Standards Make Rapid Software Releases Workable

jfruhlinger writes "There was a bit of a kerfuffle when the Mozilla Foundation's community coordinator brushed aside concerns from enterprises that Mozilla's rapid release schedule clashed with organizations' need to carefully vet software upgrades. One thing that could bridge the gap between these worldviews is a widespread adoption of open standards. After all, if IE 6 dealt with web pages in a standard way, it wouldn't have been so painful to keep it around as long as it lurked on many corporate desktops."

Comments

Version numbers

[cgeys]

Firefox's release schedule isn't any more "rapid" than it was before just because they now change major version number instead. It's just taking away the real problem and trying to be push your software to the version numbers that long term projects like IE and Opera have got over the years. Same problem with Chrome.

Re:Version numbers

[Lennie]

You are right, it isn't the version numbers. What they do now have is a more rapid release cycle. But also they don't have security updates for Firefox 4 after the release of Firefox 5.

Re:Version numbers

[Wrath0fb0b]

Firefox's release schedule isn't any more "rapid" than it was before just because they now change major version number instead. It's just taking away the real problem and trying to be push your software to the version numbers that long term projects like IE and Opera have got over the years. Same problem with Chrome.

Tell that to an addon developer, where the churn of compatibility-breaking changes (many for no apparent reason) is causing a real headache.

There is a promised SDK to land sometime this summer. We'll see if Mozilla can deliver a stable API for more than a few months.

Re:Version numbers

[Anonymous Coward]

But, the whole point is that they don't have to do this. The complaints that people "stop bitching about version numbers" is missing the point. *Even if* that is a valid view, Mozilla still has *no reason* to do it. It is pissing people off for absolutely no added benefit.

Re:Version numbers

[hitmark]

Similar to the hoopla regarding Torvalds shaving a vestigial number form the Linux version by going 3.x.

I have long wondered why various projects, if they are not maintaining a stable branch for fixes, do not simply use a single rolling number to indicate a new release.

Re:Version numbers

[hairyfeet]

Actually if you read the Moz dev's blog when the stink first hit he said they are pushing for a six week turnaround so I honestly don't see how you could say it is less rapid. We aren't talking minor point releases here, we are talking about extension breaking serious alterations to the underlying code.

Watching the backlash after this stupidity I'd say there was one clear winner...Chrome. After pulling this BS I noticed the extensions for Chrome started going up as developers started jumping ship. Most of these developers are either working for donations or on their spare time and trying to keep their code working with a 6 week release schedule simply isn't possible and the Chromium codebase doesn't seem to be broken with their extensions with regards to updates. I myself have upgraded from Comodo Dragon 6 through to Comodo Dragon 12 with ZERO broken extensions, compared to more than a third of my extensions broken on the single update from 4 to 5 with Firefox.

And finally lets get to the meat of the matter...testing. With a schedule so fast testing simply is NOT possible, not on extensions, not on the codebase itself. With the push to Intranet and Internet based applications having no way to see if your organization is gonna end up crippled by an update really is inexcusable. And when they talk about "standards" everyone's bullshit o' meter should shoot off the charts. What standards? HTML V5 is still a draft at best yes?

Final prediction...the Google juggernaut will crush Firefox thanks to their own stupidity. Many people use at home what they learned at work and no admin with a brain is gonna touch FF now. Chrome offers .MSIs and GPO controls and don't seem to be planning any major ripping out of the internals (We saw here a few months ago one of the devs talking about how there was gonna be serious work on the memory and CPU usage, which will mean major guts ripping) and more importantly you can write an extension for it and it "just works" between updates.

The only thing FF had going for them anymore was the extension framework IMHO. The GUI was more and more becoming a Chrome ripoff, in most performance tests both Chrome and Opera usually stomped all over them, the only real difference was the extensions. by adopting this "fuck you, get on board or piss off" attitude they have succeeded in running off a LOT of extension writers to the greener pastures of Chrome. Out of the extensions I use the ONLY one left exclusive to FF is NoScript and I hear the Chromium guys are working on giving the NoScript guy an API that will give him the hooks he needs.

So goodbye Mozilla, it was great while it lasted but after breaking a third of mine and my customers extensions I have removed FF from my standard installs and replaced it with Dragon. You are quickly becoming like Netscape before you, arrogant while having a codebase not able to back up that arrogance. So goodbye Moz and thanks for all the fish.

Re:Version numbers

[mrmeval]

I've stopped proselytizing open source to any business or user. Mozilla's attitude is wide spread. Businesses and users want to get work done. They don't want proselytizers for the one true way who can't be bothered to understand their needs.

Has the GUI prophets managed to get cut and paste working? Is sound working? Is 3D graphics working? Am I going to be forced to use firefox as this prophet dictates? How much of the usability I expect will be ripped out of some project this week?

Re:Version numbers

[Anonymous Brave Guy]

But the people complaining don't want Firefox 4.1, they want Firefox 4.0.1 - aka fixes for security holes and other serious bugs, but with minimal chance of incompatibility

That's exactly it. The whole version numbering thing is a complete red herring. The point is that with such a rapid release cycle, and with the failure to distinguish between bug fixes and new feature releases/UI changes, it is no longer possible to aim at a stable, secure, standardised browser platform within an organisation if you rely on Firefox as your browser.

So, If the browser functions to "standards", .....

[aix tom]

.... anyway, what the hell do they change from version to version?

If they tell you "Changes are not *dangerous*, because we stick to standards", then that is bullshit. If a change is "not at all dangerous" then it is also "not at all necessary", since it would imply the change does not change anything. What I have seen in 15 years in IT is that even some pretty minor thing that changed in a software product can bring your work flow to a halt. And you can lose business for hours or days.

Add-ons

[tepples]

They wouldn't have security updates for Firefox 4 separate from Firefox 4.1 and you wouldn't complain about it but instead of calling it Firefox 4.1 they've called it Firefox 5

The problem here is that extensions don't automatically work on the next major version, especially if they have a component written in a language other than JavaScript.

Missing the point...

[Junta]

Sure, implementing standards in *theory* should mean the browser choice doesn't matter. The problem is the difference between theory and practice. You think you write in standards, but you only validate that in one browser, you may accidentally not be standards compliant. Conversely, you may fairly be totally standards compliant, but a browser defect results in your site not behaving correctly. Or a standard could be sufficiently vague as to have multiple implementations vary in behavior without being able to point at any particular one as non-compliant.

All this is ignoring that things like browser crashes, memory exhaustion, and security issues are critical issues to worry about that generally have no bearing on standards compliance.

If standards meant the choice and version of a browser wouldn't matter, then why would there be a choice of browser and version in the first place?

Just aim for standards? This meme needs to die.

[Anonymous Brave Guy]

Don't aim for one or a few browsers, aim for standards.

I'm sorry, but that argument doesn't get any more sensible as more people parrot it.

For one thing, there are no standards that cover a lot of the newer technologies yet, and if you're going to force updates every few weeks then "This is in beta and is subject to change" just doesn't cut it.

For another thing, even if there were, standards are only ever a means to an end, and that end is producing useful tools that help people get things done. Firefox can push for trendy new standards all it likes, but it's not written by super-human developers who can avoid regressions, and those regressions hurt.