Firefox Adopts a 6-8 Week Variable Release Schedule (mozilla.org)

AmiMoJo writes: Four years ago Mozilla moved to a fixed-schedule release model, otherwise known as the Train Model, in which we released Firefox every six weeks to get features and updates to users faster. Now Mozilla is moving to a variable 6-8 week cycle, with the same number of releases per year but some flexibility to 'respond to emerging user and market needs' and allow time for holidays. The new release schedule looks like this:

• 2016-01-26 – Firefox 44
• 2016-03-08 – Firefox 45, ESR 45 (6 weeks cycle)
• 2016-04-19 – Firefox 46 (6 weeks cycle)
• 2016-06-07 – Firefox 47 (7 weeks cycle)
• 2016-08-02 – Firefox 48 (8 weeks cycle)
• 2016-09-13 – Firefox 49 (6 weeks cycle)
• 2016-11-08 – Firefox 50 (8 weeks cycle)
• 2016-12-13 – Firefox 50.0.1 (5 week cycle, release for critical fixes as needed)
• 2017-01-24 – Firefox 51 (6 weeks from prior release)

Comments

Wake up, Mozilla morons

[Anonymous Coward]

You keep on doing what YOU want, while ignoring what the USERS want.

Year after year, your popularity goes downhill. Do you even stop to think about that?

Somehow you've been frittering away over $500,000 every DAY for the last several years, and for what?

Your deliberate self-destruction is annoying and pathetic.

Re:Wake up, Mozilla morons

[Anonymous Coward]

Users want a consistent interface, which is to not be arbitrarily mucked with, or copy annoying features of That Other Browser, which do not serve the User. They want their browser not to routinely dumb down features to better accommodate the lowest common denominator.

They want a lightweight, extensible, secure browser that respects their right to privacy, and enhances their privacy on the Internet. So far, they've done just about everything in their power to make sure Firefox does not supply these features; for example: Pocket. No reason for it to be anything other than another extension. Promising to do away with the extension framework down the road. Doing absolutely nothing about browser fingerprinting, when they're in the ideal position to fight it. Not allowing power users to whitelist sites that don't abuse Javascript. Etc. Etc.

They want the extensions that attracted them to Firefox in the first place to continue to work, which is again something they've promised to do away with in some future release.

In other words, if you had a goose which laid golden eggs, you'd probably want to take care of it and stuff, right? These guys do anything in their power to neglect and annoy that goose. The goose is going neurotic and plucking its own feathers and banging its head against the wall, and they think it's great.

Re:Wake up, Mozilla morons

[drinkypoo]

What do the users want that they aren't supplying? I haven't noticed a problem.

Performance and reliability. Graphics updates are slow and after I resume from suspend I have to restart the browser or it chokes every few seconds. Sadly, I went to Pale Moon and it has the same problem.

Re:Wake up, Mozilla morons

[Barefoot Monkey]

Sounds like you want Firefox Extended Support Release (ESR). ESR releases receive a major update only once every 10 months, but still receive the same bugfix and security patches as the regular releases. Also, when you do get updated to a new ESR version, you know that it's one that's already been supported as a regular release for 2 months, so there's very little chance of surprise problems.

The current version is Firefox ESR 38, which was released as ESR on 11 August last year. The next one is ESR 45, coming on 31 May, which will last all the way until 21 March next year.

Hope that helps you.

Re:Wake up, Mozilla morons

[hackertourist]

It's more what they are taking away. It started with the status bar, then there was the ill-conceived move to Australis, version 44 removed fine-grained cookie permissions, next they're planning to kill off extensions.
Over the past few years they've spent countless hours on integrating features few people cared for, and more hours taking away features we actually used.

Removing features faster than ever

[thegarbz]

With the new faster release cycle they can alienate the existing user base with more efficiency and at a faster pace than ever before!

Re:Holy Cow

[phantomfive]

How can you possibly know if that is a tough schedule or not, without knowing what they are going to put into each release?
Release cadence is like a CPU clock speed.....it tells you nothing unless you know how much work is done during each cycle.

Dumb

[110010001000]

The idea of having a "regular schedule" of releases is stupid. What if you didn't have any compelling features to add? You are just going to do a release because that is what the release schedule says? Here is a hint guys: writing software is not supposed to be just to keep you busy. It is supposed to deliver a product that is useful.

Re:Dumb

[CrankyFool]

You don't think they're figuring out what features they're going to build into a release when they start the 6-8 week cycle for that release, do you? I suspect that, like any other sane engineering organizations, they have a large backlog of features and issues to pick from for each release.

Re:Dumb

[PsychoSlashDot]

You're mostly right. I'm sure they have a very detailed plan for what features to remove at each release. By end-of-year 2017 the goal is to have Firefox resemble Notepad, only without mouse, keyboard, or clipboard input support.

Re:Dumb

[Junta]

Conversely, if you have no planned cadence, you can land in development hell, churning eternally without actually releasing because in the time feature A has matured, you decided to squeeze in feature B and decided it can't release without B, rinse and repeat.

Re:Holy Cow

[sunderland56]

QA still needs to test the entire product every single release, no matter what was added or changed.

And just because you *can* slip an unfinished feature to the next release, doesn't mean your bosses will be happy about it.

What about instead waiting until it's ready?

[Anonymous Coward]

Pushing out releases just to check a checkbox off is very Agile. Instead, you should work towards making better software instead of trying to hit metrics.

Re: What about instead waiting until it's ready?

[phantomfive]

Ever since Mythical Man Month came out, there has been plenty of focus on managing programmers, on how to improve managers (which isn't necessarily bad, and MMM is a great book).
The end result is that sometimes programmers run around like mindless monkeys, following process, and trying to figure out what is wrong with their process and what change they need to make to make things better.

In reality, it's not the process, if you want a good team and a good product, you need to focus on improving the programmers: the process is secondary. The best Agile proponents actually do focus on improving the skill of the team, but too many of the consultants out there are focused on process, process, process,, which leads to mindless monkeys.

But getting back to MMM, Fred Brooks pointed out that unless you have a giant team, the exact processes you use are kind of irrelevant.

Re:Holy Cow

[fustakrakich]

This upgrade treadmill is getting ridiculous. Can't anybody build anything that will last more than a few weeks? Am I that old to believe long tern stability is a good thing?

Flexibility?

[MtHuurne]

some flexibility to 'respond to emerging user and market needs'
(snip)
2017-01-24 – Firefox 51 (6 weeks from prior release)

I don't understand where they'll get the flexibility from when they're planning releases a year ahead...

Re:Holy Cow

[phantomfive]

Yeap. Let it be a lesson to any software development team: every time you write a line of code, or add a feature, or remove a feature, ask yourself: "Does this feature clearly make the product better?"

If you answer no to that question too often (or if an unbiased observer would answer no), then you'll just be pushing things around haphazardly, like Google (and more likely you'll be making things worse).

Re:10 days should be enough.

[phantomfive]

Brendan Eich, then working for Nestcape now still at Mozilla, defined created and demoed the first version of Javascript in ten days.

And it shows. The web would have been better if he'd spent a little more time thinking about it.

Holy beef

[invictusvoyd]

Perhaps they leave the QA to the users .

Seriously?

[JustAnotherOldGuy]

My children will be using Firefox ver 7,462,354,846.01

They'll need to buy more memory just to keep the version number from using up all the RAM.

Here's an idea for a feature

[JustAnotherOldGuy]

Here's an idea for a feature...make it stop inexorably sucking up more and more memory until it slows to a crawl and then crashes.

Now that would a cool feature.

Re:Holy Cow

[fustakrakich]

In Mozilla's (and Google's) case that would definitely be the advertisers, they are the ones paying the bills. These 'upgrades' seem to be for their exclusive benefit. Truly necessary upgrades also go into Seamonkey, and that hasn't happened since November 8th. I have to assume that Firefox 'upgrades' are purely cosmetic and/or economic.

Re:Holy Cow

[fustakrakich]

Well, the key word now is "improvement", isn't it? That seems to be a matter of perception. I would love to know where the "improvements" are in a program that is no faster than, has grown just as fat as, and still has less than a quarter of the user features of its ancestor. Most of the processing is used to hide the garbage.

Re:Holy Cow

[fustakrakich]

I have seen the change logs, and in my opinion they reflect no improvement over the decades old all inclusive program that preceded it. And for those who remember, when it first came out Firefox was supposed to be lean and fast and *stripped of cruft*. It's not any of those things now. I don't know of any single browser right now that is not a 30+ megabyte download, and they all run about the same speed. I see no disadvantage of sticking with something a bit more familiar that I can run for years without having to think about "upgrades". And when I do upgrade, at least it still has the same familiar face from last century, hasn't aged a bit.