I also use Chrome on OS X, and the big feature that enabled me to switch was Keychain integration. I used Safari before, and the ability to share passwords between the two browsers has been invaluable. This feature has been requested for Firefox years ago, but never delivered. Until it is, I can't consider it a serious contender on OS X.

A cubic millimeter is hardly "at the size limit that the human eye can see unaided". A fleck of dust is quite a bit smaller than that, and perfectly visible.

Also, from the FAQ:

Q: Doesn't HTTP pipelining already solve the latency problem? A: No. While pipelining does allow for multiple requests to be sent in parallel over a single TCP stream, it is still but a single stream. Any delays in the processing of anything in the stream (either a long request at the head-of-line or packet loss) will delay the entire stream.

This does not make sense. You're still using TCP, which is a reliable transport protocol, which means packet loss is dealt with at the TCP level, and not seen by SPDY. So the effect of "delaying the entire stream" is exactly the same as with HTTP. The only difference is that you're using fewer TCP connections (one instead of several - in fact, this is one of your selling points!), so the probability that a request will be affected by packet loss in an unrelated request *increases*: packet loss slows down all subsequent traffic on SPDY, since it's sharing a single TCP connection, while with HTTP it only affects traffic that uses the same connection (out of several).

In the real world, packets loss rates are typically 1-2%, and RTTs average 50-100 ms in the U.S.) The reasons that SPDY does better as packet loss rates increase are several: SPDY sends ~40% fewer packets than HTTP, which means fewer packets affected by loss.

But the packets are bigger. If packets are lost due to noise, increasing the size of a packet increases the probability of having an error within it. 10 dollars says you "tested" this in a simulation by fixing the probability of losing a packet, instead of fixing the error distribution. That's going to overestimate the improvements of having fewer smaller packets.

One year doesn't strike me as particularly slow, either. And what's the point of asking people to contribute now, when you've already designed and implemented your idea and put it in production? Isn't this the same Google that froze the WebM specification? Openness for standards is not the same thing as openness for source.

Hi there, I'm a Mac user. I won't care about anything you do in Firefox until you give me Keychain integration. Keychain support is what made Chrome a viable alternative to Safari, and in fact I mainly use Chrome at the moment. But without all my passwords, the barrier to seriously trying out Firefox as my main browser is just too high.

That's not true. A scientific theory is a conceptual framework for interpreting and attempting to predict phenomena. It can be true or false, proven or unproven, or even disproven. Evolution, universal gravitation and phlogiston are all examples of theories.

This is bullshit. Mac users have been hammering Apple with criticism all the way since Rhapsody, and a lot of things you see now are the result of Apple listening to that criticism. I know this sounds strange to you, but that's because you have no idea what you're talking about.

The iMac replaced the floppy disk with the internet. That's what the "i" stood for, originally. And they explicitly said "you won't need a floppy disk because you can send files via email", etc.

Well, it's not that simple. By the terms of the GPL, anyone has the right to port the code, even if the original developers object. The issue was that distributing an application with DRM, even if for free, puts you outside the GPL, and thus at the mercy of any of the copyright holders to charge you with copyright infringement. In this case, there was one core VLC developer (and Nokia employee, lol) who really wanted VLC pulled from the App Store, and neither Apple nor the other developers felt like going to court over it, so they had to pull the app.

Non sequitur. Google's campaign is not based on VP8 being as good as H264, but on it being "patent-free". MPEG-LA is just telling the world that the choice is not between a better, but patent-encumbered codec and a worse, but patent-free one; it is between a better codec with a known licensing model, and a worse one whose patent status is simply unknown - and they're looking into it.

I just checked and XMPP now seems to work on port 443 as well, so I can finally use it from university.

What's important to browser developers is getting the upper hand in their constant pissing contest over Javascript execution speed. Nothing else matters. NOTHING.

Cell phone manufacturer had a chance to get it right, but for years they chose to use custom connectors and make a killing selling overpriced chargers and cables.
Sometimes the invisible hand of the market needs a little nudging from the mighty foot of the state.

Among other things, because BitTorrent has very poor latency (ie it takes a long time from the user's request to when you actually start receiving data).