Firefox 5 Scheduled For June 21 Release

An anonymous reader writes "Mozilla has updated its Firefox 5 release schedule and is apparently upbeat that it can release the browser even earlier than previously anticipated. The release was pulled in by a week to June 21. Mozilla is now also using a Chrome-like versioning system for Firefox — where the final Firefox 5 may be called Firefox 5.6.44.144, for example."

Comments

High version numbers

[Sonny Yatsen]

Of all the stupid features from Chrome to pick up, the version numbers is, by far, the dumbest. Has anyone considered how stupid a version number in the high double digits might be? Firefox 81 seems kind of clunky, doesn't it?

Re:High version numbers

[dotancohen]

Of all the stupid features from Chrome to pick up, the version numbers is, by far, the dumbest. Has anyone considered how stupid a version number in the high double digits might be? Firefox 81 seems kind of clunky, doesn't it?

I think that Windows went from 3.x to 90+something and even got up to the low thousands, before coming back down to single digits.

Re:High version numbers

[Bloodwine77]

Yeah, I personally don't like the major version number scheme used in this way, especially if there are going to be three or more versions of Firefox per year. I am old-fashioned and prefer the X.Y.Z approach. I could maybe see a YYYY.X approach, such as 2011.1, 2011.2, 2011.3, etc. that would track major versions per year. I never realized how close the new Firefox 4 was to Chrome with respect to the UI until I downloaded and installed Chrome the other day. Firefox seems to be hellbent on ripping off Chrome.

Re:High version numbers

[kvvbassboy]

What's the big deal? Frequent releases are good, it keeps crowd interest in your browser alive. It doesn't matter for me though, I use minefield which I presume will keep getting updated.

pAnd I am sure you don't have to worry about version 81, they will switch to a different version naming scheme, or even just fork off with a different browser name, who knows.

Re:High version numbers

[AmigaHeretic]

What actual features and improvements could they possibly have added in "8 WEEKS" since the release that they have had time to actually put through an Alpha test, Beta test, and then full release that would warrant a VERSION 5!?!

This seems crazy lame to me. The browser has slowly gotten bloated, now the number? Why?

Re:High version numbers

[MosX]

It was a joke man.

Re:High version numbers

[heptapod]

I disagree. I see a program that's in its 20th iteration and believe it's reliable but lacking anything competitive or exciting being used by new alternatives. Around the 8th iteration they should rebrand and start out at 1 again or hint that it's "beta" making it more attractive to users who can claim eliteness by getting in on the ground floor.

Re:High version numbers

[TheWanderingHermit]

I guess FF people think so. Why else would each version require more and more and MORE memory and CPU time?

I can't even use FF in a multi-tasking environment anymore because it's gotten so big.

Re:High version numbers

[jav1231]

Maybe all the features they took out of it from 3.6 to 4 they're putting back and calling 5.

Re:High version numbers

[Anonymous Coward]

I'm trying to find an older version of Linux to use on a really old computer. Does anybody know where I can find Slackware 6? I can't seem to find it anywhere! It's like it doesn't even exist!

Re:High version numbers

[dhammond]

I think what they're moving towards is a time when end users don't care about version numbers at all, just like they don't care about the version numbers of websites. I'm sure many websites attach version numbers to development milestones as a way of organizing development. I do this for some websites I work on, but when we launch new features, we never broadcast to the world, "Check out version 8.1 of mysite.com!" If there is an announcement at all, it is focused on the actual features that were added.

Chrome updates automatically and without fanfare, similar to a website. I have lost track of the version number of my version of Chrome, and I don't see anything wrong with that.

Re:High version numbers

[N0Man74]

I think that Windows went from 3.x to 90+something and even got up to the low thousands, before coming back down to single digits.

And just imagine how confusing it is going to be waiting for the next version of Windows when we get to Windows 94! Will they have to come up with a new naming system so they don't repeat 95 again? ;-)

Re:High version numbers

[tlhIngan]

Also annoying, and you end up with people who just really don't want to keep upgrading every month. I held off upgrading a bunch of PCs from FF2 expecting FF4 "any day now" for months, and did the same for others running 3.5 - I didn't update them to 3.6 expecting to update them to 4.

But if 5 is comig out in a few weeks, it may be best to just hold off until then, no? Having to do the whole update and fix plugins thing gets annoying.

Re:High version numbers

[_0xd0ad]

Posts that have been moderated once don't have a moderation adjective. It's been that way since before the site redesign, so I think it was intentional. A post with an Excellent karma bonus and one +1 mod will be rated +3 Normal.

Re:High version numbers

[BagOBones]

What if we put a leading 0. on it so it is 0.81.0 then it would follow the naming standard of a bunch of open source projects that almost never seem to get to or past version 1.0

Re:High version numbers

[Bert64]

Which is why Ubuntu uses dates for the version number...

Re:High version numbers

[cjb658]

I was 11 at the time, and I would always ask my friends "Do you have Windows 95?" They'd often respond with, "No, I have Windows 94."

Re:High version numbers

[SimonTheSoundMan]

Whoosh!

Re:High version numbers

[IB4Student]

Yes, after 30 years, they are still going to be called Firefox and using the same numbering scheme.

Re:High version numbers

[Eponymous Coward]

I wouldn't call the version number a feature.

If you listen to the Chrome people, they will say the version number is interesting only from a engineering or technical point of view and don't expect their users to pay attention to it. I think it's working because you rarely hear talk about Chrome 10 whereas Firefox users often state a version number (like Firefox 4).

If Firefox continues to release frequent updates, by the time version 81 rolls around, nobody will be quoting version numbers any more. It will just be Firefox.

Re:High version numbers

[josgeluk]

What actual features and improvements could they possibly have added in "8 WEEKS" since the release that they have had time to actually put through an Alpha test, Beta test, and then full release that would warrant a VERSION 5!?!

Who says they have just started doing that?

Re:High version numbers

[kingbilly]

The latest "That's What I Call Music!" is now at 37.

Imagine "That's what I call a browser 22!"

Re:High version numbers

[slapout]

Maybe they should just have seasons -- Firefox 4 Summer edition, Firefox 4 Winter edition, etc...

Re:High version numbers

[armanox]

That would be because there was no Slackware 6.

Re:High version numbers

[houghi]

OK, I'll bite:
Windows 95

Re:High version numbers

[jonadab]

> Of all the stupid features from Chrome to pick up,
> the version numbers is, by far, the dumbest.

I disagree. I think that's relatively harmless. Users won't even notice.

The dumbest has got to be either the stupid idiotic moronic "tabs on top" or the completely insane lack of menubar (pioneered by IE7, granted, but Chrome has had it for a good while now too). I can't decide which is worse. (Yeah, I know, you can turn them both off and get a more or less normal UI. For now. That will change in a future version, mark my words.)

Meh. Firefox 5? I don't even like Firefox 3. Version 2 was the best so far. Okay, page loading was a little slow, and version 3 improved that. But performance has never been as important to me as stability (by which I mean not crashing and even more importantly not losing data), and Firefox 2 is a great deal better in that regard. If the new CSS features in recent versions of Gecko could be backported into Firefox 2, I'd be pleased as punch.

Re:High version numbers

[Nimey]

You don't think it likely that they can work on more than one thing at once?

Re:High version numbers

[thePowerOfGrayskull]

Of all the stupid features from Chrome to pick up, the version numbers is, by far, the dumbest. Has anyone considered how stupid a version number in the high double digits might be? Firefox 81 seems kind of clunky, doesn't it?

Far less ridiculous than "0.12.6-12ubuntu3" (apachetop) or even just "0.48.1" (inkscape) - presumably because those packages aren't at "1.0" quality yet, whatever that means.

If you're making user-facing tools, using decimal numbers is confusing at best. "Did you say install 2.3.1 or 2.1.3?" Simple, whole-number revisions make it much easier to manage.

Re:High version numbers

[Jugalator]

Of all the stupid features from Chrome to pick up, the version numbers is, by far, the dumbest. Has anyone considered how stupid a version number in the high double digits might be? Firefox 81 seems kind of clunky, doesn't it?

Only because you/we are not used to it. Who cares? By Chrome 26 and Firefox 12 we won't really look at version numbers like we've done in the past (or look at them at all...), and only see two web browsers that are releasing time-based releases as opposed to feature-based releases where the point won't be to market by features and version numbers, but by following the latest web standards and web browsing trends well.

AFAIK, Firefox is also moving to silent updates, yet a move aimed to remove the old "launch party" thing with new releases that were once released as rarely as once per year, with long, long QA phases at the end to test the huge beasts of new features, bogging everything down before the developers could pick up speed again.

Re:High version numbers

[kripkenstein]

What's the big deal? Frequent releases are good, it keeps crowd interest in your browser alive. It doesn't matter for me though, I use minefield which I presume will keep getting updated.

Yeah, minefield will continue to update every night.

Re:High version numbers

[Jugalator]

What actual features and improvements could they possibly have added in "8 WEEKS" since the release that they have had time to actually put through an Alpha test, Beta test, and then full release that would warrant a VERSION 5!?!

It's not about "warranting" stuff anymore. Forget all about what you learnt about that, like you did with Chrome. The releases are now time-based - period. Not feature-based. Read these version numbers more like "milestones". "Version number" has too many assumptions associated with them nowadays, so maybe we'd be better off to just call them milestones like I know many Chromium devs already do internally today.

So what Firefox 5 will be released with simply depends on how many features Mozilla has finished by June 21.

As for that question... Many features and fixes has already been checked into the Firefox trunk that were too risky to be included in Firefox 4. I don't think Mozilla will have too many problems in "warranting" an update this coming summer.

Re:High version numbers

[heptapod]

Indeed, I feel like a gigantic idiot for upgrading to FF4 when FF5 is coming out in June. Plus flash on FF4 w/ linux is really awful.

Re:High version numbers

[foobsr]

Yes, after 30 years, they are still going to be called Firefox and using the same numbering scheme.

"As of March 2011, the latest stable release of GNU Emacs is version 23.3." (Wikipedia)

Possible, at least.

CC.

Re:High version numbers

[six]

4.0 was NT4 ... my guess is 95/OSR2/98 were all 3.2x

Re:High version numbers

[ifrag]

Just look at NVidia. Their driver package is up to version 270.51.

I've started to like 3 part version numbers lately, although only if proper change significance is applied. The standard GNU version numbering with major.minor.revision seems useful and appropriate.

Re:High version numbers

[devent]

Totally agree. Chromium is now with version 11 and I still can't use the browser, because I can't change the fonts. Finally I can change the minium font size (I think that was with version 11) but I still can't fix the font. I really don't like TimesNewRoman or Arial. For me Chrome is still like a 0.11

Re:High version numbers

[Fallen Andy]

it's called (grabs the brown paper bag and barfs) marketting...

or as a famous ex-intel design engineer puts it "Blue Crystals (R)"

and... mozilla was *always* a bloaty program, it's just got more re-engineered (sic) in it's bloatiness...

(used to build the first MPR as a machine burn in lolz)

Maybe i'll not live to see Firefox XP or Vista if i'm lucky...

After fixing stuff for a friend so he can have a familiar browser (he updated to 4.0) i'm becoming more and more a convert to chrome...

Andy

Re:High version numbers

[tuppe666]

What actual features and improvements could they possibly have added in "8 WEEKS" since the release that they have had time to actually put through an Alpha test, Beta test, and then full release that would warrant a VERSION 5!?!

This seems crazy lame to me. The browser has slowly gotten bloated, now the number? Why?

Many Features, One of the problem with Firefox 4 was timing all the new features to coincide all at once. Large projects do not work like small ones. The Linux kernel is released every 12 weeks, but new features that have been in development for years are only included in this release. Not that I think you care judging by the in fashion "bloated" comment. Its half the size of Chrome and take up less memory, Show me the Bloat!

Re:High version numbers

[sconeu]

I like the menubar on my desktop.

On my netbook, I want as much screen space as possible, so I turn off the menu bar and put the "tabs on top". That extra vertical space makes a different.

Re:High version numbers

[JustinOpinion]

Hm. When thought about this way, the "escalating version numbers" bother me much less.

When you think about it, I have tons of software on my machine that goes through version numbers without me noticing. I do my "apt-get upgrade" and all my software is just a little bit better. If Firefox really goes to this "milestone model" then it will be like that also. Rather than me getting excited about a new release, it will just appear in my normal updates and I won't pay it much heed.

From a user perspective this is probably better. Fixes and features get to me sooner and more reliably. My browser changes more incrementally rather than in big jumps. (This is a debatable advantage, since many users hate it when their software slowly changes as they update.) I don't have to think as much about Firefox releases.

This may, however, be a net loss from a hype/advertising point of view. Rather than getting all kinds of jazzy articles about the "next awesome version of Firefox", please will be less excited and it will just become "oh, yeah, there were some bugfixes in the last milestone... whatever..." Time will tell, I suppose.

Re:High version numbers

[VGPowerlord]

4.0 was NT4 ... my guess is 95/OSR2/98 were all 3.2x

Windows and Windows NT were considered two separate products lines until the old Windows line died after Windows ME. Thus, both Windows 95 and Windows NT4 had versions starting with 4.

Re:High version numbers

[rhook]

Windows 2000, XP, Vista, Windows 7...

Re:High version numbers

[McNihil]

I believe that version 1.0 in these kind of projects is equal to the perfect implementation... effectively unattainable by its very nature. This approach can be used for smaller closed problem sets that have very definite size of "impact."

Larger stuff like a browser, paint program, music composing are more open ended and will thus require many different major versions but sooner or later it will end up being irrelevant what it is.

So while the Chrome and Firefox version numbers (I argue the Fedora kernel version numbers too) have a crufty digitatis associated to them they do convey a lot of important information to the users who are in the know.

Re:High version numbers

[robot_love]

Maybe because at the end of the day your opinion about an arbitrary number doesn't really mean anything? I'm not trying to be rude, but what the hell does it matter what they call it? Will they offend the International Software Versioning Board? No. Is the planet going to spiral into the sun? No. You think numbers should go up in smaller increments. They think they should go up faster. Who gives a crap?

And +5 Interesting to one man's opinion about an arbitrary number? Come on!

Re:High version numbers

[LordLimecat]

Everyone remarking on the new firefox release system seems to think this has something to do with the numbers "catching up". Everything Ive read indicates that that has NOTHING to do with what theyre doing here.

Major (x.0) releases tend to indicate "we added new features"-- this is what chrome does, this is what firefox (usually, not counting 1.5 and 3.5/6) does, this is what IE does. Point releases (3.6.11, 3.6.12) indicate fixes.

This isnt changing. What IS changing is that rather than having a new "dot-oh" release every 1-2 years, theyre making them more rapid, and adding fewer features. This has several benefits, as chrome has shown in its rapid dev cycles:

• * New features are released sooner, keeping firefox competitive
• * Fewer major changes per release means easier regression control and testing
• * Releasing a feature thats "ready" sooner means that it gets public testing (not to the detriment of pre-release testing either, in theory), and that it gets any bugs ironed out sooner
• * Makes sure new changes are relevant-- it would be unproductive if firefox spent 2 years perfecting Tab management, if by the time it comes out, tabs have been obsoleted by the NextBigThing (aka Hurd Syndrome).

The numbering on releases is utterly unimportant, except as an indicator of whether major, or minor changes have been made. I dont know of any non-techies who even know what version of firefox or IE they are using (which is, in fact, part of the problem); and half the time I dont even know what specific number chrome is up to (because they dont really trumpet the number at all). All these complaints are over nothing; where has someone actually asserted that "Opera 11 is better than Chrome 10, because it goes to 11"? This is a nothing complaint.

Incidentally, slashdot seems awfully broken now; neither UL or OL seems to have any "list" delimiters any more (mine are manually inserted).

Re:High version numbers

[LordLimecat]

Chrome has done a LOT right. In the 3 years since it has been released, here are the major, user-noticeable changes chrome has had...

• * Several (3-4) major JS engine revisions (newest version is several times faster than initial release, which was itself quite fast)
• * Added extensions
• * Seperated extensions to be per-process
• * Added theming
• * Added MSI packages and AD GPO templates
• * Added Private browsing
• * Baked in respectable native "flashblock" / "noscript" controls (though not as full featured)
• * Added Cloud sync
• * And a lot of other stuff (HTML5, hardware rendering, autofill, cloud print, automatic page translation, etc)

Given how effective its development model seems to be, it might not be a bad idea for firefox to take a few pages from their book (even if copying the interface wasnt really necessary)

Re:High version numbers

[rsborg]

Agreed. My favorite versioning schemes are Ubuntu and OSX. I prefer Ubuntu for it's dual-version system, and OSX for it's simplicity (though I think Snow Leopard was a horrible name considering Leopard was right before it).

Android's versioning scheme was upended by their decision to compete with the iPad (since Honeycomb is such a departure in OS). If they acutally manage to do what Apple did, and merge the code bases, then it won't be so bad, but for right now, there is a tablet OS and a phone OS and a single-dimension versioning system does not work.

At the silly end are Microsoft's wildly differing brands for Windows...

Re:High version numbers

[ildon]

The joke is Windows "3.1", Windows "95", Windows "98", Windows "2000", and Windows "7". I thought it was pretty obvious.

Re:High version numbers

[OffaMyLawn]

I don't know, I kind of like it. Could lead to even more ridiculous sounding arguments among friends. Probably seems funnier in my head though...

Re:High version numbers

[Anonymous Coward]

And Chrome copied Opera's UI design.

Re:High version numbers

[hedwards]

The big deal, is that you're watering down the number system. Changing it to one that conveys no meaningful information at all. I get that the folks over at Google like to be disruptive, but this is retard stuff. You increment the major number when you break or modify backwards compatibility, make substantial changes to the way the program functions or if it's been a long time since the last upgrade. Making minor releases into major releases just confuses everybody and removes any hints about how much caution one ought to be using when upgrading.

Re:High version numbers

[hedwards]

The issue is that up until relatively recently there was some degree of agreement about roughly what a numbering system should look like. It wasn't prefect and it wasn't universally accepted, but you could be relatively sure that if you were hitting the 1.0 release that it should be relatively stable and feature complete. That a 1.1 release shouldn't require retraining or make any significant changes to the way the program was used or operated. An Alpha release wouldn't be feature complete typically, but a beta release should and a release candidate had better be in the ball park.

The reason for that is that if you're offering these things up to the public, then courtesy dictates that you give them some hint as to what state the code is in. Release notes are nice, but I don't think that it's a good idea to waste people's times looking at the release notes, if they know that using release code isn't OK in their environment.

Google OTOH, is using a revision system that's in keeping with their asinine perma-betas that they like to have. For a situation like that it makes some sense, but for organizations that realize the impact that beta code has on people, it's a stupid version naming scheme to use.

Re:High version numbers

[Chris Mattern]

That was my first thought. I use Firefox as my primary browser, and, at the moment, I am running version 4 absolutely nowhere, because it only released a couple of weeks ago! And now we're gonna get 5 in two months? 3 lasted us nearly three years--3.6 alone was the latest and greatest for over a year. C'mon, guys...

Re:High version numbers

[Chris Mattern]

When you had Windows 98 and Windows 95, you had a amazing number of users who thought Windows had model years, like cars.

"I have Windows 97"

Re:High version numbers

[Necroman]

But version numbers have nothing to do with features. Google threw a lot of money at Chrome to get it where it is today. Firefox moving to the Chrome version scheme is a marketing move only.

Re:High version numbers

[LordLimecat]

No, theyre moving to a "ship more major versions" model, since major version releases are where features are added.

Theyre simply shipping new features more often.

Re:High version numbers

[juancnuno]

Emacs is at version 23.3

Re:High version numbers

[LordLimecat]

More clunky than "Linux 2.6.31-5"?

Is it possible the number will be less important as they ship more often, especially if they implement automatic update?

Re:High version numbers

[Lennie]

Actually, NT3.x uses the same graphical design as Windows 3.x;
Windows NT 4 uses the same as Windows 95
Windows 2000 (NT5) the same like Window 98. Well kind of.
And Windows XP was for both home and office. (NT5.1) (Server side: Windows 2003)
Windows Vista was NT6 (Server side: Windows 2008)
Windows 7 ironically is NT6.1 (Server side: Windows 2008 R2 ?)

Re:High version numbers

[Lennie]

Actually, I would say Firefox copies the UI from Opera. Firefox Sync actually existed as a seperate Firefox extension from Mozilla before Chrome added cloud sync.

The Firefox developers just want to release more often so they are changing the because there are many, many HTML5/CSS features which are now only implemented not at all or half (same for any other browser). These are just very large specs.

So they need people to upgrade frequently to get that code out to users (thus webdevelopers) and to get attention from users they are doing something for them to keep marketshare ?

Re:High version numbers

[Lennie]

My guess would be they frequently break API's, that is why they frequently release a 'major' version.

Re:High version numbers

[Lennie]

Ubuntu also has these strange code names (Apple is almost as bad) which to me sounds more like someone using the Firesomething add-on.

Re:High version numbers

[Lennie]

I think the PR will be about features, like it always mostly has been.

Like:
"Firefox just released their new version with features X and Y" (so maybe not even a version number in that title)

Re:High version numbers

[ratzetob]

Sigh. In my company management asks for rolling out version 4 which still gives issues in production and now this ...

Re:High version numbers

[O( inf)]

What do you mean by "can't change the fonts"? It's Options -> Under the Hood -> Web Content -> Customize Fonts for me here (Chrome 10 / Win7). Did they remove it in 11?

Re:High version numbers

[Lennie]

Actually Firefox 4 has had a lot of speed improvements, among other things start up and size. It is now efficient enough to run on your mobile. Even though the features (and webtechnologies) support increased tremendously.

Re:High version numbers

[ColdWetDog]

We've got > 2E6 UIDs. Nothing is that obvious.

Re:High version numbers

[darthdavid]

For a while they tried to level up with XP...

Re:High version numbers

[ColdWetDog]

Just look at NVidia. Their driver package is up to version 270.51.

I thought that was the file size in GB.

Re:High version numbers

[mrmeval]

It's becoming inebriated pablum like Gnome. So far I've had several friends try and reject Firefox 4 and ask me if it's a joke on someone's part.

Why are they trying to be as dumbed down as M$?

Re:High version numbers

[metrix007]

Chrome did not invent the majority of those features. It's main innovations are in security, per process tabs and plugins, and significantly slimming down the user interface.

Of course, this seems to be at the cost of have even basic configurability, a decent extensions framework and an extreme lack of configurability. I'm aware I mentioned configurability twice, but that's because it's extremely jarring to see such a great browser technically that doesn't let you customize it in the slightest.

Re:High version numbers

[Haeleth]

No, theyre moving to a "ship more major versions" model, since major version releases are where features are added.

This is untrue.

Firefox 3.5 added features. The major version number did not change.

Firefox 3.6 added features. The major version number did not change.

Firefox 4.0 made fundamental changes to the entire interface.

There is no reason why this new version could not be called 4.1, or 4.5 if they think the new features are really big changes. Switching to a completely different numbering model is just a marketing move. Indeed, it is a step backwards, since we can no longer distinguish versions that add a few features from versions that involve fundamental changes to the way the entire interface works.

Re:High version numbers

[sdstuart]

For your wife, the size queen, yes.

Re:High version numbers

[Gilmoure]

Maybe if they changed the numbering scheme to using the year it's released?

Re:High version numbers

[IB4Student]

No, Chrome is going to run into the hurdle first, and after they fix it via a name or numbering scheme change, Fx will surely just follow suite ;)

Re:High version numbers

[no known priors]

1.0 normally means feature complete for such projects. The programs will often be perfectly valid, bug-free, stable etc. However, because they don't yet implement all the features that the developers want, they aren't yet called 1.0. I believe with Inkscape that they want to have support for the full SVG 1.1 standard before they put the version number at 1.0. See the Inkscape Roadmap.

Also, numbers are the easiest way to distinguish between two different versions of a program. Unless you would rather switch to version control hashes?

Re:High version numbers

[tuxicle]

They dropped it now - it's just Altium Designer 10. Their old version system confused me at times, since the Winter 09 edition was older than Summer 09 - seasons are flipped in Oz.

Re:High version numbers

[akayani]

Firefox 69 wins by head over the competition.

Re:High version numbers

[xnpu]

Developers should work with internal build/version numbers. The public version number is a marketing tool. Using it otherwise would be a waste of a marketing opportunity.

Assuming there will be a version 81 isn't very smart on your part. It's like expecting a brand to stick to pastel colors forever because that was their theme last year. It doesn't work that way. Engineering logic may inspire marketing decisions, but no more than that.

Re:High version numbers

[ProfessionalCookie]

Firefox 42 on the other hand...

Re:High version numbers

[ProfessionalCookie]

That makes a lot of sense. I think I would shorten the year for the next 89 years (my kids can deal with it after I'm gone) and add a leading zero to releases prevent confusion: YY.xx Build numbers would go 11.xx.bbbb but would get truncated for release. We could increment xx only on release as well, then anything with a build number is beta.

Re:High version numbers

[ProfessionalCookie]

YY.xx.bbbb

• Year: my kids can revise this in 88 years.
• Release number with one leading zero, incremented on release date.
• Build number: Only exists for non-final releases, truncated on release date.

I think that's a pretty sane, simple, informative and intuitive system.

Re:High version numbers

[devent]

Yes, I can change the font there. But the webpage is still using Arial and TimeNewRoman and not my font. In Firefox is a checkbox "Allow web pages to chose their own font instead of my selection above". I don't want that web page are choosing their own font, I want that every web page using the same font so that my eyes don't have to adjust every time I switch the site.

Also the font size in Chromium don't work. The page is first shown with the default font size, after the page loads Chromium adjusts the page's font size to my settings. Ah no, that was the MiniumFontSize addon. Chromium still doesn't have a minimum font size settings. So Chromium is totally unusable to me because every page's text is just too tiny and I don't know how to change it.

Re:High version numbers

[AmiMoJo]

Nah, I think it just because they love fucking with the UI and can't justify breaking people's workflow in a point release. By calling it Firefox 5 they can make big changes like Chrome does.

We should start a betting pool on what feature will be removed next. My money is on the stop/reload button.

Re:High version numbers

[atisss]

Cool: Windows 7 6.1.7600 Can't wait for Windows 6.22, it may finally put the end to it :)

Re:High version numbers

[atisss]

I wonder if 2009 Winter is later than 2009 Summer or other way around.
Normally winter starts at November, but the coldest is in February :)

Re:High version numbers

[hardwarefreak]

Of all the stupid features from Chrome to pick up, the version numbers is, by far, the dumbest. Has anyone considered how stupid a version number in the high double digits might be? Firefox 81 seems kind of clunky, doesn't it?

Yes, it does. However, "Firefox 451", "Firefox 666", and "Firefox 911" (as in Porsche, not terrorism) seem kinda catchy to me, so maybe they should skip double digits and go straight to triple.

Re:High version numbers

[LordLimecat]

3.5, 3.6 and 4.0 were all major versions. 3.5 was seen as a part-way big change, and 3.6 a further improvement on the JS engine. It wasnt really a major version in the way 2.0-3.0 was.

Their versioning scheme right now makes very little sense; its completely arbitrary. I think Chrome rather gets it right-- a new major number on the stable branch means "new features have been added", while the "point" numbers mean "security and bug fixes". Incremented "point" numbers on the beta and dev branches mean "new features we're baking in in prep for stable".

Re:High version numbers

[DanielSmedegaardBuus]

I don't give a flying fuck how many digits are present, what do you mean? I thought we were talking about software?

Re:High version numbers

[Jesus_666]

Up until now there were a few "major" releases like 3.0, 3.5, 3.6 and 4.0. It was well known in advance that these would be "major" and that they would break compatibility with some old extensions because the GUI/API/XPCOM would change. Extension authors tested their extensions with the next major version's betas once they were out and everything was fine since apart form these releases there was usually no breakage.

Now we have a new major version every couple weeks, nobody knows whether Firefox 7 will break compatibility with Firefox 6 and extension authors will have to do a lot more testing to see if te next major version is supported. I expect a lot of false negatives here as authors don't have the time to check for every new major version or stop developing altogether, leaving theri extension compatible with "up until 6.0".

I'll just assume that a) Firefox uses a rolling, version-less release scheme and b) all extensions and APIs will be compatible forever. I'll enforce the latter assumption by running the compatibility tester extension and enabling everything regardless of what the extension says.

Re:High version numbers

[Jesus_666]

I generally turn it off because not having the tabs on top makes the window smaller by a pixel or two on platforms where Fx isn't allowed to replace the window decorations.

Of course Chrome isn't any better, what with wasting about twenty vertical pixels on empty space. It does get better when you use a butchered ChromeOS as a browser, though, as with OS they actually realized that heroically saving space by hiding the menu bar doesn't work when you immediately turn around and waste the saved space.

I think that Chrome should add an optional Fennec-like interface. Having no always-present GUI elements is maximally efficient and Fennec puts everything just one scroll away.

Re:High version numbers

[Luckyo]

Not sure if it's relevant, but some of the latest nvidia drivers have been atrocious, and in general, driver quality has nosedived. Great example is when an AAA title that every major gfx card manufacturer got to test ages before release doesn't work on nvidia cards says a lot.
I'm talking about dragon age 2 in case you lived under the rock last couple of months.

Before it used to be ATI that failed with drivers, and nvidia was the trusty one. Now, it's the other way around. Funnily, ATI's drivers are still v8.

Re:High version numbers

[jonadab]

> On my netbook

Okay, but that's a special case. Smartphones, being an even more extreme case, don't show scrollbars or toolbars or anything. That's all well and good, as long as it stays confined to said special devices, but as a desktop user I shouldn't be inconvenienced by that.

Re:High version numbers

[jonadab]

> not having the tabs on top makes the window smaller by a pixel or two on
> platforms where Fx isn't allowed to replace the window decorations.

There are window managers that allow the application to replace its own window decorations?

Seriously? WHY? I cannot imagine anyone EVER wanting that.

I just turn off "tabs on top" because it's the most moronic thing ever. I mean, why on earth would I want all the browser and browser-extension chrome (toolbars and so on) moved down inside the page content area? Also, why would I want to have to move the pointer a couple hundred pixels farther up out of the page area to get to the tabs?

The only reason anyone would use tabs on top, near as I can tell, is because it's the default. Firefox did it is because they were copying Chrome, which had suddenly become popular, but I'm absolutely certain *that isn't why* Chrome suddenly became popular. (Speed might've been part of why. I'm certain aggressive advertising on popular Google web services is part of why. There may be other reasons as well, why Chrome got popular so fast. Tabs on top isn't among them.) I don't know why the Chrome team put tabs on top, but my first guess would be some developer (or their manager) wanted to change things up and make Chrome's UI look different from other browsers just for the sake of being different. (Some people mistake new differences for innovation, even if the specific new differences in question are obviously completely superfluous and useless.)

Re:High version numbers

[Jesus_666]

There are window managers that allow the application to replace its own window decorations? Seriously? WHY? I cannot imagine anyone EVER wanting that.

You don't replace them, you just create a borderless window and then draw your own widgets in there. Chrome does this; at least I know it does on Windows and Linux. IMO borderless windows should be restricted to full-screen windows; at least I can't think of a scenario where it's beneficial to forego the WM's decoration handling.
Look to Windows for a platform where rampant ignorance of UI consistency is the norm: Lots of apps are borderless with self-provided (and self-themed) interface elements. To someone who values consistency that can make a Windows installation with lots of those apps rather ugly and in some cases the application's iconography can be confusing. But hey, what's a bit of ergonomy if we can shove more of our corporate design down the user's throat?

I mean, why on earth would I want all the browser and browser-extension chrome (toolbars and so on) moved down inside the page content area?

The idea behind that is that the URL bar affects only the current tab, thus it logically belongs inside the tab. That's already the case as far as the browser is concerned; URL bar content has been tab-bound since tabbed browsing took off.

Of course that doesn't change the fact that tabs-on-top isn't everyone's cup of tea. It's especially unfitting if you set up your browser so that pages opened through the URL bar open in a new tab, in which case the URL bar as a navigation tool deinitely isn't bound to the current tab. Thankfully, the Mozilla devs made tab-on-taop easy to turn off.

Re:High version numbers

[jonadab]

> You don't replace them, you just create a borderless
> window and then draw your own widgets in there.

Same difference. Either way they'll be wonky and non-standard. They won't fit with whatever visual theme the user has selected, and, heck, the buttons probably won't even be in the right places. Who would want that?

> Chrome does this

Its window decorations look normal to me here. They even change their appearance when I change my window manager's settings, just like every other window. (I'm using sawfish. I have no idea why the Gnome folks switched the default to metacity, but it probably has something to do with their apparent desire to make Gnome noticeably worse and worse with every passing version.)

> IMO borderless windows should be restricted to full-screen windows.

Okay, the root window, sure, but that's different. It's not borderless because some hair-brained application decided to replace its usual decorations with funky special ones. It's borderless because it's the root window. Something similar could be said for panels and docks (I have gnome-panel running, for example, and the panels don't have window decorations), but again those are not application windows; they're part of the desktop environment. It wouldn't make sense for them to be decorated like an application window.

> Look to Windows for a platform where rampant ignorance of UI consistency is the norm:
> Lots of apps are borderless with self-provided (and self-themed) interface elements.

And yet, the user's ability to customize anything is strangely absent. So, yeah, *you* look to Windows if you want. I'll be over here doing my level best to forget about it.

> The idea behind that is that the URL bar affects only the
> current tab, thus it logically belongs inside the tab

I ain't buying it. In the first place, they didn't just move the URL bar down. They moved *everything* down into the page content area: the entire navigation bar, the links toolbar, any other toolbars that happen to be installed, ... In the second place, only *certain* widgets on the navbar are related specifically to the current tab (mainly, the really old ones that have been around since Mosaic). Other things on the same toolbar do things like show or hide various toolbars, open new tabs or sets of tabs or even whole windows, fire up add-ons, and who knows what else, limited only by what extensions the user decides to install. And that's ignoring the possibility that the user might have the location bar set up to open new tabs.

No. They didn't do tabs on top because the location bar logically belongs inside the tab. That's a sophistry somebody thought up to defend it when people complained. The reason they did tabs on top is because Chrome has tabs on top and somebody thought copying unimportant superficial features of Chrome was a good idea just because Chrome had picked up a whole bunch of new users all of a sudden, and Firefox usage has plateaued.

> Thankfully, the Mozilla devs made tab-on-taop easy to turn off.

That will change. They make controversial new features easy to turn off at first, removing them from the preferences dialog in a later release under the guise of usability, and then subsequently removing them from about:config without apology because due to some unrelated controversial new feature the option stopped working.

Have you tried turning off the stupid "bookmark tabset replaces the content of the current tab" feature in recent versions? Haha. Dataloss? What are you talking about? The contents of your tabs are not data, they're, umm, well, this is the way it's designed to work now, mkay?

Re:High version numbers

[Jesus_666]

Same difference. Either way they'll be wonky and non-standard. They won't fit with whatever visual theme the user has selected, and, heck, the buttons probably won't even be in the right places. Who would want that?

Companies who really want their app to look unlike anything else the user has installed. Right off the bat I know that Razer does this with their mouse settings panel for Windows. Some virus scanners too (Avast being an example). Steam too.

Its window decorations look normal to me here. They even change their appearance when I change my window manager's settings, just like every other window. (I'm using sawfish. I have no idea why the Gnome folks switched the default to metacity, but it probably has something to do with their apparent desire to make Gnome noticeably worse and worse with every passing version.)

I use XFCE with XFWM and Chromium does provide its own window decorations there. And yes, the buttons are in the wrong place and look wrong. Maybe Sawfish is more restrictive about its decorations.

Okay, the root window, sure, but that's different. It's not borderless because some hair-brained application decided to replace its usual decorations with funky special ones. It's borderless because it's the root window. Something similar could be said for panels and docks (I have gnome-panel running, for example, and the panels don't have window decorations), but again those are not application windows; they're part of the desktop environment. It wouldn't make sense for them to be decorated like an application window.

Good catch. Panels definitely don't need window decorations.

And yet, the user's ability to customize anything is strangely absent. So, yeah, *you* look to Windows if you want. I'll be over here doing my level best to forget about it.

Unfortunately, as other platforms become more popular, we might see some of the silliness spill over. One instance would be Steam for Mac, which looks just like Steam for Windows, which looks nothing like a Windows or Mac application. If there was a Steam for Linux you could be certain that it would be doing its best to ignore your desktop's look-and-feel.

No. They didn't do tabs on top because the location bar logically belongs inside the tab. That's a sophistry somebody thought up to defend it when people complained. The reason they did tabs on top is because Chrome has tabs on top and somebody thought copying unimportant superficial features of Chrome was a good idea just because Chrome had picked up a whole bunch of new users all of a sudden, and Firefox usage has plateaued.

I think it went like this:
Chrome devs: We now have our tabs on top! It's the wave of the future!
Other browser devs: Why?
Chrome devs: Er... It's because everything else is per-tab anyway?
Other browser devs: Ooooooh. That sounds like something that makes sense. Wait, what about toolbars?
Chrome devs: What's a "toolbars"?

Have you tried turning off the stupid "bookmark tabset replaces the content of the current tab" feature in recent versions? Haha. Dataloss? What are you talking about? The contents of your tabs are not data, they're, umm, well, this is the way it's designed to work now, mkay?

Truth to be told, I just use Tab Groups Manager and have completely outgrown the need for bookmarks. (I just put all tabs related to a concept in a group and hibernate the group.) I think Tab Mix Plus fixes the issue, though. Just like Status-4-Evar fixes the missing status bar and Element Properties fixes the missing "Properties" context menu icon. Admittedly, it's annoying that you have to install extensions to retain useful functionality while pointless things like Personas get baked into the browser, but then again you can do it. If they turn off the normal tab position I'll either install an extension that puts them back on top or one like Tree Style Tabs that provides its own freely-positionable tabs.

Why..?

[elfguy]

It used to be versions were about feature sets. If you added a small feature to a program you'd increment the minor version, if you added big features you'd release a major update. The idea of having versions increase on specific dates seem weird.

Re:Why..?

[Anonymous Coward]

Marketing

Re:Why..?

[Kpt Kill]

Maybe to catch up to IE 9? Nine is higher than four, so regular users think that IE must be more mature and feature complete. We'll see what happens when Microsoft decides to change to IE XP or some other version identifier that Mozilla won't be able to copy.

Re:Why..?

[Bloodwine77]

Then Opera is in the lead at version 11 Of the major browsers, only IE and Opera should legitimately have high version numbers due to how long they've been around. I know Firefox comes from Netscape stock, but Firefox is a different beast than Netscape. I do understand Firefox not wanting to look like they are behind due to their version number being small, but I'd have more respect and understanding if they went to a different versioning scheme altogether instead of inflating minor releases with major version numbers.

Re:Why..?

[Grygus]

I question whether computer illiterate users actually think that way. Ask one what version of Office they are running, and they will reply, "Microsoft." I think that whole premise is deeply flawed.

Re:Why..?

[Abstrackt]

Wouldn't make make more sense for Mozilla to release Firefox 2011 then?

Re:Why..?

[thePowerOfGrayskull]

It's actually always been pretty arbitrary. There's no committee that defines standard version number practices; and as such they're all over the board. From the countless 0-dot releases of open source projects, to Chrome's "major" release number. In the end, it doesn't really matter to anyone except developers. In reality each release contains changes from the prior release. Some releases introduce incompatibilities - but those are documented anyway, and don't gain much by using a special numbering to indicate it. (Especially because no two project can agree on that numbering scheme.) By using whole numbers, you're reducing confusion from the typical major/minor/minorer*/minorest* pattern. A version is a release of the software - end of story.

*poetic license #312002110033121

Re:Why..?

[Jugalator]

It used to be versions were about feature sets. If you added a small feature to a program you'd increment the minor version, if you added big features you'd release a major update. The idea of having versions increase on specific dates seem weird.

It's for many reasons - especially to cut down on the QA wind-down time that keep stalling the trunk (less features at a time means less time spent winding down and testing - a long time doing that means web standards will race ahead before you've even tested the version you were working on... the other extreme causes the "Internet Explorer effect" - often outdated before it's done, and not because the devs suck, but because the releases are too rare which causes a crapload of testing requirements for each release).

So this is a more "organic" model that should be able to follow new web standards better, and the needs and wants of Mozilla's user base.

Google explained all this pretty well too when they also moved to this model:

http://blog.chromium.org/2010/07/release-early-release-often.html

Re:Why..?

[jonadab]

That's debatable, especially if you take into account the amount of development that took place from one version to the next. Firefox 2 was arguably a more mature release, with a larger number of major releases preceding it, than IE7. (Opera I'll grant, though. It's been continuously maintained since the days of Trumpet Winsock, so big version numbers are warranted there.)

IE basically skipped versions 1 and 2 (they were minor feature-incomplete dev milestones; normal users never saw them), and even versions 3 and 4 were not feature-complete compared to other browsers of the time (notably Netscape). They just pumped the number up real fast so people would *think* it was equivalent to Netscape 4. Granted, 5.5 could arguably be considered worthy of major-version-number status. Still, being *very* generous, major IE releases are 3.0, 4.0, 5.0, 5.5, 6.0, 7.0, 8.0, and now 9.0, for a total of eight, max. A less magnanimous assessment might peg it at more like six without being completely unfair.

Meanwhile, when development on the Mozilla codebase (that eventually became Firefox) started, IE was only about three or four years old. Then after the release of 6.0 the IE team at Microsoft was completely disbanded and NO significant development was done for several years (until finally it was so antequated that Microsoft was legitimately concerned they might lose ALL of their browser userbase if they didn't get off their tails and make IE look somewhat less like using stone knives and bear skins). If you throw out the years when browser development at Microsoft had completely ceased, I'm not at all sure that the IE codebase has been developed for more years than the Mozilla codebase.

That brings up another point: the codebase that gave rise to what we now call Firefox has changed version number schemes and application names repeatedly, but starting from around Mozilla 0.8 or so it was essentially feature-complete and stable (compared to the other browsers available at the time, particularly IE). If you count from there, major releases with significant new features include Mozilla 0.9, 0.9.5, 1.0, 1.5, and 2.0, then Phoenix 0.something, a couple of Firebird releases, and then there were a couple of Firefox releases *before* 1.0...

Firefox 2 was a MUCH more mature release than IE7, with I would say a larger history of preceding major releases. Okay, the UI got a big overhaul in the aviary move, but for that matter the IE7 UI doesn't look much like IE6, either. The rendering engine is built on the same codebase in both cases, so I would argue that it's basically a contiguous development history.

Granted, there haven't been a lot of improvements *since* Firefox 2. A small handful of new CSS features (of which about three are any practical use) and a couple of perf improvements -- and a whole raft of seriously undesirable "let's screw up the UI for no good reason until nobody can stand it anymore, then let's do it some more" nonsense, plus a couple of major new stability bugs. Meh.

Re:Why..?

[VGPowerlord]

Then Opera is in the lead at version 11 Of the major browsers, only IE and Opera should legitimately have high version numbers due to how long they've been around. I know Firefox comes from Netscape stock, but Firefox is a different beast than Netscape. I do understand Firefox not wanting to look like they are behind due to their version number being small, but I'd have more respect and understanding if they went to a different versioning scheme altogether instead of inflating minor releases with major version numbers.

No, Chrome is in the lead with Chrome 12.

But, like you said, IE and Opera are the only ones who should legitimately have such high number.

Re:Why..?

[LordVader717]

It's always been more about practicality than logic. Thanks to the wonders of positional notation we're able to increment arbitrarily until we decide it's time to go for a new number.

Re:Why..?

[Fred Or Alive]

I'd personally count the stable version of a browser (which seems to be Chrome 10) as the current one. But it's all a bit moot, as Opera will probably lose the "lead" in this somewhat meaningless contest shortly.

Re:Why..?

[jez9999]

In the end, it doesn't really matter to anyone except developers

Developers, developers, developers, developers!

Actually one of the truer things Ballmer has said. MS keeps a bunch of developers on board and hence Windows is successful. Firefox risks pissing off a load of extension developers at its peril. I'm an extension dev and the new numbering system pisses me off. Incrementing maxVersion every 3 months after being expected to fix API-breaking changes? What a fucking PITA.

Wordperfect vs Word

[Anonymous Coward]

This just reminds me of when Microsoft Word for Windows jumped from version 2.0 to 6.0 just to appear competitive with WordPerfect. This will make version numbers irrelevant and nigh pointless.

Re:Wordperfect vs Word

[Xtifr]

An even better example is the Watcom C/C++ compiler, where the initial release (in 1988) was v6.0, to suggest it was competitive with compilers from MS and Borland (which it was--more than). Version numbers have been irrelevant and nigh pointless for decades. For an interesting example of the opposite effect, where the version numbers are much lower than one might expect, consider the Linux kernel, which still hasn't reached v3.0 after nearly 20 years. How long ago was Win3.0 released? (And how long was the period between Win1.0 and Win3.0?) Or consider the X Window System, which made it up to v11 fairly quickly, but has stalled out there for longer than Linux has even existed.

I guess my point is that if you're trying to derive any sort of useful information from version numbers, without investigating how a particular project uses version numbers, you're simply wasting your time.

Re:Wordperfect vs Word

[IICV]

And it's not like Microsoft has stopped doing that. Why do you think it was the Xbox 360 that went up against the PS 3, and not the Xbox 2?

Re:Wordperfect vs Word

[DarkVader]

Oh yeah, Word 5. The last good product to ever come out of M$.

I'd almost forgotten how bad 6 was. And it's never gotten any better, just more bloated - and now we're up to 14.

And the Windoze version is even worse - what an abomination of a user interface that bloatware is!

Re:Wordperfect vs Word

[Hooya]

This is a versioning system I can understand.

Re:Wordperfect vs Word

[flimflammer]

That's just silly. The product name has nothing to really do with the version number. Just because Sony happens to increment the number after Playstation on the product name doesn't mean anyone else has to (or does for that matter)

Color versioning!

[Anonymous Coward]

Firefox Red/Blue, Firefox Yellow, Firefox Gold/Silver!

or maybe Gemstone based versioning like Firefox Ruby/Sapphire!

Re:Color versioning!

[supersloshy]

Firefox Black/White would cause a lot of racism complaints though, so I'm not sure it would work out in the long run... Oh, and you forgot Firefox Green, the codename for the Beta versions ;)

Re:Color versioning!

[slapout]

Chrome is a Pokemon -- just look at the logo!

Re:Color versioning!

[AndrewNeo]

Huh? Red and Green were the versions that were released in Japan, Blue was the replacement for Green in the US.

Re:Color versioning!

[_0xd0ad]

I always thought it was a game of Simon...

Re:Color versioning!

[Fallen Andy]

For the sake of the flying spaghetti monster i hope no marketting drone read what you just suggested

Makes me glad i'm a semi retired developer - (please no more...(grins)

Andy

Re:Color versioning!

[jackchance]

No, it's definitely Samus' Morph Ball

Re:Color versioning!

[supersloshy]

Yes, red and green were the original versions. However, Blue was released later in japan as a bug-fixing special release. America didn't get any re-releases besides Yellow because the original versions in America had the bugs of the Japanese version fixed.

Re:Color versioning!

[surveyork]

Gotta catch 'em all!

Re:Color versioning!

[Jesus_666]

The central question about Firefox 5 will of course be: Team Proto Man or Team Colonel?

Well

[jav1231]

I hope it's better than 4.

Re:Well

[matrim99]

It's 1 better than 4.

It's Official. Firefox has jumped the shark!

[Ron Bennett]

After many years of Firefox being a solid, well performing browser, it appears it's going to end in bloat just like Netscape did...

And, history is set to repeat itself again with MSIE 9 having more to offer than Firefox just like MSIE did back years ago when people dumped Netscape in droves.

I'm sticking with Firefox 3.6x for as long as I can, and then when push comes to shove, just like back in the Netscape days, will likely switch to MSIE 9 unless a decent fork of Firefox comes along before then - a decent fork probably already exists, but not aware of what they are.

Many people just want a fast, compact, featured (Chrome falls short in that regard; privacy issues) browser - Firefox offered that - shame to see it becoming buggy bloatware.

Ron

Re:It's Official. Firefox has jumped the shark!

[Bert64]

Run the open source build of chrome - chromium, need not worry about privacy issues then.

Re:It's Official. Firefox has jumped the shark!

[Jugalator]

After many years of Firefox being a solid, well performing browser, it appears it's going to end in bloat just like Netscape did...

Why do you think this? More frequent updates != more bloat. The updates are also spent on fixing bugs. You can't say this before you've seen what Firefox 5 will end up covering. Sure, if it'll become much slower and crashy, I don't question you, but we've so far seen no indication that this more frequent schedule of releasing bug fixes and now less features at a time will lead to that...

Re:It's Official. Firefox has jumped the shark!

[edmicman]

What does IE9 offer that FF4 does not, and over Chrome for that matter? From what I've read it loses out in at least extensibility and speed compared to the others.

Re:It's Official. Firefox has jumped the shark!

[M. Baranczak]

Seriously. The last thing FF needs is more features. They should be optimizing the features that are in there already, instead they're wasting time on useless crap like Twitter and Digg integration.

Re:It's Official. Firefox has jumped the shark!

[Luckyo]

There is even worse thing to come with this: add-on compatibility.

As it stands now, for many like myself the only reason to use FF over chrome lies in vastly superior add-ons. But every time a major version upgrade comes, a good half of add-ons remain broken for .

In this regard, it would be much better if major version revisions were more RARE rather then more common. While this change will not drive me to competing browsers, I will likely end up staying with 3.6.x for a long time in large part because of it.
Other issue being the retarded interface and added bloat that I do not want, just like parent. Heck, I didn't want most of the features that came in various upgrades to 3 (personas was quite possibly the dumbest thing to actually force down everyone's throats - what was wrong with keeping it as an add-on?). And I end up disabling stuff like awesome bar, and if I ever upgrade to 4.x I will absolutely require a way to rearrange tabs in the way I like them (i.e. 3.6.x way).

Re:It's Official. Firefox has jumped the shark!

[BenoitRen]

After many years of Firefox being a solid, well performing browser, it appears it's going to end in bloat just like Netscape did...

Except that Netscape wasn't bloated. Version 6 was released when the code wasn't stable, and that was the end of it.

Re:It's Official. Firefox has jumped the shark!

[Zerimar]

This is totally subjective. Extensibility it clearly loses out to everyone - Microsoft and it's corporate customers do not want it to be extensible, so I doubt it will be. In terms of speed, I think Chrome/Firefox/IE/Opera are all fast enough. They all load quickly, render pages quickly, execute Javascript quickly - a lot of this is thanks to Chrome pushing that envelope, but everyone else seems to have caught up by now. For the first time in a long time, I am of the opinion that all of the major browsers for Windows are good enough. It basically comes down to personal preference at this point. I typically use Chrome as my primary browser and IE9 for the handful of random sites that don't quite work properly with Chrome.

Re:It's Official. Firefox has jumped the shark!

[Reapman]

Why on earth would you stay with 3.6? 4.0 is faster on both my Core 2 Duo Desktop and older Macbook - I've heard of no reason why 3.6 is better other then some people prefer the old interface, which as has been posted in countless /. comments is easily changed back if that's your thing.

You obviously don't remember the days of Netscape - it died for several reasons, not least of which was it was slow and painful and crashed all the time. Hell FF4 Beta on my desktop rarely crashed, the Final release hasn't yet (although I'm sure it will eventually, all software does)

Slow, painful, crashes a lot. 3 things FF4.0 is not. IE9 doesn't offer the plugins I use to make web browsing safe, and I'm not using the offspring of IE6, something I despised using at work until a few months ago.

Re:It's Official. Firefox has jumped the shark!

[Billly Gates]

Firefox 4 is quite an improvement on my Windows 7 workstation over Firefox 3. There is always Chrome as well which is multiplatform and webkit based. I have not tested it on Linux yet but I am about too. Hardware acceleration is nice. Launch times are fast and tomshardware.com benchmarked IE 9 vs Firefox 3 and while IE looked better with 3 or 4 tabs the situation changes when all the tabs are maxed with heavy duty pages. Firefox had less bloat and memory usage

Re:It's Official. Firefox has jumped the shark!

[Billly Gates]

UPDATE:
I have an old laptop 2007 era but it has 3d acceleration enabled. I have to say Firefox 4 for Linux is mediocre and lacks hardware acceleration. Chrome is smooth in comparison on Fedora Linux 13 when browsing slashdot. I go to www.msnbc.com which has huge javascript and graphics. I went into about:config with Firefox and enabled WebGL and it made it slower?.

Ok so maybe you are right with Firefox4 but it is an improvement over 3.6 on Windows. I wonder how the MacOSX version is? I may stick with IE 9 and Chrome.

Re:It's Official. Firefox has jumped the shark!

[Legion303]

Is the MR Tech plugin supported on 4.0 yet? I'm on 3.6 and like the fact that I can kill addon compatibility checking all at once, without having to manually edit files on every plugin I know still works. I installed 4.0 on my wife's new Win7 system and MR Tech was still unsupported as of last week.

(Yes, I can update its max compatibility string by hand, but as deeply as it integrates with FF, I'm afraid that will critically break something.)

Re:It's Official. Firefox has jumped the shark!

[Billly Gates]

How is hardware acceleration on Firefox 4? I was dissapointed in Firefox 4 in Linux. In Windows Firefox 4 is an improvement but the graphics seem to stutter more than IE 9. Firefox admitted not all of the HTML components are running in hardware yet. Hopefully 5 will fix that.

Anyway I assumed openGL or the lack of was the problem and the Mac would be an issue

It is because Developers love free cake

[McGruber]

Cake is even more yummy when it comes from Microsoft!

Please no

[Haedrian]

That's just horrible.

Its easy nowadays - "You use Firefox 3.6 or 4.0" ?
I'm NOT remembering a string of numbers.

If they want large numbers, maybe they should take a tip from ubuntu.

You could get "Firefox 11.6" out in 5 months. See? Big number.

Re:Please no

[thePowerOfGrayskull]

That's kind of the point (at least as far as I'm concerned - not that I'm following the internals of this, just applying logic). Why should users have to worry about x.y at all? I'm on version 6. You're on version 8. End of story.

Re:Please no

[Jugalator]

Why would you need to remember all numbers?

I don't need to remember that I'm using Chrome 10.0.648.204 stable. Chrome 10 is often more than enough. Usually "Chrome since the last year" suffice to give a good idea of the web standards it's supporting, for someone who follows the latest standards developments like a webmaster.

Re:Please no

[Merk42]

The point is the question, if any, will be "You use Firefox?"

Very impressive

[supersloshy]

As much as I dislike uselessly high version numbers, their release plan looks rather impressive. If you didn't bother to click the article link, the development for each new version of Firefox will occur in a rather "layered" fashion. After initial development on Firefox 5 is finished in mozilla-central, work begins on the initial development for Firefox 6, and after that, 7. At the same time as 6 and 7 are being worked on in mozilla-central, Firefox 5 and 6, respectively, are moved to mozilla-experimental and worked on. While experimental work is being done on 6 and 7, Firefox is moved into "Beta" and Beta work begins on 5 and 6, respectively. It's much easier to understand if you look at the image I linked to, and it certainly looks impressive, though I am not quite sure how it will work out in the long run. Only time will tell, I guess.

Re:Very impressive

[Cloudgatherer]

It is the ever so popular release train model. It sounds really good on paper, and management usually jumps on the bandwagon, however this type of release model is suited towards web deployments, NOT desktop applications. The scheduling looks very slick, doesn't it? It does until you realize you are just doing costly releases at regular intervals while the big ticket items usually end up getting delayed to future versions, and what you release is of low quality because everyone is in a hurry to "make the train" so they can jump to the next iteration. The pretty charts and graphs have schedules, but no features on them, and that's the issue. A release should be planned to deliver a set of features, not planned to release on day X regardless of what is in it (maybe nothing!).

For a web app, it works fairly well. For desktop software (which all browsers are), it is a bad idea.

They just don't get it...

[plastick]

I hope it's better than 4.

I couldn't agree more! I wish they would fix the bloat and listen to their user base instead of adding more features we don't want (what a crazy idea).

Re:They just don't get it...

[JonySuede]

what bloat ?
  On my system it takes less memory and it seems faster than any other browser except Firefox 1

Re:They just don't get it...

[JonySuede]

Firefox 1 (is only true for the memory part, not the speed, nor the brokenness of the web)

Why release early?

[Tolkien]

Take the time to fix bugs and release something even better.

Re:Why release early?

[Jugalator]

Because releasing slower requires more time testing the release instead of fixing bugs.

Re:Why release early?

[Tolkien]

Okay. Test more, then. How's that bad? :P

Re:Why release early?

[Tolkien]

I didn't mean never release. I meant don't release earlier than originally planned.

Re:What?!

[IB4Student]

They do not have any of the features of Fx 5 implemented in them, yet. Right now they are mostly just bugfixes and improvements that are going to be released in a few weeks with the minor Fx 4 update. And 4.1 is taken by mobile. And minefield build numbering is automated so you probably should not base anything off of them (e.g. Fx 4 RC was always labeled beta 13pre in the nightlies).

Patrick Volkerding should be in charge

[slackzilly]

Then the latest Firefox would be realeased when it's ready to be released. Come to think of it, he should run the world.

Re:Patrick Volkerding should be in charge

[m50d]

But the next major release would be firefox 7

Re:Patrick Volkerding should be in charge

[Jon Abbott]

As long as we don't have a repeat of "Slackware '96"-style naming, I'm all for it. :^)

Re:Patrick Volkerding should be in charge

[slackzilly]

Maybe Pat IS "Bob" :-O

I don't see a year

[jonescb]

I just see the date June 21, but which year?
Jun 21, 2014 perhaps?

I wonder how many gigs of ram I'll need

[plastick]

I really do hope they fork the code. The developers seem too arrogant to listen to the myriads of cries from the people who actually use their browser.

The new version 4 already was taking almost 400 megs of ram. That's acceptable?? Hello?

Re:I wonder how many gigs of ram I'll need

[dr. chuck bunsen]

I'm not doubting you, but just wanted to point out that, at least on 64bit Linux, I've never seen FF4 get much over 200MB. That seems pretty reasonable to me.

Re:I wonder how many gigs of ram I'll need

[BinarySolo]

Give Pale Moon a try (http://www.palemoon.org/). It's a fork that's basically like an optimized Firefox (note: only for Windows). I only found it a couple days ago but so far it seems to be a lot better with memory than Firefox 4.

Re:I wonder how many gigs of ram I'll need

[maxume]

I regularly have it over 400 MB. But the later releases of 3.6 and 4.0 have been a lot more stable in memory use than 3.5 was.

Re:I wonder how many gigs of ram I'll need

[dr. chuck bunsen]

So, after posting about my Firefox RAM experience I loaded up the latest stable Chrome, the latest stable Opera, and of course FF4. I loaded them each up with more tabs than I would ever use, each one in Identical order, and then allowed them to idle for quite sometime. RAM Usage results: Firefox 4 - 284 MB Chrome - 312 MB Opera - 345 MB Which at least on 64 bit linux makes it the LEAST RAM hungry of the 3. I would also like to note that even Opera is using a very reasonable amount of RAM considering the number of tabs. I will admit that FF3 was a pretty bad performer, and I, along with most folks switched to Chrome until the release of FF4. But today, with FF4 released, all of the major browsers performance are so close to one another, that it really negates performance as a bullet point when choosing a browser. Just pick your favorite in terms of UI, features, and extensibility, and use it. Right now the browser war is down to personal preference.

Re:I wonder how many gigs of ram I'll need

[kwoff]

They couldn't just, you know, optimize Firefox....

Re:I wonder how many gigs of ram I'll need

[schnikies79]

My firefox 4 experience on OS X. Freshly opened, no extensions installed, 1 tab open browsing slashdot.

402.3mb

Re:I wonder how many gigs of ram I'll need

[dr. chuck bunsen]

Linux seems to be far better then.

Freshly opened, 3 tabs - 2 pinned (gmail, gvoice) plus igoogle = 156.2 MB

Re:I wonder how many gigs of ram I'll need

[dr. chuck bunsen]

So just out of curiousity I went an ran the exact same test on my Macbook Pro, running 10.6, and the latest FF4.

So same test as for Linux, 3 exact same sites open in seperate tabs, 2 pinned = 204.6 MB

So not as good as my Linux box, but certainly no 400 MB. Are you sure you are running latest FF4?

Re:I wonder how many gigs of ram I'll need

[Ash-Fox]

My firefox 4 experience on OS X. Freshly opened, no extensions installed, 1 tab open browsing slashdot.

402.3mb

Apple offers a fix here.

Just wait and see

[zugurudumba]

I really hope they'll make the auto-updates more aggressive, like Chrome does. Otherwise, the Firefox market share will become excessively fragmented.

Chrome and Firefox's Development Process

[kripkenstein]

What actual features and improvements could they possibly have added in "8 WEEKS" since the release that they have had time to actually put through an Alpha test, Beta test, and then full release that would warrant a VERSION 5!?! This seems crazy lame to me. The browser has slowly gotten bloated, now the number? Why?

Hi there, I work on Firefox. First thing, we didn't write the article linked to in the summary, and I don't think they gave a totally accurate description. In fact, I don't even think this was interesting enough for a blog post from them.

We are basically going to switch to a development process that is very similar to Google's with Chrome. So everything you say here is valid about their development practices as well - rapidly rising version numbers for no reason, little features in 'major' releases, etc.

Why are we doing it? There is just one reason, it helps get code shipped faster. Code does not get written faster though, in either Chrome or the new Firefox process :) It just gets shipped quicker. But that is important too, and that's why we (and Google) are doing this.

Basically, Chrome and Firefox will release quickly, with small amounts of changes each time. I agree with you 100% that the major version number rising each time is silly! Personally I would either drop the version number entirely, or use something like Ubuntu's versioning scheme (10.10 for 10th month, 2010). But oh well.

In any case, since you asked what will ship in Firefox 5, I can tell you about stuff I know about (which is platform/backend stuff, not frontend). We have several improvements to performance that should be very useful, in both JavaScript and graphics. In particular WebGL should be faster on some cool demos on Linux, which I am very happy about.

Re:Chrome and Firefox's Development Process

[dstyle5]

For those of us who work on browser-based products for large monolithic corporations this is going to be a gong show. Companies are always looking for guarantees of "Official Support" for browser X, version Y and now that you guys are going to be pumping out new major version numbers frequently that means browser QA/verification is going to have to occur far more often now. Not to mention having to test products against a quickly increasing number of versions.

Re:Chrome and Firefox's Development Process

[judeancodersfront]

But ours goes to 11.

Re:Chrome and Firefox's Development Process

[gmack]

Why are we doing it? There is just one reason, it helps get code shipped faster. Code does not get written faster though, in either Chrome or the new Firefox process :) It just gets shipped quicker. But that is important too, and that's why we (and Google) are doing this.

It may not get written faster but fewer changes each cycle will allow it to be tested and debugged faster.

Re:Chrome and Firefox's Development Process

[k8to]

Yeah, same here.

We'll basically end up saying something like IE 7-9 are supported, along with Firefox and Chrome versions no one in the world uses.

In reality, pretty much all browsers work, but sometimes there are small glitches the customers do not like. Those customers will just have to use internet explorer.

Re:Chrome and Firefox's Development Process

[kripkenstein]

We'll basically end up saying something like IE 7-9 are supported, along with Firefox and Chrome versions no one in the world uses. In reality, pretty much all browsers work, but sometimes there are small glitches the customers do not like. Those customers will just have to use internet explorer.

I think that what will happen is basically that IE and Safari will become the 'stable' browsers that big corporations standardize on, because they change infrequently, while Firefox and Chrome are the 'fast-moving' browsers that are faster, have more new features, etc.

Re:Chrome and Firefox's Development Process

[strikethree]

"Why are we doing it? There is just one reason, it helps get code shipped faster. "

Good morning sir. First, let me thank you for your efforts. Despite the rapid bloat that is emerging, I still use Firefox as my main browser. Again, thank you. I do feel the need to ask you a question:

What purpose does shipping code faster serve?

To be quite honest, if the 2.x series were still receiving security updates, I would be using it despite the memory leaks.

I guess what I am really asking is this:

Why not just improve the code you have rather than writing new code?

Yes, HTML 5 support needs to be added at some point, but surely most of the new additions should be implemented as plugins and the base browser should be lean and mean like it used to be. No?

Again, I sincerely thank you for your efforts in producing the most usable browser out there.

Re:Chrome and Firefox's Development Process

[Ginger Unicorn]

programming against the spec while ignoring the quirks means your site is likely to be broken on everything.

Re:Chrome and Firefox's Development Process

[AmiMoJo]

The problem is that every time you change the major version number the UI usually changes and all add-ons have to be tested and updated. That wasn't too bad when it was only yearly updates and there was time for people to deal with it, but 8 weeks? I only updated my own add-on to FF4 compatibility a couple of weeks ago, and I don't normally bother with beta or RC versions (it just so happened that I bought a new laptop and decided to try it).

I know there is a browser war on and it is very tempting to try to match Google with fast updates and new code, but FF is a rather different beast to Chrome. Chrome add-ons seem to transition from version to version more easily. The UI is much less flexible than FF so you are forced to work the Chrome Way(TM), and thus the changes are less likely to break your workflow or annoy you.

All I'm saying is try to see if from the user's and add-on developer's point of view.

Re:Chrome and Firefox's Development Process

[Tim C]

It's not about programming against a given release, it's about verifying against releases - in fact, we do exactly what you say and programme against the spec (then add in the inevitable tweaks to get it working right cross browser and browser version). That doesn't change the fact that if a client wants it supported in "all versions of IE from 6 up, all versions of Firefox from 3 up" etc, then releasing what would normally be Firefox 4.1 (or even 4.0.1) as Firefox 5 adds another round of browser compatibility testing that would not otherwise be there.

Re:Chrome and Firefox's Development Process

[kripkenstein]

What purpose does shipping code faster serve?

The reason Google do it, and why we are now starting to do it too, is to get code into users hands faster. For example, we had WebM support last year, but only shipped it to users last week. (Beta users had it since last year, but not most of our users.) The same is true for JavaScript performance - we improve it all the time, so the quicker we get it into users' hands, the faster their browser is.

Why not just improve the code you have rather than writing new code?

We do both. Some people do mainly one or the other, others do a combination, like me. For example, I did a lot of work on improving performance, so that is improving existing code, and right now I am working on adding new functionality (IndexedDB in Fennec).

Re:Chrome and Firefox's Development Process

[kripkenstein]

The problem is that every time you change the major version number the UI usually changes and all add-ons have to be tested and updated.

I agree that this will be challenging for addon writers. Part of the solution is the Jetpack API, which is very stable, but not all addons can use it. So I completely understand your concerns here. I am not on the addons team myself, but from what I understand they are trying hard to make this as easy as possible. And there really should not be many UI changes between FF4 and FF5, just like there are hardly any between Chrome X and X+1 - simply because not a lot of time passes between them.

I just want a browser that can filter out the crap

[KarlIsNotMyName]

Don't turn that around on me.

Comment removed

[account_deleted]

Comment removed based on user account deletion

Marketing

[MaWeiTao]

All these versions and frequent updates are good for one thing: marketing. Frequent releases keep the application on the minds of the consumer.

It keeps people engaged, like something's happening even if when it's not. It's not that dissimilar to the Twitter mindset. And with carefully metered feature implementations a company can ensure that there's something to offer in subsequent versions. And with others playing the same game they're probably thinking it's best to hold back on certain features until they know what the competition is up to.

It's inevitable that what we've seen with gadgets would spread to software. Except with an even more absurd timetable.

5.6.44.144?

[Arancaytar]

That address isn't even assigned... :P

Re:Dumb idea

[ZorinLynx]

People who have browser requirements for their online system need to be able to say "Requires an HTML 5 compliant browser" and code their site to work across all modern browsers.

The days of "Netscape NOW!" are long past.

MajorFeature.MinorFeature.Bugfix.build

[Tiger4]

I realize this is petty, but why the rush to bump up the numbers? I mean is the only way to give your product some eye appeal is to give it a bigger flashier new number? Of course I'm assuming there is some kind of defined (and designed) spec being worked on here. Every time you implement a feature in the spec, you tick up the MinorFeature number. You write a new spec with more Stuff in it, you tick up the MajorFeature number.

But maybe not. Maybe there is no spec and no design. People just keep gluing stuff on whenever they feel like it and push it out the door when it doesn't crash (too much). OK, that model is R-ReleaseNumber.Bugfix. Not as pretty, but at least it warns people you are just driving without actually navigating to a goal.

User identification

[saikou]

I think the really important aspect is if users actually remember/care about the version number of browser.
Do they know it's Firefox 3 (16.6?) or Firefox 4?
I don't think Google ever advertised the version number in any significant way.
It's always "Download Google Chrome", not "Firefox 4, Free Download".
If Firefox moves away from major version numbers completely then yeah, call it 5-6-17-293-whatever. It's "Firefox"

I like it

[kelemvor4]

I guess I'm the minority, but I like the straight number version system opposed to things based on date or quarter like some others use. v94 doesn't bother me at all, and neither would v765. It's simple, easy to understand, and can go on forever without a "reset" or change to the version system.

Why mess with what works?

Site support?

[haeger]

Currently my bank doesn't even support FF4. It does work, but I get a nastygram when I login. I suppose that's their way of protecting themselves if the browser should fsck something up on the site. They're quite picky about which browsers they support. They will eventually support almost anything, but they'll have to test them first, and with a new release every three months I think I'll always be out of compliance with this site. And some day I guess something WILL break.
Not fun.

And Chrome is a carbon copy of Safari

[pizzach]

Serously. Look at how similar the guis are. The dev tools haven't even been modified at all, icons and all. I suppose the sincerest form of flattery in imitation or something.

Re:And Chrome is a carbon copy of Safari

[xouumalperxe]

Chrome's UI is vastly different from safari's (as "vastly" as the difference between two browsers can be anyway). The devtools aren't an imitation, they're the same: the webkit inspector. Seeing as webkit provides fully fleshed-out dev tools, why would Google re-implement or re-skin them? Just so they look different?

Re:compatibility nightmares

[indiechild]

If you're writing modern standards-compliant CSS and HTML you shouldn't have a problem. None of the modern browsers have given me trouble for several years now. Life as as a web person is so much easier nowadays. 99% of problems come from IE7 which we still have to support at work, but that's an ancient browser so it's hardly surprising.

Coming this winter...

[Isaac Remuant]

Knowing Firefox, they will release "Coming this Winter! Firefox Summer Edition! (Cause' remember, it's summer in the southern hemisphere)"

Re:compatibility nightmares

[Ash-Fox]

If you're writing modern standards-compliant CSS and HTML you shouldn't have a problem.

Bullshit. Many browsers and websites are built following recommendations set out by W3C instead of the standards published by ISO/IEEE/IEC on CSS, XHTML and HTML. I know for a fact that if I follow even old stuff like ISO/IEC 15445:2000 to the letter, there are still instances where the major browsers won't follow the standards correctly and the same issues exist with the latest published documentation where browsers simply didn't implement the standards.

Re:OH GOOD JAVASCRIPT

[Ash-Fox]

I don't have any of these problems the parents is having normally. The only time I've had memory annoyances with Firefox was when the Skype toolbar or Adblock plus was installed.

Re:Dumb idea

[Ash-Fox]

People who have browser requirements for their online system need to be able to say "Requires an HTML 5 compliant browser" and code their site to work across all modern browsers.

I guess everyone without a time machine would get screwed then, since only HTML5 drafts are available currently to browser developers and website developers.

Re:Firefox was never about high performance

[Jon Abbott]

Firefox was never about speed or minimalism.

Phoenix (which eventually became Firebird and then Firefox due to trademark issues) was immensely faster than the kitchen sink that was Mozilla Application Suite. It was certainly minimalist compared to Mozilla.

Prefer a 2-part number

[GerryHattrick]

Leave me the UI that I use all day (part 1) and make whatever engine improvements are needed (part 2). Do auto-updates on (stable) part 2 but only *offer* updates on part 1. Why should I have to use add-ons to get things back the way I want them?

Re:Firefox 4 is so bloated!

[surveyork]

True that the new interface is very minimalistic, but remember that a few clics will bring you the old interface back. And there is also 'status bar foreva'.

Re:OH GOOD JAVASCRIPT

[surveyork]

Some add-ons are known offenders: https://addons.mozilla.org/en-US/firefox/performance/ You can also try 'Configuration Mania', 'Bar Tab', creating a new profile...

Oh No!

[Errol backfiring]

Frirefox Green is made from people!