mook

60 ms·

841.
by mook 13y ago
I don't think it is necessary in this case for Mozilla to even be in possession of transiently-non-aggregate data, though. If they want to analyze click-through rates or whatever, they can have URLs that are constant across users and
842.
by mook 13y ago
Having a email-specific link is not aggregate data; that can potentially be used to identify the user. Aggregate data would be a campaign-specific link that doesn't identify the user, such that a click from any person looks equal. In
843.
by mook 13y ago
I don't think tmpfs is involved at all; it's just moving things out of the way instead of deleting, first. The actual delete runs in the background, so you can get an interactive shell back and keep working while the delete happens without
844.
by mook 13y ago
I believe Thunderbird's phishing detection does something along the lines of matching link text with link destinations. Mostly noticed it because mail from Mozilla tend to trip it... It also (by default) disables scripts in mail anyway, so
845.
by mook 14y ago
The identity indeed doesn't have to be an email address; I'm currently using a random thing I set up on AppEngine's free tier (serving a static file for the public side; the auth side is client-local). There's no way I'll be receiving @*.a
846.
by mook 14y ago
That would end up requiring your identity to be a key of some sort, though, since ultimately the thing you're logging into needs to check _something_. Email-like things are useful because somebody has to control the domain; if you don't ha
847.
by mook 14y ago
I have a Firefox extension that does this: https://github.com/mook/browseridp It only requires a static file; currently I have it hosted on Google AppEngine's free tier. Since it stuffs the keys into the normal password storage, sync sho
848.
by mook 14y ago
While it would indeed be enormous, it wouldn't be infinite - after all, they are writing Servo (See https://github.com/mozilla/servo/wiki/Design for details) that's hopefully parallelizable. Of course, whatever that ends up being might n
849.
by mook 14y ago
Small correction: Chrome and Firefox teams are using basically the same implementation of WebRTC; see http://mxr.mozilla.org/mozilla-central/source/media/webrtc/t... - it's an import of the Chromium code into Mozilla. Having the source a
850.
by mook 14y ago
According to https://lwn.net/Articles/525459/ filling the entropy pool with /dev/urandom is a bad idea. Of course, checking my Nexus S, /dev/hwrng just plain doesn't exist; I assume that means the kernel would normally get entropy from o
851.
by mook 14y ago
The code works fine (enough); that's why they will (now) ship nightlies of Win64 builds. The problem is two-fold: 1) They don't want their engineers to spend time fixing Win64 errors, and rely solely on people who care a lot about Win64 (t
852.
by mook 15y ago
If you create a JS macro that is triggered on file open, with the contents "komodo.view.scimoz.indentationGuides = komodo.view.scimoz.SC_IV_LOOKBOTH", it should give you the indentation guides without the whitespace actually there. http:/