Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
mbrubeck
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
37 ms
·
361.
▲
by
mbrubeck
12y ago
Fixed, thanks.
362.
▲
by
mbrubeck
12y ago
Note: nnethercote is a Mozilla employee (as am I). These details are straight from the horse's mouth (dino's mouth?).
363.
▲
by
mbrubeck
12y ago
Compared to the previous deal (where Google was the global default), the new deal also leaves Mozilla more free to pursue different partnerships worldwide. For example, Yandex is now the default Firefox search engine in Russia (as it was t
364.
▲
by
mbrubeck
12y ago
This is true for most sites now, but is being solved gradually, with hard-coded certificate pinning already shipping in Firefox and Chrome, and the HTTP Public Key Pinning extension coming soon.
365.
▲
by
mbrubeck
12y ago
The Readme file at https://github.com/mcsoto/cosmos is a nice concise introduction to the language. The blog post is more of a manifesto.
366.
▲
The Top Ten (Percent) Tech Rules
(modelviewculture.com)
1 points
by
mbrubeck
12y ago
|
0 comments
367.
▲
The Theoretical Minimum
(theoreticalminimum.com)
1 points
by
mbrubeck
12y ago
|
0 comments
368.
▲
by
mbrubeck
12y ago
1/4955 means your fingerprint is shared by fewer browsers than the other commenter's 1/2915. If your fingerprint is totally unique (like mine, with 22.16 bits of information) then it is shared by only 1 in 4,697,672 browser
369.
▲
by
mbrubeck
12y ago
Try it and see! Tor Browser contains modifications specifically to make fingerprinting harder. Panopticlick gets 11.36 bits of information from Tor Browser on my computer, compared to 22.16 bits from unmodified Firefox on the same machine
370.
▲
by
mbrubeck
12y ago
html5test.com doesn't include generators or other JS language features. Instead, see http://kangax.github.io/compat-table/es6/
371.
▲
Distrusting Our Own Infrastructure for Safer Software Releases [video]
(air.mozilla.org)
2 points
by
mbrubeck
12y ago
|
1 comments
372.
▲
by
mbrubeck
12y ago
Yeah, pretty much. "Developer Edition" is the new name for the Aurora channel. There is no more "Aurora." (Well, almost. The "Developer Edition" stuff is desktop-only for now, so the alpha-branch builds for A
373.
▲
by
mbrubeck
12y ago
We've done that frequently with the Aurora channel too, so in terms of release management rules, it's just a new name and focus for an existing channel.
374.
▲
by
mbrubeck
12y ago
Yup. To time the release with the tenth anniversary, it had to go around some of the normal release engineering and localization processes.
375.
▲
by
mbrubeck
12y ago
Yes.
376.
▲
by
mbrubeck
12y ago
These aren't new rules; this has been the Firefox release process since Firefox 5 was released (3.5 years ago).
377.
▲
by
mbrubeck
12y ago
Since last week, Firefox Nightly has process separation enabled by default: https://groups.google.com/forum/#!topic/mozilla.dev.platform... A patch to enable the content process sandbox by default was posted for r
378.
▲
by
mbrubeck
12y ago
The DuckDuckGo addition hasn't landed in Nightly yet. It has rolled out to the release channel, and should be in all channels shortly.
379.
▲
by
mbrubeck
12y ago
This is very similar to Mozilla's proposal to the FCC last spring: http://arstechnica.com/tech-policy/2014/05/mozilla-offers-fc... Mozilla: "The petition calls on the FCC to designate last-mile d
380.
▲
by
mbrubeck
12y ago
It blocks all network requests to URLs on the blocklist. This can include ads, if those ads are listed as known tracking resources. I believe there'll eventually be a mechanism for publishers to declare DNT-compliance or something, t
381.
▲
by
mbrubeck
12y ago
There are "light" and "dark" themes in the Developer Tools options pane. Click the "gear" icon near the top right of the toolbox: https://pbs.twimg.com/media/B2Fj8vbCUAMfjwj.png
382.
▲
by
mbrubeck
12y ago
Here's a direct link to the docs on enabling the experimental tracking blocker in Firefox: https://support.mozilla.org/en-US/kb/tracking-protection-fir...
383.
▲
by
mbrubeck
12y ago
For reference, here the starting point of the Servo implementation: https://github.com/servo/servo/pull/3590
384.
▲
by
mbrubeck
12y ago
I think I'll discuss some "advanced topics" like this in a separate post near the end of the series.
385.
▲
by
mbrubeck
12y ago
I doubt that I'll manage to implement incremental layout or display-list invalidation in robinson. If you're curious about how it works other browsers, though, check out http://www.html5rocks.com/en/tutorials
386.
▲
by
mbrubeck
12y ago
You're welcome! :D
387.
▲
Let's build a browser engine, part 7: Intro to Painting
(limpet.net)
90 points
by
mbrubeck
12y ago
|
9 comments
388.
▲
EPIC 2014: recalling a decade-old imagining of the media’s future
(arstechnica.com)
1 points
by
mbrubeck
12y ago
|
0 comments
389.
▲
by
mbrubeck
12y ago
This thread on the Rust subreddit compares Blow's language ideas to Rust: http://www.reddit.com/r/rust/comments/2gwi11/jonathan_blow_i... He specifically says that Rust "cares too much about sa
390.
▲
by
mbrubeck
12y ago
So you also need to make sure that your phone's browser doesn't have your Google password stored, and/or your phone's storage is encrypted with a strong-enough key.
More ›