|
|
Subscribe / Log in / New account

Mozilla Delivers New Version of Firefox

Mozilla has announced the release of Firefox 5. "The latest version of Firefox includes more than 1,000 improvements and performance enhancements that make it easier to discover and use all of the innovative features in Firefox. This release adds support for more modern Web technologies that make it easier for developers to build amazing Firefox Add-ons, Web applications and websites."

to post comments

Mozilla Delivers New Version of Firefox

Posted Jun 21, 2011 18:33 UTC (Tue) by markhb (guest, #1003) [Link] (9 responses)

Wow... why such a rush for a full version upgrade? They just released FF 4 not very long ago.

Mozilla Delivers New Version of Firefox

Posted Jun 21, 2011 18:34 UTC (Tue) by dlang (guest, #313) [Link] (3 responses)

they decided to move to a new release numbering scheme where every release is a full version upgrade, and they are making a new release every few months.

Mozilla Delivers New Version of Firefox

Posted Jun 21, 2011 19:02 UTC (Tue) by nix (subscriber, #2304) [Link] (2 responses)

And those few people who thought that linking their apps against xulrunner was a good idea will soon change their minds. API/ABI stability? What is that?

Mozilla Delivers New Version of Firefox

Posted Jun 21, 2011 20:04 UTC (Tue) by joib (subscriber, #8541) [Link]

I recall quite a while ago (1 year maybe?) at least Ubuntu was trying to shift apps to webkit due to poor ABI compatibility of xulrunner, so that's not anything new brought on by this version numbering change.

Mozilla Delivers New Version of Firefox

Posted Jun 22, 2011 1:49 UTC (Wed) by kripkenstein (guest, #43281) [Link]

> And those few people who thought that linking their apps against xulrunner was a good idea will soon change their minds. API/ABI stability? What is that?

Gecko has definitely made the decision to focus on moving forward faster, as opposed to supporting stable APIs or ABIs (which WebKit does).

Is that a good or a bad thing? It's good for the reasons it's good Linux doesn't have an internal stable API/ABI, and bad for the same reasons that that is bad.

Mozilla Delivers New Version of Firefox

Posted Jun 22, 2011 1:43 UTC (Wed) by kripkenstein (guest, #43281) [Link] (4 responses)

> Wow... why such a rush for a full version upgrade? They just released FF 4 not very long ago.

Is your concern that a new stable FF was released, or that the major version number was bumped (or both)?

Regarding the first, I think it's a good thing to release stable versions often. Each such version has few new things, obviously, but users get new features and improvements faster. Google started doing this with Chrome and Mozilla is following their example with Firefox, because it makes sense to do. It is sad when your developers just finished something really amazing, but the next major release is a year in the future, so no one (except for alpha testers) will see it anytime soon...

Regarding the second, I do agree that bumping major version numbers so fast is kind of silly. I said that when Google did it, and I say that now that Mozilla is doing the same. But, it is just a number, and with two major browsers now doing it, it won't seem odd fairly soon I guess.

Mozilla Delivers New Version of Firefox

Posted Jun 22, 2011 12:52 UTC (Wed) by markhb (guest, #1003) [Link] (2 responses)

Actually, my concern / surprise was that they had released a new major version so quickly; I don't really follow FF development much now and wasn't aware that they had changed numbering schemes. Now that I've installed it and find that it would probably have been FF 4.0.1 under the older versioning, I'm no longer surprised, but I agree that single-segment version numbering is lame.

And really, it's not "just a number". If you build significant web applications, your QA department needs to be able to define the browsers it will run test cases for. If you can't efficiently do that for FF because the versioning no longer reflects potentially incompatible changes in rendering or scripting capabilities, then the financially responsible choice may be to declare FF to be unsupported. Same goes for Chrome.

Mozilla Delivers New Version of Firefox

Posted Jun 22, 2011 14:31 UTC (Wed) by kripkenstein (guest, #43281) [Link] (1 responses)

> If you can't efficiently do that for FF because the versioning no longer reflects potentially incompatible changes in rendering or scripting capabilities, then the financially responsible choice may be to declare FF to be unsupported. Same goes for Chrome.

I believe that is in fact the case - Firefox and Chrome's frequent releases *can* be incompatible with the one right before them. So technically bumping the major version number is the proper thing to do.

Mozilla Delivers New Version of Firefox

Posted Jun 22, 2011 18:58 UTC (Wed) by markhb (guest, #1003) [Link]

I believe that is in fact the case - Firefox and Chrome's frequent releases *can* be incompatible with the one right before them. So technically bumping the major version number is the proper thing to do.
Yes, now, with the "frequent release" plan (I'm only discussing FF). It wasn't the case before when potentially incompatible changes were held for truly major releases. I saw a thread in mozilla-dev-planning (I think) where an annual LTS release was suggested; hopefully they will pursue that.

I probably should say here that in all my posts here I'm speaking exclusively for myself and not my employer.

Mozilla Delivers New Version of Firefox

Posted Jun 23, 2011 10:15 UTC (Thu) by job (guest, #670) [Link]

It is a false dichotomy to have to choose between stable interfaces and rapid development.

Linux has had a stable API for a very long time yet no one can accuse it of evolving slowly. I believe it is a bad decision to play the version game against Chrome, as xulrunner and the extensions ecosystem is it's biggest feature.

Mozilla Delivers New Version of Firefox

Posted Jun 22, 2011 1:00 UTC (Wed) by Hausvib6 (guest, #70606) [Link]

People were complaining that they're too slow once Google entered the web browser market with Chrome, now.. they're too fast. Anyway this is a great development and there are still numerous other browsers available for those who wants slower pace. With the rapid development cycle of Chrome and Firefox, I think they'll be on the forefront of web standard/feature development.

Thanks to backport and Debian Mozilla team repo, Debian stable user will be able to use the latest Iceweasel. I wonder what will be the version number of Firefox just before Wheezy release... 10? For Chromium... 20?

Mozilla Delivers New Version of Firefox

Posted Jun 22, 2011 7:38 UTC (Wed) by jiu (guest, #57673) [Link]

...And it's very fast! faster than the previous version, the difference being very noticeable on a mac.

Plug-ins?

Posted Jun 23, 2011 12:41 UTC (Thu) by NRArnot (subscriber, #3033) [Link] (1 responses)

What's the Firefox policy on plug-ins now?

It used to be that if a plug-in worked on (say) 3.6.1, one could expect it to continue working on all 3.6.n releases.

I'm still on 3.6 and don't wish to move, until the Tabkit plug-in works with a higher release, or until some other plug-in offers me similar functionality. And I really don't want to see a situations where plugins get broken by 5, then 6, then 7 ... a few months apart.

Plug-ins?

Posted Jun 24, 2011 14:15 UTC (Fri) by gerv (guest, #3376) [Link]

If by plug-ins you mean add-ons, then the policy hasn't changed. However, instead of relying on addon authors to update the "maxversion" of their plugins on addons.mozilla.org, as far as possible this is done automatically by systems which know about what has changed in Firefox and analyse the code for patterns which might have broken. People can also submit feedback from the addons manager about what's working and what isn't.

Then, when the new version arrives, if all of your addons are known compatible you will be upgraded, and if they aren't, you'll be warned about the issue.

Gerv


Copyright © 2011, Eklektix, Inc.
Comments and public postings are copyrighted by their creators.
Linux is a registered trademark of Linus Torvalds