Which is completely fine. I'm simply saying that if they're going to do this fast release schedule, they can't do it half-way like they are right now. Either ditch the versions, and remove the option to reject upgrades (in the interest of both home users and enterprise) and force them, or they need to split off API versioning and only increment them when changes will break extensions/plugins (which, honestly, I think will only muddle matters more).

The key difference, I think, is that Mozilla still thinks versions are an important signifier, whereas for Chrome, it's the browser platform, not the version, that matters. Mozilla's trying to move to what Chrome's doing without realizing why Chrome does it the way they do it.

Personally, I dislike this e-peen^W version envy Mozilla's got. Their yearly update schedule wasn't broken at all. Since they've taken to this in half-measures, it ends up being a hassle and making nobody happy at all.

At least to some extent. I think Mozilla's trying to switch over to the fast release schedule because of serious version envy, as well as wanting to push out releases faster, but what I don't think they realized is the fundamental difference between how Google's Chrome engineers seem to view the version number.

A user downloads Chrome. That's it. They don't ever need to download updates, since once Chrome is launched, it patches the binary with newest updates automagically. It updates automagically in the background, only occasionally asking for a restart. The version number is buried in the "About Chrome" dialog box, and almost never referred to elsewhere. Extensions aren't tied to versioning, but rather, features. Chrome's fast release schedule works because they've taken some choices and options out of the equation. They're asking people not to support a specific version of Chrome, but rather the Chrome browser platform--which makes it easier: instead of asking what browser version someone's using, one asks what browser, and the supporter can be reasonably assured that the user has the most recent version, with all the most recent bug fixes and features. Whatever the version number is, at best, incidental.

Firefox, on the other hand, still requires some user interaction to download and install updates; updates still go through an installation/update process that's visible to the user, and worse yet, they give the option to the user to ignore them. Extensions are tied to versions, which is why there's frequent breakage on updates. Finally, on first launch, and whenever you visit the page, the version number is there, like it's important or something. Mozilla is still thinking with versions--which makes it important to ask what version of the browser someone's using, and leaving the supporter wondering what updates and what features have been installed and applied.

Mozilla's only gotten the fast releases half-right.

False rumors about their service? I could see their tower from my living room window in Chicago. Nothing in the way. They advertised 6Mbps, but I was lucky to get 2. Watching Netflix was a chore, since the quality would be atrocious and would take forever to buffer; YouTube fared little better. I'd call and call, and get the runaround about problems with the tower that would never be fixed, or blamed that I put the modem in the wrong place (except I had 5 bars...). To cap it all off, it'd cut out for about thirty minutes every night between midnight and one.

I was with them a year. A whole year, because I wrote a lot of it off to being an early adopter. They had to comp my service four months because it was so bad. When I called to cancel? $40 "restocking" fee for the modems.

They're not false rumors. The company is shitty and the service is worse, so much so that it was a *relief* to go back to the awfulness that is xfinity/Comcast.

Which makes me sad. I'm willing to be patient for a while, and this is a beta, but there's been almost no real discussion on the Opera Desktop Team blog about it. I may well eventually cave and reinstall the QT3 dependencies, but I do hope that I won't have to.

Richard Garriot wasn't a founder. They hired him to open up NC-Austin in 2000 or so, but they published their first MMO, Lineage, in Korea in the late '90s. (For a while, it was one of the largest, in terms of subscribers).