It might not be long before the phrase "new version of Firefox" ends up being not very thrilling at all. On schedule, just six weeks after the organization gave the final go-ahead to release Firefox 5 (the "dot-oh" is oh, so 2010), this afternoon installed versions of version 5 were buzzing their users that Firefox 6 was available.
It's part of an ambitious new scheduling plan by the Mozilla organization to expedite new features as rapidly as possible, in the wake of increased competition from Google Chrome, whose latest stable release is version 13, and whose dev channel release is on the tail end of 14. But the plan has met resistance recently from developers and IT personnel who claim six weeks is not enough turn-around time to test a new release before distributing it to users in their enterprise.
progress element (which is also supported by Google Chrome). This is a gadget found on basically all operating environments today. Contrary to some documentation, the
value attribute so the progress bar doesn't have to stay static.
Last year, Firefox 4 was subject to multiple delays, whose cause appeared at the time to be stability. Field telemetry generated by testers of Mozilla's beta channel showed the number of crashes per 100 active users, whose statistics are generated daily and posted to the general public, had yet to fall below critical benchmarks to be declared stable. Some major additions to the product, such as the long-awaited WebSocket API, a critical feature of HTML5, had to be delayed, and are only now making their premiere with Firefox 6.
XMLHTTPRequest for this task, a critical function for AJAX. Along with WebSocket comes a new, full protocol that developers believe will be more secure and less susceptible to man-in-the-middle attacks. Chrome has supported WebSocket since version 4.
Last year during the Firefox 4 delay, there were plenty of interim point releases to Firefox 3 in the meantime. Memory of that fact has led many to ask whether Mozilla's new scheduling scheme is actually just a multiplication of the old system by a factor of 10. In terms of general appearance to the end user, Firefox 6's differences from Firefox 4 are best pointed out with a magnifying glass.
Just six weeks ago after the Firefox 6 release, e-book author Aaron Gustafson told the developer blog NET Magazine, "In talking with end users I have found that the accelerated release cycle of Firefox is causing confusion. Many people are surprised when they are asked to upgrade because they feel that they just upgraded to Firefox 5 (or 4 in some cases). I am somewhat concerned that this shift may hurt the Firefox brand in terms of perceived reliability and stability because of the way they engage users in the upgrade process."
But it's not that Mozilla takes just six weeks to make a new browser. The number's actually more like 24, but the new scheduling system is staggered by six weeks on four different tracks. The dev track, formerly called "Minefield" and renamed "Aurora" last April, also takes six weeks, after which there's a six-week public beta period.
"Each release channel is backed by a Mercurial repository containing a distinct copy of the Firefox source code," explains a draft document of Mozilla's official explanation of its development process. "As a set of changes progresses through the repositories, features that aren't quite ready are disabled. New features are never directly added to the mozilla-aurora or mozilla-beta channels. Features that end up disabled or miss the scheduled transition to the experimental channel can be pulled again the next time the schedule permits. This policy does allow for features that take a long time to develop. It's just that they'll be present only on mozilla-central until they're ready."
This would appear to suggest that new Firefox versions will not be feature-driven, like previous releases up to version 4. If new features take just as long as they take, to borrow a phrase from a pastry chef's kitchen, then indeed the only major change from developers' perspectives should be which side of the decimal point the release number falls on.
Incidentally, there have been plenty of reports over the past week suggesting that Firefox 6 had already been released unofficially, and there was no real need to wait to download it on the release date. This graph from Mozilla's crash telemetry report suggests otherwise. It clearly shows a prohibitive number of crashes per 100 users - over the 80 mark - for browsers reporting themselves as non-beta Firefox 6 over the weekend, and that number subsiding dramatically to a normal figure yesterday. The same browser that crashed so many times Saturday is probably not the same one that works like a charm today.