|
|
Subscribe / Log in / New account

Mozilla to Businesses: We're Not Interested (PC Mag)

Here's a PC Magazine article taking a sharp look at the apparent disconnect between the new Firefox development process and the needs of businesses. "[Asa] Dotzler's comment is both sneering and contemptuous of the businesses that have deployed Firefox in their organizations. And, sometimes, at their own peril, may I add. While Firefox is a wonderful browser with its own unique set of features, the frequent updating, occasional lack of good documentation, extension breaking whenever a new update comes out - it all makes it a dicey choice of browser for businesses."

to post comments

Editor's comment

Posted Jun 25, 2011 19:26 UTC (Sat) by corbet (editor, #1) [Link] (49 responses)

[I didn't want to put this on the article itself]

One could argue that Mozilla has chosen to adopt a process which is quite similar to that used by the Linux kernel - major releases on a fairly frequent basis. What's missing is an organization which plays the role of the distributors, supporting older releases and backporting fixes when needed. That is an expensive effort, of course; it's not clear how it might be funded now that Mozilla has shown even less interest in funding that work itself.

Editor's comment

Posted Jun 25, 2011 20:07 UTC (Sat) by rahulsundaram (subscriber, #21946) [Link] (11 responses)

I thought about the exact same thing. It is important to realize however that if a third party organization decides to do this work, they will likely have to fork the project and rename it and at that point, it might be possible and even likely that the organizations which are risk averse are unlikely to adopt it anyway because of the level of stickiness to known brands. This is a similar problem for Libroffice as well because there is higher level of usage of Openoffice.org in Windows and without the presence of distributions, there isn't a push to Libreoffice. This is something IBM recognizes and will want to bank on.

Editor's comment

Posted Jun 26, 2011 15:03 UTC (Sun) by mingo (guest, #31122) [Link] (10 responses)

Well, "business" has a standardized way of expressing its business needs: cold, hard cash.

If true "business interest" wants Mozilla to care (i don't think PC Mag putting up an inflammatory article counts as such), it needs to fund the features it wants to emphasise - and stabilization/backporting work is such a feature. (and it's a very expensive one)

Business interests are doing this for the Linux kernel: distributions are funding the -stable effort, directly and indirectly as well - despite Linus not wanting to do this himself originally. They are doing it to such a degree that most of the active kernel developers care about it thus stabilization and backwards compatibility has become a development process component as well.

For a kernel it also makes sense technically: it's much easier to develop and test the kernel if a strict ABI is maintained. A browser is not a kernel though and browser plug-ins are more similar to binary kernel modules and the Linux kernel project emphatically does not support backwards compatibility with binary kernel modules. Extensions are borderline but i think the Mozilla Project has the right to do that as long as it considers it in its own best technical interest. I personally don't mind starting out a new Firefox version bare-bones and adding extensions only when i really need them.

The only truly hostile approach Mozilla could embark on here is if it actively destroyed stabilization/backporting efforts - i don't see this happening here - they simply don't have the resources to care themselves. So if business wants Mozilla to care, it needs to fork over either some cash to the Mozilla Foundation or needs to pay some Firefox developers that care.

"Expecting" Mozilla to be "business friendly" with binary plug-ins and extensions is in itself a fairly one-sided affair - why should Mozilla care if the other side does not? Mozilla should only care about the attributes driving its code, it should care about its developers and about its users.

So looking at this from a kernel project perspective i think Mozilla made the right call here: care about those who care about you (quid pro quo), and those who care about Mozilla right now are either using your latest (stable) version or are at least not whining about your earlier versions ...

In any case "businesses" should not really be surprised if they see a competent negotiator as a counter-party.

Editor's comment

Posted Jun 26, 2011 17:06 UTC (Sun) by rgmoore (✭ supporter ✭, #75) [Link] (9 responses)

A browser is not a kernel though and browser plug-ins are more similar to binary kernel modules and the Linux kernel project emphatically does not support backwards compatibility with binary kernel modules.

I don't think the analogy is especially apt. The kernel takes an indifferent attitude toward ABI compatibility largely because they want to force developers to push their modules upstream. Firefox is equally indifferent to ABI compatibility, but they are much less willing to accept extensions into the mainline tree. That combination is a much more hostile attitude toward out of tree code than the kernel takes.

The only truly hostile approach Mozilla could embark on here is if it actively destroyed stabilization/backporting efforts - i don't see this happening here - they simply don't have the resources to care themselves.

But Mozilla is fairly hostile toward stabilization/backporting efforts- at least ones that want to call themselves Firefox. If you want to maintain an old version and call it Firefox, you have to get permission from Mozilla for each and every code change you make. That's why Debian calls their version "Iceweasel" instead; they don't want to be in a position where they have to wait for Mozilla's approval before they can release a critical security update. It's unreasonable to complain about lack of resources to maintain old versions and then throw up arbitrary stumbling blocks- ones that use your supposedly scarce resources, BTW- to other people who try.

Trying to maintain control over the project is OK, but Mozilla can't plausibly try to maintain ironclad control over their project and then cry about lack of resources to do the job right. They can't call their project user friendly when their default is to break their users' configuration periodically with automatic, incompatible updates. What good is it to add the latest features that few web sites are using if the cost is breaking old features that you know your users want because they have gone out of their way to enable them?

Editor's comment

Posted Jun 26, 2011 19:57 UTC (Sun) by mingo (guest, #31122) [Link] (2 responses)

Now yours is probably a fair argument to make - and if enough developers share your sentiment it's a powerful force the Mozilla Project should better not ignore.

(Sidenote: the fact that IceWeasel has not taken over the browser market by storm seems to suggest that at least for the time being most developers are behind Firefox, right?)

In any case, businesses should not "expect" the Mozilla guys to care about them just for the sake of being business-friendly.

Editor's comment

Posted Jun 27, 2011 17:02 UTC (Mon) by rgmoore (✭ supporter ✭, #75) [Link] (1 responses)

Sidenote: the fact that IceWeasel has not taken over the browser market by storm seems to suggest that at least for the time being most developers are behind Firefox, right?

It shows that Debian was far sighted enough to go to the trouble of rebranding when most other distributions weren't. I suspect that Firefox's mandatory upgrade policy is going to make every other distributor that wants to maintain its own browser upgrade schedule instead of being dragged along by Firefox will be strongly considering following suit.

Consider Fedora, for example. They released Fedora 15 about a month ago, and the version of Firefox they shipped with is no officially obsolete and unsupported upstream. Since they are trying not to upgrade important packages during a product's lifecycle, they're faced with the unpleasant choice of making an exception for Firefox; living with the need to get Mozilla's approval for any update, including critical security updates, in FF4 for the next year; or trying to replace FF4 with a rebranded version partway through Fedora 15's life. Something tells me that there will be no Firefox- though there may be an Iceweasel or similarly rebranded version- in Fedora 16 unless FF changes its update policy.

Editor's comment

Posted Jun 27, 2011 17:09 UTC (Mon) by rahulsundaram (subscriber, #21946) [Link]

There is no basis for such assumptions. Firefox 5 is now available as an update for Fedora 15 and it is will within the policy for updates with no explicit exceptions required since Firefox 5 despite the major version change really is just a incremental update that doesn't change the end user UI drastically or anything like that.

http://fedoraproject.org/wiki/Updates_Policy

I would be very much be willing to bet that Firefox is going to stay with no renames or forks by default.

Editor's comment

Posted Jun 26, 2011 20:23 UTC (Sun) by kripkenstein (guest, #43281) [Link] (4 responses)

> But Mozilla is fairly hostile toward stabilization/backporting efforts- at least ones that want to call themselves Firefox. [..] It's unreasonable to complain about lack of resources to maintain old versions and then throw up arbitrary stumbling blocks- ones that use your supposedly scarce resources, BTW- to other people who try.

I understand that from the perspective of people using the code, the trademark stuff is an annoyance. It isn't an arbitrary thing though, it is done for very good reasons (I am sure you already know the reasons I mean, if not I am happy to explain myself). And it is just the name.

> Trying to maintain control over the project is OK, but Mozilla can't plausibly try to maintain ironclad control over their project and then cry about lack of resources to do the job right.

This isn't control over the *project*. It's control over the *name*. The code is fully open source and not controlled by anyone.

Compare that to other major browsers, which are either partially or entirely closed source, and even when they are open source, are often not developed in the open like Firefox (and closed development is a form of control).

Editor's comment

Posted Jun 27, 2011 11:38 UTC (Mon) by geofft (subscriber, #59789) [Link] (1 responses)

If names are things important enough that Mozilla feels the need to protect theirs to such an extent, isn't it obvious that names are definitely things important enough that any serious downstream redistributor of Firefox would want to continue calling their redistribution by the same name?

It can't be "just a name" for one person, and so important to file for trademark protection and actively defend it for another.

Editor's comment

Posted Jun 27, 2011 14:54 UTC (Mon) by kripkenstein (guest, #43281) [Link]

You make a fair point, if names were meaningless, they can't be important enough for someone to care about trademarking.

But this isn't symmetrical, as you imply. Mozilla has been releasing Firefox for a while, and people expect something from it. If someone else releases something they also call 'Firefox', but it is extremely buggy and unstable, Mozilla is hurt more than the other party. People will blame 'Firefox', not the other party.

Of course, it does matter to the other party as well. Calling it 'Firefox' helps people know that it is closely related to the Firefox they already know.

So I agree with you that names are not meaningless. However, they have different meanings to Mozilla and to people using the name.

Note that if another party calls their browser 'SnowBear, a browser based on Firefox', then there is no trademark issue, and the other party also gets most of what they want, since people will understand it is in fact closely related to Firefox. Compromise is possible here.

Editor's comment

Posted Jun 27, 2011 16:09 UTC (Mon) by nye (subscriber, #51576) [Link] (1 responses)

On the other hand, compare it to the majority of free software which does not have that requirement - modified versions of Linux are called 'Linux'; modified versions of Apache are called 'Apache'; modified versions of KDE are called 'KDE'. All of these are trademarked, and control over that trademark is occasionally exerted, but not in such a sweeping, blanket fashion - and the sky didn't fall.

(Okay, arguably the KDE project should have asked Kubuntu's first few KDE4 releases to be called something else; Kubuntu really did give a lot of people a falsely bad impression of KDE4, so maybe I've defeated my own point...)

Editor's comment

Posted Jun 27, 2011 21:49 UTC (Mon) by kripkenstein (guest, #43281) [Link]

> On the other hand, compare it to the majority of free software which does not have that requirement - modified versions of Linux are called 'Linux'; modified versions of Apache are called 'Apache'; modified versions of KDE are called 'KDE'.

I think there is a very big difference between Linux, Apache and KDE on the one hand, and Firefox on the other.

Linux, KDE and Apache are used by tech people like you and me. When we try some Linux distro, if we see it fail in some embarrassing way, we know it isn't *Linux*'s fault. It's almost certainly the distro.

Firefox however is a consumer product. Hundreds of millions of people use it, almost all of them *not* tech people like us. If they download something called 'Firefox' and it is horribly buggy, they will think that Firefox is to blame, and not place their trust in the Firefox name anymore.

Linux of course is used by many consumers. But they basically never hear that name. They hear Android, TiVo, Kindle, etc. My parents don't know what Linux is, but they know what Firefox is, even though they use both.

Historical trivia: Early in the Mozilla project, the plan was to not ship a consumer product at all. It was to just ship code, and let others make the final products. In other words, exactly like the Linux kernel works. But this didn't take off so a shift was made, and Mozilla started to ship Firefox as a consumer product.

Editor's comment

Posted Jun 28, 2011 19:12 UTC (Tue) by kripkenstein (guest, #43281) [Link]

> I don't think the analogy is especially apt. The kernel takes an indifferent attitude toward ABI compatibility largely because they want to force developers to push their modules upstream. Firefox is equally indifferent to ABI compatibility, but they are much less willing to accept extensions into the mainline tree. That combination is a much more hostile attitude toward out of tree code than the kernel takes.

The analogy is not exact, but I think it is closer than that. One factor is that addons.mozilla.org is in between being in the mozilla codebase, and being entirely separate. If your addon is in there, it will get automatically checked for version compatibility issues, and either get its allowed versions either bumped, or a warning shown.

That isn't the same as being in the mozilla codebase itself, of course. But it practice almost all addons can be automatically verified as compatible, so if your addon is on addons.mozilla.org, your life as an addon developer is much easier.

The bigger issue is for addons that aren't in addons.mozilla.org. Mozilla has no access to them, so it can't help ensure their compatibility. But Mozilla does encourage that everyone put their addons in addons.mozilla.org, just like the kernel encourages developers to push their code upstream. (Again, it isn't an exact analogy, but there is some general similarity.)

Editor's comment

Posted Jun 25, 2011 20:49 UTC (Sat) by bronson (subscriber, #4806) [Link] (9 responses)

Except that the Linux kernel is very good about not breaking userspace.

For me, the Firefox 5 upgrade broke all plugins except one. That rather suddenly changed my whole browser experience. And, I must say, I don't find stock Firefox very compelling compared to Chrome.

So, unless Mozilla gets serious about not trashing plugins every 3 months, I wonder if I'll continue using it.

Editor's comment

Posted Jun 25, 2011 21:21 UTC (Sat) by Julie (guest, #66693) [Link]

unless Mozilla gets serious about not trashing plugins every 3 months

How about every six weeks? :-)

------
(from the comments below the blog)

--- Simon June 23, 2011 at 9:25 pm #
I *do* follow Planet Mozilla, which is how I spotted this blog post. But even following that, I believe it wasn’t at all clear just how quickly FF 5 was going to be arriving. Plenty of talk of a shorter release cycle, but I was anticipating that to mean annual releases, or bi-annual at the most. Not every three months!

Reply
Asa June 23, 2011 at 10:10 pm #
It’s actually going to be 6 weeks :-) ---


------
(from the comments below the blog in reply to someone else)

--'Asa June 23, 2011 at 5:54 am #

As for John’s concern, “By the time I validate Firefox 5, what guarantee would I have that Firefox 5 won’t go EOL when Firefox 6 is released?” He has the opposite of guarantees that won’t happen. He has my promise that it will happen. Firefox 6 will be the EOL of Firefox 5. And Firefox 7 will be the EOL for Firefox 6. - A' --

Editor's comment

Posted Jun 25, 2011 21:53 UTC (Sat) by dlang (guest, #313) [Link] (1 responses)

the problem is that there is not really an extension interface for firefox, extensions are allowed to muck around with anything they want in the firefox internals.

it's impossible to have this freedom and stability across versions (or security against bad extension writers for that matter), but it does allow extension writers great flexibility to do anything they want.

hopefully the end result of this (after the dust settles, which will take a while), is going to be a proper API for extensions to use that is stable across versions (and while they are at it, they may add some security as well)

Editor's comment

Posted Jun 26, 2011 8:57 UTC (Sun) by oblio (guest, #33465) [Link]

Jetpack: https://addons.mozilla.org/en-US/developers/

They thought of this. But it will take time to be adopted fully, of course.

Editor's comment

Posted Jun 26, 2011 2:24 UTC (Sun) by cesarb (subscriber, #6266) [Link] (5 responses)

> Except that the Linux kernel is very good about not breaking userspace.

I would say that, for a browser, the equivalent of userspace is the "HTML space" where web pages run. AFAIK, they are good enough about not breaking that.

Plugins and extensions would be more similar to kernel modules.

Editor's comment

Posted Jun 26, 2011 4:39 UTC (Sun) by jrn (subscriber, #64214) [Link] (4 responses)

> Plugins and extensions would be more similar to kernel modules.

Right, and the Linux kernel developers are pretty good about not breaking in-tree kernel modules and about creating a path for out-of-tree modules to find their way into the tree.

Editor's comment

Posted Jun 26, 2011 6:56 UTC (Sun) by drag (guest, #31333) [Link] (2 responses)

It doesn't really matter.

Businesses will use extensions when they need to. This is one of the things that makes Microsoft Office so powerful and so useful for many businesses.

Do you think that accountants use special proprietary software to run banks and such? Hell no. They use Excel with a load of visual basic with a connection to some big database somewhere backing it all up.

That's sort of thing is why software with extensions is so powerful. With the proper extensions you can use Firefox as a IDE and a dozen other things.

If your going to break extensions all the time then it's pointless to even have them. Nobody can use them for anything serious.

They will just use something that takes end user's needs seriously... which would be Chrome or Internet Explorer.

Editor's comment

Posted Jun 26, 2011 8:58 UTC (Sun) by oblio (guest, #33465) [Link] (1 responses)

They are providing an additional extension mechanism, with an API: http://lwn.net/Articles/449269/

Editor's comment

Posted Jun 28, 2011 18:03 UTC (Tue) by xtifr (guest, #143) [Link]

Might have been nice if they'd finalized that and waited until it was widely deployed before pulling the support rug out from underneath their users.

I know, I know, ask for my money back if I don't like it. :) Actually, I'm already using Iceweasel instead.

Editor's comment

Posted Jun 26, 2011 16:04 UTC (Sun) by bronson (subscriber, #4806) [Link]

Once Mozilla bundles plugins with Firefox downloads, and discourages out-of-tree plugins, THEN I will agree that they are similar to kernel drivers.

Today, though, plugins and kernel drivers are quite different.

Editor's comment

Posted Jun 25, 2011 21:19 UTC (Sat) by ejr (subscriber, #51652) [Link] (6 responses)

Another issue is that any such distributor cannot use the Firefox trademarks.

Editor's comment

Posted Jun 26, 2011 1:52 UTC (Sun) by kripkenstein (guest, #43281) [Link] (5 responses)

> Another issue is that any such distributor cannot use the Firefox trademarks.

Not without permission. But just like Ubuntu can use the Firefox trademarks, some company offering long-term support for Firefox should not have a problem getting Mozilla's cooperation, I assume.

Editor's comment

Posted Jun 26, 2011 15:27 UTC (Sun) by mattdm (subscriber, #18) [Link] (4 responses)

> Not without permission. But just like Ubuntu can use the Firefox trademarks, some company offering long-term support for Firefox should not have a problem getting Mozilla's cooperation, I assume.

Actually, I very much doubt it. Long-term support would run counter to the "everyone get the latest and greatest!" policy they want to encourage. Having an older codebase also selling itself as a currently-supported Actual Firefox doesn't fit in.

Editor's comment

Posted Jun 26, 2011 20:25 UTC (Sun) by kripkenstein (guest, #43281) [Link] (3 responses)

I see what you are saying, but I still believe the advantages outweigh the disadvantages you mention, and that Mozilla would go for it.

Editor's comment

Posted Jun 26, 2011 21:00 UTC (Sun) by ejr (subscriber, #51652) [Link] (2 responses)

They did not. This is what Debian proposed, and Mozilla wasn't interested.

Editor's comment

Posted Jun 26, 2011 21:09 UTC (Sun) by kripkenstein (guest, #43281) [Link] (1 responses)

It's more complicated than that.

For one thing, trademarked logos and so forth are not DFSG friendly, for example, so even if Mozilla did agree for Debian to use it, Debian couldn't include it by its own self-imposed (not a criticism, btw) rules.

In the current case, we are talking about some company selling support for old versions of Firefox. I don't see why Mozilla wouldn't agree to the use of the brand Firefox with that, under the usual conditions, which are that all the branding is used, and modifications to the code are acceptable to Mozilla. In this case, of course all the branding is used, and all the modifications are security fixes, mainly backported from Mozilla's later codebases anyhow.

This was unacceptable to Debian, and again I am not criticizing Debian - they have very high criteria for the DFSG, which I applaud. Both Debian and Mozilla had a valid position, but a perfect agreement was just impossible. The solution in that case, to ship all of Mozilla's code with just a small branding change, seems perfectly fine to me. But in the current situation, selling support for older versions of Firefox, there aren't the same limitations and the brand should be usable.

Debian Firefox

Posted Jun 27, 2011 4:27 UTC (Mon) by jrn (subscriber, #64214) [Link]

> trademarked logos and so forth are not DFSG friendly

Restrictive trademark policies are obnoxious but the DFSG is not about them. From <http://glandium.org/blog/?p=1650>:

> It is still not certain that this will actually lead to Debian shipping
> something called Firefox some day, but things are progressing, even if
> at a rather slow pace, and I have good hope (discussions are promising).

So do not be deterred by what you thought was true. There is a bug filed on bugzilla.mozilla.org tracking progress toward trademark permission and some day it might be a footnote in history.

Editor's comment

Posted Jun 25, 2011 21:55 UTC (Sat) by mattdm (subscriber, #18) [Link]

Given that they have a distribution-hostile trademark policy, they're not really encouraging that part of the ecosystem anyway.

Editor's comment

Posted Jun 25, 2011 22:40 UTC (Sat) by pboddie (guest, #50784) [Link] (3 responses)

That is an expensive effort, of course; it's not clear how it might be funded now that Mozilla has shown even less interest in funding that work itself.

As others have pointed out elsewhere in response to this article, such an organisation doing the support, maintenance and so on should not necessarily be doing that work for free. If the "needs of businesses" are to be met, those businesses should be prepared to pay, just as they expect their customers to pay for whatever it is they do or provide.

Editor's comment

Posted Jun 26, 2011 21:51 UTC (Sun) by jd (guest, #26381) [Link] (2 responses)

A fair comment, but remember that they're competing against Chrome, IE, Opera, Safari and other free-as-in-beer browsers. If they try to charge for unbroken ABIs, they'll be dead in a week. Microsoft is already exploiting this for all they're worth and I would not be surprised if IE gains some of Firefox' marketshare. Suicide, even with good intent, is still suicide.

Editor's comment

Posted Jun 26, 2011 22:42 UTC (Sun) by butlerm (subscriber, #13312) [Link] (1 responses)

If they try to charge for unbroken ABIs, they'll be dead in a week

Two things: (1) Charging for supported long term stable versions does not prevent Mozilla from giving away short term auto-upgrading versions. (2) If no one is willing to pay for a long term stable support, then these complaints are just a bunch of whining. Businesses that want this need to put their money where their mouth is. Otherwise they can just stick with IE on Windows.

And Internet Explorer is in no sense "free". It is an operating system feature, one that probably consumes a healthy part of Microsoft's development budget, one where Microsoft has of late been spending an extraordinary amount of effort trying to preserve backward compatibility. It is not like you can get a free-as-in-anything version of Internet Explorer for any other environment.

Similarly, if there is a market for a stable version of Firefox, it ought to be similar to the market for a stable version of desktop Linux. I don't know what Redhat does in desktop sales, but they of all people ought to be interested in long term stable version of Firefox or something like it. That is what they do.

Editor's comment

Posted Jun 27, 2011 16:16 UTC (Mon) by nye (subscriber, #51576) [Link]

>And Internet Explorer is in no sense "free".

Er, there's no charge for it. How is that not the criterion for 'free-as-in-beer'?

If your argument is that you need Windows to run it then my reaction to that is twofold:
1) By the same token, no software at all is free since you need hardware to run it
2) To all but a near-zero minority of people, Windows is free. It comes with the machine, and you can't get an equivalent machine without it for less. Indeed, that often costs extra.

Editor's comment

Posted Jun 26, 2011 1:59 UTC (Sun) by kripkenstein (guest, #43281) [Link] (4 responses)

> What's missing is an organization which plays the role of the distributors, supporting older releases and backporting fixes when needed. That is an expensive effort, of course; it's not clear how it might be funded now that Mozilla has shown even less interest in funding that work itself.

If there is a big need for this by corporations, someone should found a company that sells this kind of support. The only thing Mozilla would need to do is allow use of its trademarks, and I don't see why it wouldn't (and if it doesn't, all that needs to change is the name).

This is exactly one of the things that make open source *more* business-friendly. Corporations are not tied to the vendor for services and support, anyone can take the code and sell support for it. Although granted this may be more complicated for businesses (get the tool from one organization, buy support from another).

Editor's comment

Posted Jun 26, 2011 22:56 UTC (Sun) by butlerm (subscriber, #13312) [Link] (3 responses)

Although granted this may be more complicated for businesses (get the tool from one organization, buy support from another).

The only sane way for a sizeable business to provide long term software support is to take the role of the software vendor. Otherwise you are essentially just providing a consulting function, with little or no power to do anything about anything.

Enterprise Linux vendors may well cover this

Posted Jun 27, 2011 10:36 UTC (Mon) by Cato (guest, #7643) [Link] (2 responses)

Red Hat shows you can simply sell support without owning proprietary software. The problem for Firefox in the enterprise is that browsers are viewed as a free-as-in-beer commodity, despite the investment that goes into them. So it's not clear anyone would actually buy this support.

In any case, I would hope the enterprise Linux vendors will take this on within RHEL etc - since they sell some workstation Linux, it's reasonable for them to continue to backport fixes to older Firefox versions, which is certainly what Ubuntu does. The only difference is that there are more "major" versions, and over time (say Firefox 5 to 12) it may get harder to backport the fixes because the core browser is changing more frequently.

Anyone else who needs an enterprise supported Firefox with a stable version could start with an Enterprise Linux version.

Enterprise Linux vendors may well cover this

Posted Jun 27, 2011 10:48 UTC (Mon) by zwenna (guest, #64777) [Link] (1 responses)

> it's reasonable for them to continue to backport fixes to older Firefox versions, which is certainly what Ubuntu does.

Reasonable or not, Ubuntu does not backport fixes to older Firefox versions. Rather, they upgrade to a supported version:

https://wiki.ubuntu.com/DesktopTeam/Specs/Lucid/FirefoxNe...

Enterprise Linux vendors may well cover this

Posted Jun 27, 2011 13:16 UTC (Mon) by Cato (guest, #7643) [Link]

They used to backport fixes, but not any longer. Oh well...

Editor's comment

Posted Jun 26, 2011 4:24 UTC (Sun) by cyd (guest, #4153) [Link]

Mozilla is sitting on mountains of Google cash. Refusing to fund something like this makes no sense.

Editor's comment

Posted Jun 26, 2011 15:20 UTC (Sun) by jzb (editor, #7867) [Link] (7 responses)

The problem here is, why would any business buy Firefox support when they get IE support bundled in? (Put on your corporate IT hat, not your FOSS supporter hat here.)

The majority of businesses are still using Windows on the desktop - and they already get IE. The rapid-fire releases without any long-term support just seem to me to be an way to push companies right back to IE. Which means that we go back to companies telling their developers "just code for IE," which means we start losing all the gains we've made the past few years away from IE-only sites.

I may be wrong, but this does not strike me as particularly adept long-term planning from Mozilla.

Code for whom?

Posted Jun 26, 2011 20:24 UTC (Sun) by pboddie (guest, #50784) [Link] (6 responses)

Which means that we go back to companies telling their developers "just code for IE," which means we start losing all the gains we've made the past few years away from IE-only sites.

Yes, but what kind of companies and what kind of applications are we talking about? If companies want their internal applications to only work on IE, ActiveX and all, who really cares? They can all go back to writing Visual Basic applications - that's not the Web, either.

If all the people outside "the enterprise" are using Firefox, those companies that deploy applications for everyone else to use will have to support Firefox, and the best way they can do so is by writing stuff which adheres to Web standards. Targeting the corporate people who will always find some excuse not to use one's stuff anyway is a distraction - the Mozilla people have called this one correctly.

Code for whom?

Posted Jun 26, 2011 21:11 UTC (Sun) by vblum (guest, #1151) [Link]

The mere fact that the publicity is a disaster shows that the call was not made correctly.

It is always a nice idea to give the impression that you care about your users. That would be the people that use your product in order to get a different, time consuming job done, different from (re)coding for your project or filing bugreports as a feature. If the product ends up costing more time and tells them outright "fix it yourself, not our problem", product dies an ugly death. Why would I use a browser that is allowed to break every few weeks, by stated admission of the team behind it?(*)

The Linux kernel process does not do this. KDE does not do this. Gnome does not do this. nethack does not do this. emacs doctor mode does not do this. Firefox did not use to do this.

If nothing else, the idea that stability is not a useful thing to guarantee has already done some damage by merely being worded as such.

(*) so far I do, anyway, I don't really believe that firefox is doomed just yet.

Code for whom?

Posted Jun 26, 2011 22:04 UTC (Sun) by jd (guest, #26381) [Link] (1 responses)

Let's start with Selenium, a common integrated test suite. That relies heavily on stable ABIs. It's used heavily by corporations to test their products before release across multiple browsers. If Firefox stops being on the list, no matter what the reason, Firefox will stop being considered not only on corporate websites but also on all corporate products used to make websites (be that commercial servers or commercial scripting languages).

Since nobody likes learning multiple browsers, if a person encounters an important Firefox-disabled site they will stop using Firefox at all.

With Microsoft abandoning Silverlight in favour of HTML5, Firefox COULD have become more important. Now it'll become a footnote in history.

Code for whom?

Posted Jun 26, 2011 23:38 UTC (Sun) by pboddie (guest, #50784) [Link]

Last I heard, Selenium consulting and support was a healthy area of business with money changing hands and everything. I think people can get a few dollars together and keep up with whatever ABI churn the Mozilla people can produce.

And I'm sure all this Web site testing isn't entirely without cost, and since corporate people do not merely pay for trinkets like Eclipse plugins offering marginal functionality benefits, but frequently pay good money for such trinkets, I'm sure the budgets will withstand the necessary demands to keep Web testing and automation tools functional.

Code for whom?

Posted Jun 28, 2011 21:27 UTC (Tue) by rodgerd (guest, #58896) [Link] (2 responses)

I work for a bank. Our stats show most of our customers do their banking in the morning, when they get to work. As Mozilla have issued their "fuck off, we don't want you" statement to corporate users, I expect to see the numbers of Firefox users plummet. As it does, the interest in properly testing our applications to work with Firefox will disappear.

Code for whom?

Posted Jun 28, 2011 21:39 UTC (Tue) by dlang (guest, #313) [Link] (1 responses)

I'd like to know what companies that have these silly version dependant policies actually blessed Firefox in the first place?

I suspect that in most cases it's being demanded by users, if not just installed by the users in spite of the corporate policies (that's how firefox got into the businesses in the first place)

as for the need to be able to purchase support, where did they buy support for firefox version 3 or 4?

Ars Tech has it right that going with a date based versioning would have avoided this major number panic, and that there is really very little difference between the firefox 5 release and any of the 18 different firefox 3.6.x releases in terms of risk. they all have security fixes in them, and they all have non-security changes in them

Code for whom?

Posted Jul 7, 2011 20:37 UTC (Thu) by wtanksleyjr (subscriber, #74601) [Link]

Actually, the "silly version dependent policies" involved here are Firefox's own extension checking policies, together with the breakneck and undocumented change in release policies.

I can't tell you which enormous org I work for, or why they blessed Firefox. I can only hope that your question was intended to ask whether there were any such orgs, and my answer can somehow reassure you that there are such orgs, and they blessed Firefox, and then the Microsoft reps who consult for us pointed out Asa's comments, and now the future of Firefox in our org is now in doubt. Since our org has a large (but violently disputed) voice in the control of as much as 3/4 of the surface area of the globe, that's a pretty big loss for Firefox adoption.

I don't blame Asa for his decision; I think it's worked well for others and hope it works wonderfully for him and his team. I do blame him for failing to communicate his decision. Such communication, if delivered to the right people, might have arranged to have a backporting team already arranged, thus removing the need for the announcement of ZERO support for the immediately prior release.

The ideal outcome of this would be for some distributor to step up and form a team to build a stable "Firefox EE" build, with a roadmap based on the expected future of Firefox. This roadmap would fairly quickly become populated with features that are actually present in Firefox 6 or 7, but are considered too big or extension-breaking to use in the EE.

My main expectation for this kind of thing would be Debian and Redhat.

-Wm

Editor's comment

Posted Jun 26, 2011 18:26 UTC (Sun) by Kluge (subscriber, #2881) [Link]

I think this is understating the problem. For one thing, the original Firefox model was to focus on making a core browser that encouraged the adding of extensions to enhance function. The new Firefox model makes is significantly more difficult to use extensions to enhance function.

Also, the kernel is very serious about maintaining stability of the extensive unix ABI for userspace. I would argue that extensions and plugins *are* the userspace apps for Firefox. And now there's no guarantee of any lasting interface stability for those apps.

The kernel also has to keep up with rapidly evolving hardware. I don't think web standards or usage change at nearly the same pace as hardware does. So why the need for a development model focused on flexibility for core developers above everything else?

Mozilla to Businesses: We're Not Interested (PC Mag)

Posted Jun 25, 2011 20:26 UTC (Sat) by fest3er (guest, #60379) [Link] (3 responses)

A few points. Mozilla may be a non-profit corporation, but it is still a corporation, albeit one that need not make a profit. Robert Bosch Foundation is a non-profit corporation; they make scads of tools and appliances, but the business doesn't need to make a profit.

Businesses that rely on a particular browser should arrange with that browser maker to pay for long-term support. Or they can band together and fund their own captive business that supports the chosen browser for a number of years.

No business has the resources to upgrade major software systems every year, let alone every three months. They need stability far more than they need new features.

HTML is, in a manner of speaking, much like the X.25 protocol: everyone implements it differently, and no one implements it right (Attribution: F. Bronzo). Hence there are always interworking problems. No two browsers render any particular page the same way. Many of them seem to have problems displaying 'simple' pages correctly; pop-up menus are hidden behind other elements; a final 'small-font' element causes a vertical scrollbar to be rendered in one particular browser. It gets to be a real PITA having to user three different browsers. And remembering which browser must be used with with web site.

I would like to see browser makers band together and support a common, core set of functionality that implements the HTML standard(s). This would almost guarantee that the basic functionality of HTML would work regardless if which browser is used.

U.S. law may have something to say about how long a product must be supported after it has been released to the market. For instance, Ford Motor Company must support vehicles it sells for some years after the vehicle is introduced. So if they introduce an 2012 F250 diesel pickup truck in 2011, then release a a 2012-1/2 model 1/1/2012, they must still support the original 2012 model. From a bird's-eye view, Firefox 4 and 5 aren't that different from the automotive analogy.

Dropping support for a browser 'model' the instant it is superceded is a good way to lose market share. It will be perceived as one of two things; either the old model was fatally flawed and shouldn't have been released in the first place, or the manufacturer and employees look upon their customers with disdain.

Mozilla to Businesses: We're Not Interested (PC Mag)

Posted Jun 25, 2011 21:01 UTC (Sat) by Julie (guest, #66693) [Link] (2 responses)

A few points. Mozilla may be a non-profit corporation, but it is still a corporation, albeit one that need not make a profit.

According to the article, Mozilla were doing OK with making a profit, at least in 2009:

--'Telling businesses that they are not "a focus" nor should be for Firefox developers is also potential profit suicide. Yes, Firefox makes a profit and a hefty one, with lucrative deals made with search engine companies. Mozilla raked in $104 million in profit in 2009, largely though a deal with Google.'--

Businesses that rely on a particular browser should arrange with that browser maker to pay for long-term support.

The problem reflected in the article, the blog and the blog comments really seems to be that business use and 'normal' use is rather different, so this might really work...FireFox Enterprise Edition, anyone? :-)

Mozilla to Businesses: We're Not Interested (PC Mag)

Posted Jun 26, 2011 9:22 UTC (Sun) by elanthis (guest, #6227) [Link] (1 responses)

> According to the article, Mozilla were doing OK with making a profit, at least in 2009:

That's not what a "for-profit" company means. Any and all companies can and are really expected to pull in a positive income. Otherwise they go bankrupt. A not-for-profit still can still pull in a hefty income; it's just expected that the cash go back into the organization's chartered goals rather than being equitable gain for the corporate shareholders.

Mozilla to Businesses: We're Not Interested (PC Mag)

Posted Jun 26, 2011 15:01 UTC (Sun) by tialaramex (subscriber, #21167) [Link]

Also, Mozilla have both a not-for-profit foundation, and a for-profit corporation whose profits fund the foundation, or something like that.

There are things a not-for-profit isn't allowed to do, but a for-profit company can do those things and donate the proceeds to a not-for-profit without the not-for-profit losing its tax status.

Update: Kev Needham, Channel Manager at Mozilla has responsed to a request for comment:

Posted Jun 25, 2011 20:47 UTC (Sat) by SilverWave (guest, #55000) [Link] (9 responses)

"The Web and Web browsers continue to evolve rapidly. Mozilla's focus is on providing users with the best Web experience possible, and Firefox needs to evolve at the pace the Web's users and developers expect. By releasing small, focused updates more often, we are able to deliver improved security and stability even as we introduce new features, which is better for our users, and for the Web."

"We recognize that this shift may not be compatible with a large organization's IT Policy and understand that it is challenging to organizations that have effort-intensive certification polices. However, our development process is geared toward delivering products that support the Web as it is today, while innovating and building future Web capabilities. Tying Firefox product development to an organizational process we do not control would make it difficult for us to continue to innovate for our users and the betterment of the Web."

Well that sounds reasonable...

If business finds Firefox useful... well they could try paying to have an older fork of the code maintained.

They pay for other services that they find useful... how is this situation different?

Update: Kev Needham, Channel Manager at Mozilla has responsed to a request for comment:

Posted Jun 26, 2011 0:52 UTC (Sun) by Lennie (subscriber, #49641) [Link] (7 responses)

Please, this would be a terrible time to create a fork or keep using older versions. There is so many new features that got added to HTML5 and friends which are all being implemented by the different browsers.

There are probably something like 30+ catagories of new features in all these standards. Some browsers have implemented some things not at all, others have 90% coverage of the same feature. While they may only support 50% of an other specification. They are all really busy adding all of these things. Even Microsoft.

(OK, it is pretty obvious IE will not support WebGL any time soon)

This is a time with lots and lots of new development and bugs being fixed and new specifications being implemented.

It is not the same things as a server environment, where you have RedHat supporting RHEL for 10 years.

As an example IE6 is gonna turn 10 in August. The same with the Windows XP desktop, which is still used by 50% of the website visitors all around the world.

Which is stupid and crazy. Atleast the developers of Mozilla, Chromium and Opera allow people to upgrade their older browsers on Windows XP so they don't have to be stuck in the darkages.

Unlink Microsoft which doesn't allow people to upgrade from IE8 to IE9+ on Windows XP.

Update: Kev Needham, Channel Manager at Mozilla has responsed to a request for comment:

Posted Jun 26, 2011 5:19 UTC (Sun) by butlerm (subscriber, #13312) [Link] (6 responses)

Large businesses that needed a stable edition of a browser like Firefox would probably qualify and install a new version every three to five years, depending on their needs. The Internet doesn't change so fast that that is unreasonable.

Update: Kev Needham, Channel Manager at Mozilla has responsed to a request for comment:

Posted Jun 26, 2011 16:28 UTC (Sun) by Lennie (subscriber, #49641) [Link] (5 responses)

"The Internet doesn't change so fast that that is unreasonable."

That was the whole point of my post, it is changing that fast right now.

I know that after IE had the really large marketshare and the W3C just wanted to make an incompatible XML-based XHTML standard nothing seemed to happen.

But things are now completely different. IE6 will be completely gone next year.

Apple took the engine from the KHTML-project and created 'their own': Webkit. By adding a lot of new code.

Mozilla, Opera and Apple created the WHATWG which started the HTML5 effort, to prevent new the XML-only XHTML-standard to be backwardscompatible and add more features.

Safari started the (JavaScript) speed battle and Mozilla joined them.
Then Google started the Chromium-project and created the V8-engine
after which they took the lead for a while after Opera took it and then Chromium again.

Just before Microsoft's marketshare dropped below 50% Microsoft got the old IE5/IE6 developer team together again and created IE9. And most browservendors have introduced hardware acceleration. Now Microsoft are working on IE10 which might have the fastest JavaScript-engine yet.

IE10 will be used heavily as presentation-engine of the new version of Windows, Windows 8.

Most of the user-visible changes happend in just a little less than 2 years time.

Update: Kev Needham, Channel Manager at Mozilla has responsed to a request for comment:

Posted Jun 26, 2011 23:08 UTC (Sun) by butlerm (subscriber, #13312) [Link] (4 responses)

That was the whole point of my post, it is changing that fast right now.

Just because the browsers change that fast doesn't mean the Internet is changing that fast. Most large websites are just getting around to the point where they can consider dropping support for IE7. Why? Because IE7 was released almost five years ago. No one with a large, public facing website is going to consider dropping support for IE8 for at least three more years, for the same reason.

Update: Kev Needham, Channel Manager at Mozilla has responsed to a request for comment:

Posted Jun 27, 2011 10:06 UTC (Mon) by Lennie (subscriber, #49641) [Link] (3 responses)

Actually, the normal Microsoft browser version replacement cycle takes about 2 years. For everyone that does normal updates, so like IE7 to IE8. That is pretty much everyone, including businesses. So if you think 5 years is normal for businesses you are wrong.

The reason IE7 is still supported is because IE6 still had such a broad userbase. If IE6 had a normal upgrade cycle no one would still be using IE7. But IE6 didn't have a normal upgrade cycle, see below.

But IE is the showest being updated, most other browsers take a few days or a few months.

And IE10 will be out in 2012, IE9 came out in 2011.

But IE8 is the exception.

The real reason IE8 will be slow to replace is because of Windows XP, on Windows XP IE8 is the latest version and there will be no IE9 for Windows XP.

So it depends on the time it takes to replace Windows XP.

That is why IE8 is new IE6.

Because of the marketshare of IE6 at the time, people creating intranet-websites for businesses created IE6-specific websites. So when an upgrade came available they couldn't upgrade. This is why IE6 has a bigger marketshare on a lot of sites than IE7.

Similair with IE8, it is has a barrier to upgrade, a dependency.

Windows XP has dropped below 50% worldwide, below 40% in Europe and around 30% in North America. It seems it will take atleast 4 years to completely replace Windows XP.

It doesn't mean IE will be the preferred browser for people on Windows XP, especially at the home. And it also doesn't mean people won't be dropping support for IE8 before it is completely replaced.

These are reasons IE doesn't get updated, but they don't apply to Firefox, Chrom(e)(ium), Opera or Safari.

But even Microsoft is pushing hard to get people to upgrade, they are replacing Silverlight on their own websites with HTML5 where they can.

It also doesn't mean that certain websites need the new features, it just needs certain websites to adopt it and see the benefits in other browsers to move a large userbase from IE8 to other browsers on Windows XP.
___

Anyway, let's get back to the issue at hand.

Businesses that want to deploy Firefox on their corporate desktops what parts do they depend on the most which might break ?

Is it extensions or just rendering webpages/webapps/whatever ?

Update: Kev Needham, Channel Manager at Mozilla has responsed to a request for comment:

Posted Jun 28, 2011 8:45 UTC (Tue) by Wol (subscriber, #4433) [Link] (2 responses)

But XP will probably need to be supported for a while yet ...

Not this machine, but my old machine is still in regular use. With three slots taking a maximum of 256Mb each, my machine is upgraded to the max yet it's below the recommended minimum for any upgrade.

What do I do if IE8 support is dropped? I had difficulty finding the money for my (wife's) new machine! Old machines live on a long time in some hands - if they don't break poor people can't afford to replace them ...

Cheers,
Wol

Update: Kev Needham, Channel Manager at Mozilla has responsed to a request for comment:

Posted Jun 28, 2011 10:24 UTC (Tue) by njwhite (guest, #51848) [Link] (1 responses)

Perhaps you should consider a Linux distribution.

Update: Kev Needham, Channel Manager at Mozilla has responsed to a request for comment:

Posted Jun 28, 2011 15:08 UTC (Tue) by Lennie (subscriber, #49641) [Link]

Totally agree, my parents use Linux on the desktop and I have a lot less work with the OS/apps; All the software is easy to keep up to date.

I do however have the same amount of work with just doing user-support. "How do I do X", they wouldn't know how to do certain things on any computer.

Linux is not more complicated for most users, maybe some are stuck in their ways on their existing Windows XP for example.

But they eventually need to move to something else anyway, Microsoft stopped supporting it, only extended support till 2014.

Update: Kev Needham, Channel Manager at Mozilla has responsed to a request for comment:

Posted Jun 27, 2011 8:22 UTC (Mon) by Seegras (guest, #20463) [Link]

"...may not be compatible with a large organization's IT Policy and understand that it is challenging to organizations that have effort-intensive certification polices."

THAT is the problem. It's not really a technical problem, but policies made for entirely different software, probably for operating systems which do not have a package management. And also, maybe in respect of web-applications which do not adhere to standards and need to be fixed everytime the browser changes.

The comment about add-ons looks a little out of touch....

Posted Jun 25, 2011 20:58 UTC (Sat) by SilverWave (guest, #55000) [Link] (2 responses)

>...extension breaking whenever a new update comes out...

The add-on issue is being addressed in the short-term by auto bumping all working add-ons.

Long term Jetpack and other initiatives will sort this out.

see here:
http://www.oxymoronical.com/blog/2011/06/Why-do-Firefox-u...

"How does the Add-ons SDK fit in?

With the most recent update to Firefox we also made the first official release of the new add-ons SDK available. These are tools designed to make developing add-ons for Firefox easier and faster and they largely serve as an add-ons API sitting between the add-on and Firefox with one difference, the API is actually a part of the add-on so you can actually write your own additional APIs to supplement those in the core SDK.
Add-ons written with the SDK (and only using the core APIs) gain a lot of the benefits of add-ons for other browsers. The APIs in the SDK can be far more stable than those in the browser itself, as the browser code changes the internal SDK code can adapt to match allowing add-ons to work just by rebuilding with the newer SDK. This makes the problem of Firefox updates much smaller for those that can move their add-ons to the SDK.
I don’t foresee a future where the only add-ons are SDK based though. We’d lose too many really important add-ons that way so the SDK doesn’t completely solve the problem. "

The comment about add-ons looks a little out of touch....

Posted Jun 25, 2011 21:37 UTC (Sat) by Julie (guest, #66693) [Link] (1 responses)

Mmm, here's some more comments from the blog:

---'Don June 23, 2011 at 10:08 pm #
I upgraded to Firefox 5 and it shut down my antivirus saying it was not compatible. I was unable to turn it back on. It also disabled my my System Restore and The Weather Channel Add On did not work at all. Fortunately I had a mirror image and back up of my computer and was eventually able to restore it. That brought me back to Firefox 4 which works just fine. If this is hwat we have in store, time to go back to IE.

Reply
Asa June 23, 2011 at 10:33 pm #
You should ask your Firefox add-on vendors to use Firefox’s stable APIs available through the Add-on SDK. If they do that, their add-ons won’t break with Firefox releases.

Reply
matt June 24, 2011 at 11:05 am #
if its that simple, can you please ask major vendor Log Me In to fix their plugin that no longer works when behind a proxy on FF4? they dont seem to think it’s such a simple issue.'---

The comment about add-ons looks a little out of touch....

Posted Jun 26, 2011 0:34 UTC (Sun) by Lennie (subscriber, #49641) [Link]

My guess is the JetPack API has coverage for 90% of API the older extensions used.

And Log Me In and so on will need time to change their plugin or actually are in that 10% catagory that might need to be adapted when a new version comes out or atleast properly tested.

Mozilla to Businesses: We're Not Interested (PC Mag)

Posted Jun 26, 2011 0:58 UTC (Sun) by slashdot (guest, #22014) [Link]

Well, they can just continue to use the older version they apparently like so much.

If they are worried about security, they can restrict the older version to internal sites, and also install the newest version side-by-side for Internet browsing.

Or they could just trust Mozilla and upgrade automatically to the latest version, rolling back should any problem pop up (or fixing the problem itself).

No idea where the issue is.

Mozilla to Businesses: We're Not Interested (PC Mag)

Posted Jun 26, 2011 6:29 UTC (Sun) by Hausvib6 (guest, #70606) [Link] (1 responses)

So businesses want a gratis web browser that is supported for a long long long time so that they can minimize the cost of migration and testing even further at the expense of Mozilla's developer time and countless other souls, especially people like me who likes to see a lot of new exciting features get added to Firefox. I'm surprised! Really!

Mozilla should consider spending some resources to help these poor businesses so that cutting-edge developments in Firefox can be slowed down, these busineeses don't like changes and unpredictabilities especially if they are occuring too fast.

Like Internet Explorer?

Posted Jun 26, 2011 18:04 UTC (Sun) by gmatht (guest, #58961) [Link]

Businesses still have an effectively gratis web-browser with long term support: IE8/9. I don't think business in general cares what Mozilla does, and this affects more people who want to advocate the use of FOSS, particularly those who have already rolled out Firefox on the the assumption that it would be supported for a year or more, and are now left with egg on their face. Mozilla can't please everyone, but lets not shoot the messenger.

Also, it is not just businesses that are wary of Firefox updates. Home users have learnt that upgrading Firefox breaks most of their favorite extensions. Jetpack etc. may help with this, but users are once bitten twice shy.

Mozilla to Businesses: We're Not Interested (PC Mag)

Posted Jun 26, 2011 8:25 UTC (Sun) by MaartenM79 (guest, #73036) [Link]

Sounds to me like the same people unwilling to transition from ie6 because of some intranet app you can do without. The kind of people who do endless compliance reports. Get you hands dirty, talk to some people with the right knowhow and stop whining.

Intranet apps drive browser support

Posted Jun 26, 2011 8:33 UTC (Sun) by Cato (guest, #7643) [Link] (1 responses)

Unless Firefox has good enterprise support of a stable version for a year or two, as discussed, there will be more intranet apps that only support Internet Explorer. In turn, there will be some employees who never get introduced to Firefox at work, and never look to install it at home.

All very short-sighted on Mozilla's part - it wouldn't cost much to do a stable version, or just license the trademark at very low cost to a company willing to do this.

Outstanding Enterprise Integration

Posted Jun 26, 2011 17:45 UTC (Sun) by SilverWave (guest, #55000) [Link]

Mozilla to Businesses: We're Not Interested (PC Mag)

Posted Jun 26, 2011 9:23 UTC (Sun) by lmb (subscriber, #39048) [Link] (8 responses)

It is the same political bullshit that is the basis for most of the demand for "Enterprise Linux" distributions; a fear that changing version numbers and frequent releases imply breakage.

Granted, they have reasons to believe that from past experience, but more and more software products are shifting - the Linux Kernel has already been cited, Pacemaker is another. Continuous Integration and Continuous Delivery are taking over eternal release cycles. Sure, there will be some new breakage initially, but the new model will allow that to be fixed quickly as well.

The breaking add-ons for FF are a good example; there is a new standard/stable API, or the ability to auto-bump extensions if they work. Extensions checking a specific FF version string need to be fixed, simple as that.

The "Enterprises" are screaming bloody murder and spreading FUD because they're associating the FF5 and FF6 releases with such releases as (FF2->)FF3->FF4 were; which they are simply not. They'll eventually get over it, or find a distributor that lies to them about the version number in exchange for their money. (Just like with Enterprise distros and the Linux kernel versions they include.)

Admittedly, communicating the changed release process better is a key task for the FF team these days ("No, the world will not end, and your fears won't come true - here's why, and the steps we're taking to ensure that"), but tech journalists really should know better and help, instead of making the problem worse. Too bad.

Mozilla to Businesses: We're Not Interested (PC Mag)

Posted Jun 28, 2011 23:35 UTC (Tue) by ThinkRob (guest, #64513) [Link] (7 responses)

> Extensions checking a specific FF version string need to be fixed, simple as that.

Uh... that would be all of them. All extensions contain a max version field.

And no, much to Mozilla's amazement, not every extension is deployed via addons.mozilla.org

This is kinda missing the point though; the talk of what is/is not a sane corporate testing policy, etc. is all ignoring the bigger question:

Why are they doing this?

Why on earth is Mozilla bumping the major version with (nearly) every couple releases? I can understand doing it "just cause" if it were, say, a pet project with no significant users. Or if it were an internal app, hey, use whatever versioning policy you want. But Firefox is neither of those things. It's a well-funded browser used by millions including some *very* high profile businesses. At this point Mozilla has to know -- even if they didn't when they first considered doing it -- that bumping the major version is going to result in a fair bit of pain and irritation for a not-insignificant chunk of their userbase.

As long as I've been developing and using software, I've been under the impression that if you know that a given change is not "free" in terms of user frustration, you'd better have a decent reason for making it. Apparently, Mozilla thinks otherwise.

The lack of a good explanation for why they're doing this, combined with the outright hostile attitude of at least one of the developers comes together to make Mozilla look (at least to some folks such as myself) not unlike a group of hipster 20-somethings sneering "get with the times Grandpa" to their users.

And that's not a particularly pleasant way to think of what was once a poster child of open source.

Mozilla to Businesses: We're Not Interested (PC Mag)

Posted Jun 29, 2011 7:37 UTC (Wed) by dlang (guest, #313) [Link] (6 responses)

clarification, they aren't bumping the major number every couple of releases, they are bumping the version number every release (there is no minor version number any longer)

Mozilla to Businesses: We're Not Interested (PC Mag)

Posted Jun 29, 2011 11:20 UTC (Wed) by tialaramex (subscriber, #21167) [Link] (5 responses)

So like xterm then. I appear to be running xterm version 261. I'm pretty sure I recall using one of the 100-ish releases in the past, and perhaps I'm old enough to have run double digit versions of xterm. Nothing terrible happened in that time, mostly bugs got fixed.

However for Firefox it does mean they can no longer distinguish "this release is a minor upgrade, it fixes sixteen bugs AND the JS implementation is up to 10% faster" from "many APIs are changed, we have replaced the core rendering engine, and removed support for SSLv3 due to a patent problem".

Now, that's never bothered me with xterm. Maybe it won't bother me with Firefox. Maybe web browsers will settle down and stop innovating and all future Firefox versions are largely bugfixes. But I doubt it.

Mozilla to Businesses: We're Not Interested (PC Mag)

Posted Jun 30, 2011 12:51 UTC (Thu) by nix (subscriber, #2304) [Link] (4 responses)

less does the same thing. But xterm and less don't change much, so in effect all releases are minor releases. I'm not sure it works so well for projects undergoing constant major disruptive change (actually I'm not sure such a project can be considered usable by any end users at all).

Mozilla to Businesses: We're Not Interested (PC Mag)

Posted Jun 30, 2011 18:03 UTC (Thu) by dlang (guest, #313) [Link] (3 responses)

that would mean that the linux kernel is not usable by any end users, and as one of the end users, I can attest that it is very usable.

Mozilla to Businesses: We're Not Interested (PC Mag)

Posted Jun 30, 2011 21:09 UTC (Thu) by nix (subscriber, #2304) [Link] (2 responses)

The Linux kernel is not undergoing constant major *user-visible* disruptive change. Internal stuff the users cannot detect is not relevant here.

Mozilla to Businesses: We're Not Interested (PC Mag)

Posted Jun 30, 2011 22:45 UTC (Thu) by dlang (guest, #313) [Link] (1 responses)

but at one point it was, it's only over time, with the experience of making many releases that they have gotten as good as they are about not making disruptive changes.

Give the Mozilla folks some time and I expect them to start getting pretty good at this as well.

for all the fuss over this change, how many people even noticed when ubuntu moved from 4.x to 5? the only reason I noticed is that I went looking to see what version I had to figure out how hard it would be to manually move from there to 5. I was surprised to find I was already running 5.

I don't run a lot of extensions, so I wouldn't have expected trouble, but I think that most people actually have the same experience.

Mozilla to Businesses: We're Not Interested (PC Mag)

Posted Jul 2, 2011 11:16 UTC (Sat) by JanC_ (guest, #34940) [Link]

I did immediately notice because of breaking extensions, and judging from the support channels I'm not the only end-user experiencing that... And to be honest: all but one of them started working again within a week after the upgrade (unfortunately it's one I use fairly often but is not on the Mozilla addon-site).

Mozilla to Businesses: We're Not Interested (PC Mag)

Posted Jun 26, 2011 12:09 UTC (Sun) by lyda (subscriber, #7429) [Link]

Oh boo hoo.

Bad documentation? It's a free software project - they accept contributions I assume. Businesses could contribute documentation they deem to be missing.

Lots of version changes? Yes, the web is evolving you muppets. Hire good web developers who don't write shit code for your internal tools. And for those third-party vendors who supply broken web interfaces that only work for certain browsers - how about firing them and getting a different third-party solution.

Sorry, there's nothing in that article that impresses me. I switched to Chrome a long time ago, but the Firefox developers are right to do frequent releases and keep their browser up to date. The functioning Firefox development process should not be broken to help paper over the brokenness in the way some people run their IT infrastructure.

Mozilla to Businesses: We're Not Interested (PC Mag)

Posted Jun 26, 2011 12:48 UTC (Sun) by KaiRo (subscriber, #1987) [Link] (6 responses)

Asa's comment has been taken significantly out of context, in my opinion. He just said that the enterprise is "not a focus" for Mozilla. He didn't say we're not interested in having corporate installs, he just said we won't change our processes to suit them. If there's low-hanging fruit to make them deal better with our new fast release process, we might take a look, but we won't set aside resources for the relatively small amount of users that corporate installations have in total, compared to our general user base.

And as someone mentioned that the current "models" of Mozilla wouldn't be supported, that's wrong. We support the current release for security by automatically updating it to the next release. That may have a small amount of feature changes as well, of course, but it's the way of supporting the new one. That's just like your car would get a makeover and be updated to the most current technology every time a service is made. I guess you would like that. :)

Mozilla to Businesses: We're Not Interested (PC Mag)

Posted Jun 26, 2011 15:04 UTC (Sun) by tialaramex (subscriber, #21167) [Link] (5 responses)

"Why doesn't my car fit in my garage any more? It was fine yesterday"

"It's an upgrade - we've widened the car free with your annual service"

"But, now it doesn't fit"

"Don't bother us with your whining"

Mozilla to Businesses: We're Not Interested (PC Mag)

Posted Jun 26, 2011 17:49 UTC (Sun) by SilverWave (guest, #55000) [Link]

yeah I would ask for my money back....

Mozilla to Businesses: We're Not Interested (PC Mag)

Posted Jun 26, 2011 23:24 UTC (Sun) by KaiRo (subscriber, #1987) [Link] (1 responses)

Sure, it wouldn't be a comparison, if it didn't fail in some way. Of course you can construct a case out of it that looks laughable.

Mozilla to Businesses: We're Not Interested (PC Mag)

Posted Jun 27, 2011 5:52 UTC (Mon) by jku (subscriber, #42379) [Link]

You set that trap all by yourself :)

I understand your position fine and think it maybe isn't a bad one. Saying that automatically updating to the next version is supporting the older versions... well, that's just plain doublespeak. Don't be surprised if you get called out for re-defining words to your needs.

Mozilla to Businesses: We're Not Interested (PC Mag)

Posted Jun 27, 2011 8:10 UTC (Mon) by Seegras (guest, #20463) [Link] (1 responses)

It's more like "you changed the radiator mascot! Now we must redeploy and retest everything!"

Mozilla to Businesses: We're Not Interested (PC Mag)

Posted Jun 29, 2011 15:22 UTC (Wed) by markhb (guest, #1003) [Link]

No, it's more like "We changed the radiator mascot. Maybe we switched the brake and accelerator pedals as well, but you're going to have to take her out on the highway to find out."

My posts are my opinion only and do not represent my employer in any way.

Mozilla to Businesses: We're Not Interested (PC Mag)

Posted Jun 26, 2011 19:29 UTC (Sun) by PO8 (guest, #41661) [Link] (18 responses)

I think the article is wrong. The new Firefox plan does not, in my experience, "work for the average user." I'm not an average user---I'm as open source as you can get, and am running Debian and Iceweasel on several boxes. Thank goodness Iceweasel is still "stuck" on a Firefox-3 based version even in unstable, due to the issues described on this forum. However, the Web has moved on, and I've been forced to upgrade my browser to Iceweasel 4. A bunch of new things worked, which is great, but a bunch of extensions I counted on didn't anymore, and my browser configuration broke. In short, I spent several hours this week unproductively messing with my browser, and I count my time as precious. If I am in this situation, I can't imagine "Aunt Tillie" is in a better one.

I'm about one more such adventure away from switching to Chrome---most of my friends (with a similar profile to mine) already have. As pointed out in the article, it looks strongly like what is happening is that users are leaving Firefox for Chrome in numbers that far exceed the numbers of new Firefox signups. Under those circumstances, I can't imagine why Google would renew its agreement with the for-profit Mozilla Corporation under nearly the favorable terms they have now. It would seem much more likely that they would give their own open source browser preferential treatment. (The answer, though, may be antitrust: with the US FTC actively investigating Google right now, they don't want to appear to be a monolith.)

I'm guessing the same moves that may drive businesses away from Firefox are already starting to drive everybody else away also. I'm hoping I'm wrong here---I certainly wish Firefox and the non-profit Mozilla Foundation every success. But they need to think carefully about their development plans as well as their messaging. In particular, they need to get Jetpack adopted widely by extension developers *before* trying to change big hunks of the current extension "interface" every few weeks.

Mozilla to Businesses: We're Not Interested (PC Mag)

Posted Jun 26, 2011 20:53 UTC (Sun) by elanthis (guest, #6227) [Link] (12 responses)

> If I am in this situation, I can't imagine "Aunt Tillie" is in a better one.

Because surely Aunt Tillie even knows what the heck an extension is, had any extensions installed besides the adware crap far too many apps come with, and thus also had a ton of extensions she cared deeply about break on her. Right?

You really don't see a lot of non-nerds using extensions. Regular users just want to open the browser, click on one of their bazillion unorganized bookmarks, and then read whatever is on that page. That's about it.

Those users upgraded to Firefox 4 just fine (Firefox 3 is near dead already in usage statistics, and Firefox as a whole has not declined in use despite an outcry from the uber-nerds about extension breakages) and are going to upgrade to Firefox 5 just fine as well.

The only threat Firefox has is from Chrome. Which, I might note, upgrades even more often than the new Firefox model and also breaks extensions fairly regularly. And users still don't care because almost nobody actually uses extensions outside of the uber-dorks who think that messing around with browser behavior is actually a fun or useful way to spend time.

Mozilla to Businesses: We're Not Interested (PC Mag)

Posted Jun 26, 2011 21:36 UTC (Sun) by fuhchee (guest, #40059) [Link] (1 responses)

Re. adware crap, I wonder whether the unavoidable firefox N++ api churn will result in greater difficulty keeping sanity-savers like AdBlock+ working. Wouldn't it be a shame if firefox N++ users started to see more ads, like those of a certain mozilla.com underwriter?

Mozilla to Businesses: We're Not Interested (PC Mag)

Posted Jun 26, 2011 21:59 UTC (Sun) by njs (subscriber, #40338) [Link]

...That's seriously your conspiracy theory?

Mozilla to Businesses: We're Not Interested (PC Mag)

Posted Jun 26, 2011 22:31 UTC (Sun) by Cato (guest, #7643) [Link] (8 responses)

Firefox extensions get used a lot, particularly Adblock Plus etc - not by Aunt Tilly but by ordinary people who are are clueful enough to know what an extension is. Small percentage of the total, but probably a larger percentage of the Firefox user base, given that Firefox is not pre-installed on most PCs.

Mozilla to Businesses: We're Not Interested (PC Mag)

Posted Jun 27, 2011 0:54 UTC (Mon) by viro (subscriber, #7872) [Link] (3 responses)

IIRC, elanthis was one of the "blocking ads is immoral freeloading and if you do that, you are probably a w4r3z d00d too" crowd...

Mozilla to Businesses: We're Not Interested (PC Mag)

Posted Jun 27, 2011 14:56 UTC (Mon) by branden (guest, #7029) [Link] (1 responses)

You mean Bruce Perens wasn't a crowd of one?

Mozilla to Businesses: We're Not Interested (PC Mag)

Posted Jun 27, 2011 18:44 UTC (Mon) by viro (subscriber, #7872) [Link]

[searching archives] there you go: http://lwn.net/Articles/247869/
Bruce Perens is not alone in that kind of demagogy...

Mozilla to Businesses: We're Not Interested (PC Mag)

Posted Jun 28, 2011 4:40 UTC (Tue) by elanthis (guest, #6227) [Link]

Uh, no, no I was not.

I have AdBlock installed myself.

Mozilla to Businesses: We're Not Interested (PC Mag)

Posted Jun 27, 2011 0:56 UTC (Mon) by PO8 (guest, #41661) [Link] (2 responses)

Surely someone has some actual numbers here? My impression was that most Firefox users have at least a few extensions installed, but I'd really like to know.

Mozilla to Businesses: We're Not Interested (PC Mag)

Posted Jun 27, 2011 6:03 UTC (Mon) by Mook (subscriber, #71173) [Link] (1 responses)

According to their addon site's blog, it's something like 85% for Firefox 4, excluding Personas and "several commonly-bundled add-ons" (I'm guessing things like anti-virus toolbars, but who knows).

Mozilla to Businesses: We're Not Interested (PC Mag)

Posted Jul 2, 2011 11:36 UTC (Sat) by JanC_ (guest, #34940) [Link]

Ubuntu includes an extension that helps to integrate Firefox into the desktop, so I guess that's one of the "commonly-bundled add-ons" too.

Mozilla to Businesses: We're Not Interested (PC Mag)

Posted Jun 27, 2011 16:31 UTC (Mon) by njs (subscriber, #40338) [Link]

And Adblock Plus is already ported to Firefox 7... I think they'll be okay.

Mozilla to Businesses: We're Not Interested (PC Mag)

Posted Jun 27, 2011 17:00 UTC (Mon) by nye (subscriber, #51576) [Link]

>You really don't see a lot of non-nerds using extensions

Sure they do. Yahoo toolbar, Google toolbar, AVG toolbar, AOL toolbar, OMG toolbar, WTF toolbar, BBQ toolbar, the list goes on...

I know, I know, you covered that with 'adware crap' but I think the point needed to be more strongly stated. Browsing the web on non-technical relatives' browsers is like browsing it through a small letterbox.

Mozilla to Businesses: We're Not Interested (PC Mag)

Posted Jun 26, 2011 21:46 UTC (Sun) by njs (subscriber, #40338) [Link] (2 responses)

If I'm counting right, I had 15 extensions installed when Ubuntu rolled out the Firefox 5 upgrade last week, and of those, 14 were detected as compatible and are working perfectly. This includes all the big complicated ones like Firebug and Zotero, and all the critical ones like Adblock and Flashblock. The only one that didn't upgrade is the Firemacs extension that came up in the comments here recently, and I'd actually already disabled it because even in Firefox 4 it did frustrating things like break keyboard control over the awesomebar.

AFAICT this is because Mozilla put a lot of effort into making sure that the 4->5 transition would not be like the 3->4 transition, and it worked. You might be jumping the gun a bit.

Mozilla to Businesses: We're Not Interested (PC Mag)

Posted Jun 27, 2011 6:01 UTC (Mon) by PO8 (guest, #41661) [Link] (1 responses)

I hope it will be better on 4->5 and beyond. I had a couple of extensions break on 3->4 that I cared about. More importantly, my carefully-constructed layout broke. I had to figure out how to set everything up again. Like I said, I just kind of reached my tolerance limit. If things are smoother for a while, I'll likely stay around.

Mozilla to Businesses: We're Not Interested (PC Mag)

Posted Jun 27, 2011 11:06 UTC (Mon) by HenrikH (subscriber, #31152) [Link]

3->4 was a major change. 4->5 is more like 4.01->4.05

Mozilla to Businesses: We're Not Interested (PC Mag)

Posted Jun 26, 2011 22:29 UTC (Sun) by Cato (guest, #7643) [Link] (1 responses)

Have a look at the Addon Compatibility Reporter - it makes it easy to force any extension to run under Firefox 5, and if it works well enough it lets you report that to Mozilla. Should be standard issue for anyone using extensions: https://addons.mozilla.org/en-US/firefox/addon/add-on-com...

Mozilla to Businesses: We're Not Interested (PC Mag)

Posted Jun 27, 2011 6:02 UTC (Mon) by PO8 (guest, #41661) [Link]

Thanks much for the pointer. I'll give it a try.

Mozilla to Businesses: We're Not Interested (PC Mag)

Posted Jun 27, 2011 15:58 UTC (Mon) by nye (subscriber, #51576) [Link] (3 responses)

"Enterprise has never been (and I'll argue, shouldn't be) a focus of ours"

Good for them; somebody had to say it.

Enterprisey people seem to suffer under the delusion that not only do they matter, but that their needs should be put in front of the vast majority of people; as a result, can anyone name 3 pieces of 'enterprise-grade' software that could at least be described as 'not that bad'? How about *one*?

Forcing widely used software to be gimped because of some faceless companys' draconian policies doesn't sound like good idea. I'm sure MS wish they weren't in that bind.

Mozilla to Businesses: We're Not Interested (PC Mag)

Posted Jun 28, 2011 9:36 UTC (Tue) by tialaramex (subscriber, #21167) [Link]

Red Hat Enterprise Linux seems rather better than "not that bad" to me.

Mozilla to Businesses: We're Not Interested (PC Mag)

Posted Jun 28, 2011 18:55 UTC (Tue) by k8to (guest, #15413) [Link]

I work on "enterprise" software that I will not name because it would verge on advertisement, but I can honestly say it's "not that bad". I wouldn't claim much better, so I think your point has some merit.

Mozilla to Businesses: We're Not Interested (PC Mag)

Posted Jul 2, 2011 11:31 UTC (Sat) by JanC_ (guest, #34940) [Link]

There is a huge difference between applications that are enterprise ready (without bragging about it) and applications that are marketed as "enterprise ready". :P

For example Apache seems to be quite enterprise ready to me (they innovate around a stable API and using a plugin-based system), and although there might be better web servers for certain workloads, it will do its job fairly well most of the time...

Mozilla to Businesses: We're Not Interested (PC Mag)

Posted Jun 28, 2011 11:41 UTC (Tue) by Karellen (subscriber, #67644) [Link] (1 responses)

Microsoft have responded

Mozilla to Businesses: We're Not Interested (PC Mag)

Posted Jun 29, 2011 11:15 UTC (Wed) by nye (subscriber, #51576) [Link]

>Microsoft have responded

Yeesh.

Every time I think that LWN has become nothing but a trollfest I just have to look at <the rest of the web> for a breath of foul air.


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