5 ms·

Firefox's security is still nowhere near up to par, unfortunately.

by Niten 11y ago

robin_reala 11y ago
156 CVEs in Firefox this year compared to 133 in Chrome. Worse, but not ‘nowhere near up to par’.
- Excavator 11y ago
  Some relevant, interesting, material on counting CVEs: https://plus.google.com/u/0/+JustinSchuh/posts/CNEgtJWYTb5 https://plus.google.com/u/0/+JustinSchuh/posts/CNEgtJWYTb5 http://lcamtuf.blogspot.com/2010/05/vulnerability-databases-and-pie-charts.html http://lcamtuf.blogspot.com/2010/05/vulnerability-databases-...
  - robin_reala 11y ago
    Good points there, will bear in mind in the future.
- Niten 11y ago
  CVE counting is a poor proxy for actual security. A browser that does not sandbox its renderers or enforce a permissions model on its extensions cannot reasonably be considered secure in 2015.
  - JupiterMoon 11y ago
    > enforce a permissions model on its extensions Why? Surely as with any software one should only install extensions that one trusts?
    - iopq 11y ago
      You trusted it, but you were wrong
iopq 11y ago
Then neither is Chrome. If you want security you probably want to look into Servo.
- Niten 11y ago
  No, Chrome has a materially better security architecture than Firefox. https://seclab.stanford.edu/websec/chromium/chromium-security-architecture.pdf https://seclab.stanford.edu/websec/chromium/chromium-securit... When it's ready, hopefully Servo will be used as the basis for a browser that can provide security comparable to Chrome. But there's much more to that comprehensive goal than just writing a renderer in Rust.
  - pcwalton 11y ago
    > But there's much more to that comprehensive goal than just writing a renderer in Rust. Which is why the latest Servo builds have very strict sandboxing. People assume that Servo's security story is "just write it in Rust and it'll magically be secure". In reality, nobody on the team is that naive. The goal is to create a browser engine that achieves and significantly exceeds the state of the art. The architecture for that exists right now.
    - Niten 11y ago
      The architecture may exist, by Servo as a day-to-day usable browser does not yet. Don't get me wrong; I absolutely look forward to the day that it does. Neat that they have a sandbox up and running, though. I wasn't aware of that. Is there a paper somewhere I can read about their sandbox architecture?
      - pcwalton 11y ago
        There is no paper per se, but you can look at the code (including readme) for the sandbox here: https://github.com/pcwalton/gaol https://github.com/pcwalton/gaol

Firefox's security is still nowhere near up to par, unfortunately.

by Niten 11y ago

Firefox's security is still nowhere near up to par, unfortunately.

robin_reala 11y ago

156 CVEs in Firefox this year compared to 133 in Chrome. Worse, but not ‘nowhere near up to par’.

Excavator 11y ago
Some relevant, interesting, material on counting CVEs: https://plus.google.com/u/0/+JustinSchuh/posts/CNEgtJWYTb5 https://plus.google.com/u/0/+JustinSchuh/posts/CNEgtJWYTb5 http://lcamtuf.blogspot.com/2010/05/vulnerability-databases-and-pie-charts.html http://lcamtuf.blogspot.com/2010/05/vulnerability-databases-...
- robin_reala 11y ago
  Good points there, will bear in mind in the future.
Niten 11y ago
CVE counting is a poor proxy for actual security. A browser that does not sandbox its renderers or enforce a permissions model on its extensions cannot reasonably be considered secure in 2015.
- JupiterMoon 11y ago
  > enforce a permissions model on its extensions Why? Surely as with any software one should only install extensions that one trusts?
  - iopq 11y ago
    You trusted it, but you were wrong

iopq 11y ago

Then neither is Chrome. If you want security you probably want to look into Servo.

Niten 11y ago
No, Chrome has a materially better security architecture than Firefox. https://seclab.stanford.edu/websec/chromium/chromium-security-architecture.pdf https://seclab.stanford.edu/websec/chromium/chromium-securit... When it's ready, hopefully Servo will be used as the basis for a browser that can provide security comparable to Chrome. But there's much more to that comprehensive goal than just writing a renderer in Rust.
- pcwalton 11y ago
  > But there's much more to that comprehensive goal than just writing a renderer in Rust. Which is why the latest Servo builds have very strict sandboxing. People assume that Servo's security story is "just write it in Rust and it'll magically be secure". In reality, nobody on the team is that naive. The goal is to create a browser engine that achieves and significantly exceeds the state of the art. The architecture for that exists right now.
  - Niten 11y ago
    The architecture may exist, by Servo as a day-to-day usable browser does not yet. Don't get me wrong; I absolutely look forward to the day that it does. Neat that they have a sandbox up and running, though. I wasn't aware of that. Is there a paper somewhere I can read about their sandbox architecture?
    - pcwalton 11y ago
      There is no paper per se, but you can look at the code (including readme) for the sandbox here: https://github.com/pcwalton/gaol https://github.com/pcwalton/gaol