biggest target needs to get exposed more.