Discussing Mac OS X 10.5 Support Plans

[Josh Aas]

I'd like to propose that we remove support for Mac OS X 10.5 in Firefox 13, which should ship on or near June 5, 2012. To be clear: this is not a decision that has been made, I’m proposing it here in order to get feedback.

First, some facts:

Mac OS X Release Dates) 10.5 was released in October of 2007, 10.6 was released in June of 2009, 10.7 was released in July of 2011.

Mac OS X User Breakdown) The following ADU (Active Daily User) numbers are from November 8, 2011. Mac OS X users make up 6.6% of all Firefox users across all versions going back to Firefox 3. Of those Mac OS X users, 9% are on Mac OS X 10.4, 24% are on Mac OS X 10.5, 53% are on Mac OS X 10.6, and 14% are on Mac OS X 10.7. If you limit the scope to the more recent Firefox 7 release, then only 20% of Mac OS X users are on Mac OS X 10.5.

Trends) Mac OS X 10.5 users have been declining by 1-2% per month (as a share of our total Mac OS X users). This means that when Firefox 13 ships, Mac OS X 10.5 users will likely make up about 13% of Mac OS X users across all versions of Firefox. This number should be around 9% for users of the most recent version of Firefox.

Apple releases new versions of its operating systems relatively quickly and each new version contains significant changes that we must adapt to. This requires resources, and with limited resources this sometimes means we have to make tough decisions about where to invest.

Maintaining Mac OS X 10.5 support consumes a non-trivial portion of the resources we have available for Mac OS X development. Not maintaining Mac OS X 10.5 support will allow us to devote more resources to the product as used by the majority of our Mac OS X users (those on Mac OS X 10.6 and 10.7).

Furthermore, there are already some significant ways in which Firefox on Mac OS X 10.5 has fallen behind Firefox on newer versions of Mac OS X. Accelerated compositing and WebGL are not available on Mac OS X 10.5. Users cannot run plugins out-of-process on Mac OS X 10.5.

Finally, Apple has stopped supporting Mac OS X 10.5. While they do not officially drop support for older OS versions, they have stopped shipping security updates and updating applications like Safari.

[Robert Kaiser]

Josh Aas schrieb:

> Maintaining Mac OS X 10.5 support consumes a non-trivial portion of the resources we have available for Mac OS X development. Not maintaining Mac OS X 10.5 support will allow us to devote more resources to the product as used by the majority of our Mac OS X users (those on Mac OS X 10.6 and 10.7).

We've always only dropped support of platforms because of significant
technical challenges in supporting them, and the fact that TenFourFox
exists tells that even maintaining something for older Macs seems to
have value at times (but it doesn't have to be us doing it ourselves
either, as that example shows).

What are the actual things it costs us?
"Accelerated compositing and WebGL" are good examples you brought, but
we don't support those for a whole range of supported users out there,
on different Windows and Linux versions at least in addition to 10.5
users. What are the real technical problems that bring 10.5 over the
top? I think a decision should mainly base on those, with all the other
data you gave being a reason why it doesn't make sense to put too much
effort into taking on those challenges, but still we need to understand
the challenges first, I think.

Robert Kaiser

--
Note that any statements of mine - no matter how passionate - are never
meant to be offensive but very often as food for thought or possible
arguments that we as a community should think about. And most of the
time, I even appreciate irony and fun! :)

[Justin Lebar]

> Trends) Mac OS X 10.5 users have been declining by 1-2% per month (as a share of our total Mac OS X users). This means that when Firefox 13 ships, Mac OS X 10.5 users will likely make up about 13% of Mac OS X users across all versions of Firefox. This number should be around 9% for users of the most recent version of Firefox.

I'm not sure this is the only relevant metric.  The proportion of all
Mac users on 10.5 is of course decreasing as people buy new Macs.  But
how quickly is the absolute number of 10.5 users decreasing?  How many
people would we be leaving out in the dust if we stopped supporting
10.5, and how does that number compare to OS'es which continue to
support?

-Justin

>> _______________________________________________
>> dev-planning mailing list
>> dev-pl...@lists.mozilla.org
>> https://lists.mozilla.org/listinfo/dev-planning
>>
>

[Jet Villegas]

FWIW, Flash Player already dropped OSX 10.5 with the latest version:
http://www.adobe.com/products/flashplayer/tech-specs.html

Also FWIW, Apple desktop software (eg. iPhoto, Aperture) engineering teams are only required to support two OS versions back.

I think we'll be OK, especially with the June 2012 ship date.

-- Jet

[Aleksandr Milewski]

On 30 Nov, 2011, at 12:59 , Josh Aas wrote:
> Finally, Apple has stopped supporting Mac OS X 10.5. While they do not officially drop support for older OS versions, they have stopped shipping security updates and updating applications like Safari.

The latest Apple has ever released a security update for version N-2 (currently 10.5) is at the same time as the release of the N.1 update (in this case 10.7.1, which shipped on August 16)

The last update that would apply to 10.5.x at all was Quicktime 7.7 on August 3.

http://support.apple.com/kb/HT1222

[Alex Keybl]

> Mac OS X User Breakdown) The following ADU (Active Daily User) numbers are from November 8, 2011. Mac OS X users make up 6.6% of all Firefox users across all versions going back to Firefox 3. Of those Mac OS X users, 9% are on Mac OS X 10.4, 24% are on Mac OS X 10.5, 53% are on Mac OS X 10.6, and 14% are on Mac OS X 10.7. If you limit the scope to the more recent Firefox 7 release, then only 20% of Mac OS X users are on Mac OS X 10.5.
>
> Trends) Mac OS X 10.5 users have been declining by 1-2% per month (as a share of our total Mac OS X users). This means that when Firefox 13 ships, Mac OS X 10.5 users will likely make up about 13% of Mac OS X users across all versions of Firefox. This number should be around 9% for users of the most recent version of Firefox.

Just to be clear, I believe the relevant metric (based upon your estimates) is that you're proposing we EOL ~.6% of our updating user base (although updates to FF8 have not yet plateaued). If we were to assume that all 10.5 Mac users would desire to update at some point in the future (and aren't on PPC), we'd be EOLing ~.9% of our users. So somewhere between a half and one percent would be affected as of FF13.

> Maintaining Mac OS X 10.5 support consumes a non-trivial portion of the resources we have available for Mac OS X development. Not maintaining Mac OS X 10.5 support will allow us to devote more resources to the product as used by the majority of our Mac OS X users (those on Mac OS X 10.6 and 10.7).

There are other similar discussions ongoing with the MSVC 2010 migration. The current criteria being used there is:

Estimated user population being EOL'd (sounds like we have a good idea of this, but we should work with metrics to get a final estimate)
User benefit for users on later versions of the platform (if any)
Engineering benefit (recent related bugs, estimated engineering time)
Strategic considerations

I think this would be good to roll into the MSVC 2010 migration discussion. Together these two scenarios will help us set precedents for EOLing in the future.

-Alex

[Cameron Kaiser]

> > I'd like to propose that we remove support for Mac OS X 10.5 in Firefox 13, which should ship on or near June 5, 2012.

[...]

> I think this would be good to roll into the MSVC 2010 migration discussion. Together these two scenarios will help us set precedents for EOLing in the future.

It seems to me that the best point to end support for an OS would be
on an ESR branch, should that proposal bear fruit; at least they would
get some modicum of updates. It would certainly help stabilize us in
TenFourFox-land, since we could be assured that there would only be
security-related widget bustage. There are certain things in 10.6+ we
probably can't find good equivalents for in the 10.4 SDK (even working
around CoreUI proved tricky, though I think I have most of the old
code merged correctly), so even if the roof collapsed and we fell off
rapid release, we would still be able to merge with and build from a
branch that still got security updates.

Cameron Kaiser

[Henri Sivonen]

On Wed, Nov 30, 2011 at 10:59 PM, Josh Aas <jos...@gmail.com> wrote:
> I'd like to propose that we remove support for Mac OS X 10.5 in Firefox 13, which should ship on or near June 5, 2012. To be clear: this is not a decision that has been made, I’m proposing it here in order to get feedback.

I think it would be valuable to wait until
https://bugzilla.mozilla.org/show_bug.cgi?id=684227 has been fixed.
All the remaining Mac OS X 10.5 users we have using current Firefox
are on Intel, since Firefox 4 already dropped support for PowerPC.
Since Mac OS X 10.6 support all Intel systems that 10.5 ran on, all
10.5 users have a realistic upgrade path that we could inform them
about if a fix for https://bugzilla.mozilla.org/show_bug.cgi?id=684227
enabled us to.

The main problem is finding a way to buy a 10.6 disc, though. B&M
Apple Stores reportedly no longer carry it and refuse to order it, but
it's available from the online Apple Store:
http://store.apple.com/us/product/MC573Z/A?fnode=MTY1NDAzOA If we end
up telling users to upgrade their OS, we should probably link all the
way to a page that allows them to order the disc.

Users might lack a realistic upgrade path only if they have other
important software that no longer runs on 10.6.

What I'm saying above about all 10.5 users having an upgrade path to
10.6 includes users of early Intel Macs with 32-bit CPUs. If 10.5
support is dropped, is the plan still to keep 32-bit code in a fat
binary only for 10.6 users who have 32-bit CPUs?

> If you limit the scope to the more recent Firefox 7 release, then only 20% of Mac OS X users are on Mac OS X 10.5.

One in five is a big portion of our Mac user base to abandon. I'm
uncomfortable with abandoning that big a proportion of our Mac users,
but I don't feel the downside of continued 10.5 support, because I
don't work on platform-specific code.

> Finally, Apple has stopped supporting Mac OS X 10.5. While they do not officially drop support for older OS versions, they have stopped shipping security updates and updating applications like Safari.

It's worth noting that Chrome and Opera haven't dropped 10.5 support
yet. If Chrome and Opera still support 10.5 when we drop support, will
we value the security of our users more than we value not giving users
to competitors and inform the users about the possibility of switching
to Chrome or Opera? Or are we willing to instruct users to run
TenFourFox on Rosetta?

--
Henri Sivonen
hsiv...@iki.fi
http://hsivonen.iki.fi/

[Boris Zbarsky]

On 12/1/11 2:21 AM, Henri Sivonen wrote:
> What I'm saying above about all 10.5 users having an upgrade path to
> 10.6 includes users of early Intel Macs with 32-bit CPUs. If 10.5
> support is dropped, is the plan still to keep 32-bit code in a fat
> binary only for 10.6 users who have 32-bit CPUs?

The plan is to keep the 32-bit code so we can run 32-bit plug-ins, I
would think. At that point support for 32-bit-only CPUs is sort of free.

-Boris

[Henri Sivonen]

Doing QA on a 32-bit Gecko seems like a very different issue from
supporting a 32-bit plug-in container process even if getting a 32-bit
plug-in container meant building the whole app as a fat binary.

[Gervase Markham]

On 30/11/11 20:59, Josh Aas wrote:
> I'd like to propose that we remove support for Mac OS X 10.5 in
> Firefox 13, which should ship on or near June 5, 2012. To be clear:
> this is not a decision that has been made, I’m proposing it here in
> order to get feedback.

A couple of questions, the answers to which which may have some bearing:

1) Is the TenFourFox team interested in expanding their OS range, or
doing a TenFiveFox?

2) What would we recommend to 10.5 users to do when it is no longer
supported? Here is our current advice:
https://wiki.mozilla.org/Firefox/Unsupported_OSes

(Relatedly: if 10.4 and 10.5 are no longer supported by Apple, should we
be recommending TenFourFox, or should we be telling people to upgrade,
for the reasons given at the bottom of that page?)

3) Is there a possibility that we might recommend an ESR to people who
are unable to run the latest Firefox but still want something which is
supported? (I suspect not, but it's worth asking the question.)

Gerv

[Cameron Kaiser]

> 1) Is the TenFourFox team interested in expanding their OS range, or
> doing a TenFiveFox?

This is really a three part question:

- We do support 10.5 PPC and have users who run it on 10.5. It is the
"minority" OS based on my version ping data, but a significant
minority.
- We only build against the 10.4 SDK, because we specifically support
G3 and Classic users. Heck, all of my PPCs and my daily driver quad G5
run 10.4; Power Macs seem to bog down in Leopard. I certainly have no
plans to drop support for it -- it's just a matter of what we can hack
to build.
- I have no plans to build or support an Intel 10.5 version
specifically. I only have a single Intel C2D Mac, and it will shortly
be running Lion so I can still use it for Android development (that's
all I use it for). People can run TenFourFox in Rosetta (unsupported)
and this is known to work, and if someone takes the 10.4Fx patches and
spins them into an Intel version I would be happy to direct people to
that project, but 10.4Fx will itself always be PPC. This would not be
difficult to manage; they would just need to undo the PowerPC-specific
stuff. The rest of it should "just work." It may need some minor
changes to build properly against the 10.5 SDK since I don't test
that.

> 2) What would we recommend to 10.5 users to do when it is no longer
> supported? Here is our current advice:https://wiki.mozilla.org/Firefox/Unsupported_OSes
>
> (Relatedly: if 10.4 and 10.5 are no longer supported by Apple, should we
> be recommending TenFourFox, or should we be telling people to upgrade,
> for the reasons given at the bottom of that page?)

All 10.5 Intel Macs can upgrade, given sufficient RAM, to 10.6. There
are good reasons for Power Mac users not moving to 10.5, but I'm not
aware of any reasons for any Intel owner to stay with 10.5. (10.4,
yes, because it could still run CFM Carbon binaries with Rosetta, IIRC
-- 10.5 only runs Mach-O.)

> 3) Is there a possibility that we might recommend an ESR to people who
> are unable to run the latest Firefox but still want something which is
> supported? (I suspect not, but it's worth asking the question.)

This to me really seems like the best option. It's pretty much the
role 3.6 is serving now, unofficially.

Cameron Kaiser

[Alex Keybl]

> 3) Is there a possibility that we might recommend an ESR to people who are unable to run the latest Firefox but still want something which is supported? (I suspect not, but it's worth asking the question.)

The ESR branch is meant for organizations and institutions with their own internal support structure in order to enable them to qualify and roll out new releases on a timeline that they're comfortable with. Sending our other users to a release which doesn't have the full support and attention that our newer releases have likely isn't the right call. If we were to start sending non-organizational users to the ESR, it would quickly grow the scope of ESR in a way we've been trying to avoid.

-Alex

[Jonathan Kew]

On 1 Dec 2011, at 17:08, Alex Keybl wrote:

>> 3) Is there a possibility that we might recommend an ESR to people who are unable to run the latest Firefox but still want something which is supported? (I suspect not, but it's worth asking the question.)
>

> The ESR branch is meant for organizations and institutions with their own internal support structure in order to enable them to qualify and roll out new releases on a timeline that they're comfortable with. Sending our other users to a release which doesn't have the full support and attention that our newer releases have likely isn't the right call.

If we ship an ESR in early 2012 (for example), my understanding is that we'll "support" that with critical security updates for at least a year or so. OTOH, a 10.5 user on the non-ESR track would receive _no_ updates even for critical security issues as soon as FF13 ships at the beginning of June (under Josh's proposal).

Surely it's better to suggest 10.5 users who can't update their OS should move to Firefox ESR than to leave them on a non-ESR release for which they can no longer receive _any_ updates, because the only supported update we offer for it won't run on their platform.

They'll still face the problem once the ESR is EOL'd, of course, but at least that would buy them some additional time.

JK

> If we were to start sending non-organizational users to the ESR, it would quickly grow the scope of ESR in a way we've been trying to avoid.
>
> -Alex
>
> On Dec 1, 2011, at 3:50 AM, Gervase Markham wrote:
>

[Post master]

I'm sorry to intrude into this discussion,
but i want you to give an user feedback (and not a developper)
(english is not my mother tongue, i hope you'll excuse any mistake i
can make)

the facts :
- i'm working in computer graphics company, and like many other we're
nearly no other choice than work on Mac (specially for print),
- Mac create a new version of Mac OS X nearly each 2 years, it may
seems a lot for software, but in fact, most user change of Mac OS X
version ONLY when computer is out of order, burn or unable to do his
job.
- In many case, you can't upgrade OS version on a Mac because of
hardware problem, or apple restriction (that was specially the case
for 10.3, but it still happens)
- Most of the time, a company changes his computer each 5 to 7 years,
and by the way, have a new OS X version/license in the same time.
- Most of the time too, a company don't install new OS X before 4 or 6
months on computer after gold version (because of stability of os, and
because of software who are not still compatible with the os, ...)
- this was really an huge problem with the end of OS 10.3 support.

If you're dropping OS 10.5 support, it means that you'll drop support
for a period going from 2 to 4 years for any user in this situation...
4 years !!!

To give you an idea of our current situation :
here, we are currently using around
- 5% 10.3,
- 90% of 10.5,
- 5% 10.6,
and this will not gonna change before at last 1 or 2 years at last,
with, or without firefox.


i know that firefox has to focus ressource,
i know you're doing a fantastic work,
but it not a solution to make the final user in a situation where
he'll not have anymore choice than using chrome

last, but not least :
i'll not speak here about checking the compatibility of website on
next version of firefox, or so one, but that may become a true problem
too.


in my mind, there'll be (i don't know if it's possible) an alternative
solution :
since the main problem is that some component are not compatible/
support by the system, why not create a "degraded" (omega ?) version
who will simply shutdown this component.
Honestly, i nearly don't know anyone who have use of webgl, websocket
and some useless feature who looks so nice on roadmap but who are and
who wont be used before years.
I prefered a browser who works perfectly than a browser full of
useless features

Like that, it'll be clear, when a version goes "omega", you'll have
not all features, but the most usefull, and security issues will be
fixed.

thank you for reading this, and please take that in consideration

regards,
Sil

[vasi]

On Dec 1, 9:54 am, Cameron Kaiser <ckai...@floodgap.com> wrote:
> ... I'm not

> aware of any reasons for any Intel owner to stay with 10.5. (10.4,
> yes, because it could still run CFM Carbon binaries with Rosetta, IIRC
> -- 10.5 only runs Mach-O.)

I am running CFM binaries perfectly well in even 10.6. Maybe you're
thinking of something else?

-Dave

PS: Classilla and 10.4Fx are awesome, thanks!

[Gervase Markham]

Hi Sil,

On 02/12/11 10:14, Post master wrote:
> I'm sorry to intrude into this discussion,
> but i want you to give an user feedback (and not a developper)
> (english is not my mother tongue, i hope you'll excuse any mistake i
> can make)

You are very welcome to contribute.

I understand that you have machines running 10.5, but if Apple is no
longer providing OS security fixes for them, then those machines are
insecure and should not be put on the Internet. This is true whether you
use Firefox or anything else.

> but it not a solution to make the final user in a situation where
> he'll not have anymore choice than using chrome

Chrome only supports 10.5+ Intel, according to this:
http://www.google.com/support/chrome/bin/answer.py?hl=en&answer=95411

Gerv

[hart...@uni-freiburg.de]

Maybe not. I read that the latest Safari-Bug isn't reproducable on
MacOS 10.5, so there was no need for an update.

Please note that MacOS 10.5 is the last version that support Macs with
PowerPC-CPU. All later versions run only on Intel hardware.

Because of this special situation we hope that Apple and Mozilla will
support 10.5 longer than usual.

[hart...@uni-freiburg.de]

On 30 Nov., 23:43, Jet Villegas <j...@mozilla.com> wrote:
> FWIW, Flash Player already dropped OSX 10.5 with the latest version:http://www.adobe.com/products/flashplayer/tech-specs.html

But they do offer an updated Flash version 10.3.183.11 for MacOS 10.5,
with all known security vulnerabilities fixed.
However they have dropped PPC-Support already with version 10.2.
Version 10.3 requires an Intel-CPU. And all Macs with Intel-CPU could
be upgraded to MacOS 10.6.

[Kyle Huey]

Mozilla no longer supports PowerPC on current releases, and has not since
Firefox 3.6.

- Kyle

[Tommy B]

firefox needs to stop supporting so many platforms.

Who is firefox trying to target?

Mac OS X 10.5 is apple's problem. It is their users. They paid big
money to apple, not mozilla. They can use safari.

Mozilla seems to be spreading pretty thin, It is losing on so many
fronts. It needs to go back and win back power users that want speed.
It's time to leave laggards behind. I dont understand the point of
being apple's end of life support, mozilla doesn't get paid for it.

Mozilla needs to allocate their resources better and put it in making
the fastest and most flexible browser.
Let apple support 10.5, let microsft deal with windows 98, me, 2000.

Moziila will cease to exist if it thinks it can be everything to
everybody. I think mozilla needs to learn how to prune a tree. You
need need to cut of some branches to improve the strength of the whole
plant.

[Post master]

On 7 déc, 23:58, Tommy B <tommy...@gmail.com> wrote:

> Mac OS X 10.5 is apple's problem. It is their users. They paid big
> money to apple, not mozilla. They can use safari.

Don't forget a large part of Mac OS X user use it on graphical
purpose, to create website, and so one
So, not only you'll tons of people who are using firefox + 10.5 to
test and support website for firefox
but you'll loose lot's of prescriber user who help to spread firefox
on "simple" user.

Like a small green man said, "Do or Do not. There is no try."
If you choose to do something, do it fully :)

[Post master]

On 5 déc, 17:50, Gervase Markham <g...@mozilla.org> wrote:

> You are very welcome to contribute.

thank you :)

>
> I understand that you have machines running 10.5, but if Apple is no
> longer providing OS security fixes for them, then those machines are
> insecure and should not be put on the Internet. This is true whether you
> use Firefox or anything else.

10.5 may be not anymore supported by apple, but there's still security
update :
by example : http://support.apple.com/kb/HT1222?viewlocale=en_US

>
> Chrome only supports 10.5+ Intel, according to this:http://www.google.com/support/chrome/bin/answer.py?hl=en&answer=95411

Well, indeed, chrome stills support 10.5, even if it's not anymore the
case with flash
By the way, Flash is not the web (and that's not a bad thing)

Sil

[Gervase Markham]

On 16/12/11 14:38, Post master wrote:
> 10.5 may be not anymore supported by apple, but there's still security
> update :
> by example : http://support.apple.com/kb/HT1222?viewlocale=en_US

The last update for Safari on that page is July 2011 (5.0.6). Since
then, 5.1 and 5.1.1 have been released - for 10.6 and above only.
Similarly, the 9th September and 12th October security updates were 10.6
and above only.

Yes, they are still releasing some software for 10.5 (e.g. iTunes), but
it seems not security updates.

Gerv

[Jerzy]

Please don't EOL Firefox on Mac OS 10.5, TenFourFox (based on it) is
the last supported browser on my PowerBook! Opera and Safari are
abandoned, Chrome doesn't work, many apps are not supported too, but
please allow us to see HTML5 at least!

[djv...@gmail.com]

It looks like Apple is at least minimally still supporting 10.5 in security updates: http://support.apple.com/kb/DL1533

-V

[emichae...@gmail.com]

It has been months since the last post, here, and a lot has happened since then, so here are my thoughts about this. Basically, the next ESR should be the last version of 10.5 support. FF16 should include an instruction that all OS 10.5 Macs automatically download the ESR of version 17, when that version is detected as available for download. That way, 10.5 users wouldn't have to manually do anything, and would be able to keep getting security updates on some form of Firefox for another year. By the end of that ESR's normally scheduled EOL (early 2014), 10.5 marketshare should be so low, that EOLing then shouldn't be a problem.

For those left on 10.5 at that time, TenFourFox with Rosetta could work. Another option would be Camino, a universal binary. As of now, this browser uses the FF3.6 rendering engine, however. Additionally, it hasn't been updated in several months, likely because Mozilla discontinued development of the FF3.6 rendering engine; hopefully, the Camino developers can find a solution to this. Yet another possibility could be OmniWeb, based on Webkit.

On Thursday, December 1, 2011 12:08:04 PM UTC-5, Alex Keybl wrote:
> The ESR branch is meant for organizations and institutions with their own internal support structure in order to enable them to qualify and roll out new releases on a timeline that they're comfortable with. Sending our other users to a release which doesn't have the full support and attention that our newer releases have likely isn't the right call. If we were to start sending non-organizational users to the ESR, it would quickly grow the scope of ESR in a way we've been trying to avoid.

It wouldn't "quickly grow" the scope of the ESR's purpose. It would grow the scope a little, technically, but it wouldn't dilute the ESR's intentions towards enterprise, which is what's important. I can't see how it would hurt to provide over another year of security updates of some kind to those that are left on OS 10.5, versus providing none at all.

[Robert Kaiser]

emichae...@gmail.com schrieb:

> FF16 should include an instruction that all OS 10.5 Macs automatically download the ESR of version 17, when that version is detected as available for download.

We've stated before that we re not willing to do things like that, as we
don't want to point normal users to ESR. Mozilla doesn't provide direct
support for ESR to users, as ESR is only there for mass-rollout by
organizations, which then are on the hook for supporting their users,
while Mozilla will only directly support the organization admins.

Robert Kaiser

[Kyle Huey]

On Sun, Aug 19, 2012 at 12:55 AM, <emichae...@gmail.com> wrote:

> It has been months since the last post, here, and a lot has happened since
> then, so here are my thoughts about this.

It has been months since the last post because a decision was made and
executed. 10.5 support has been removed for Firefox 17.

> Basically, the next ESR should be the last version of 10.5 support. FF16
> should include an instruction that all OS 10.5 Macs automatically download
> the ESR of version 17, when that version is detected as available for
> download. That way, 10.5 users wouldn't have to manually do anything, and
> would be able to keep getting security updates on some form of Firefox for
> another year. By the end of that ESR's normally scheduled EOL (early
> 2014), 10.5 marketshare should be so low, that EOLing then shouldn't be a
> problem.
>

And we decided not to do that, because 10.5 marketshare will already be
quite low when Firefox 17 is released and our policy is not to move regular
users to ESR for the reasons that have been stated by akeybl.

For those left on 10.5 at that time, TenFourFox with Rosetta could work.
> Another option would be Camino, a universal binary. As of now, this
> browser uses the FF3.6 rendering engine, however. Additionally, it hasn't
> been updated in several months, likely because Mozilla discontinued
> development of the FF3.6 rendering engine; hopefully, the Camino developers
> can find a solution to this. Yet another possibility could be OmniWeb,
> based on Webkit.
>

Yes, there are potentially other options out there for users who choose to
remain on OS X 10.5 even though it is unsupported by Apple. The
availability of those options is not our concern.

On Thursday, December 1, 2011 12:08:04 PM UTC-5, Alex Keybl wrote:
> > The ESR branch is meant for organizations and institutions with their
> own internal support structure in order to enable them to qualify and roll
> out new releases on a timeline that they're comfortable with. Sending our
> other users to a release which doesn't have the full support and attention
> that our newer releases have likely isn't the right call. If we were to
> start sending non-organizational users to the ESR, it would quickly grow
> the scope of ESR in a way we've been trying to avoid.
>
> It wouldn't "quickly grow" the scope of the ESR's purpose. It would grow
> the scope a little, technically, but it wouldn't dilute the ESR's
> intentions towards enterprise, which is what's important. I can't see how
> it would hurt to provide over another year of security updates of some kind
> to those that are left on OS 10.5, versus providing none at all.
>

It would grow the scope because it would require SUMO to support end users
on ESR, which we don't currently. It would "hurt" to provide another
year's worth of support for OS X 10.5 because we have to keep around
testing infrastructure, QA processes, etc for backporting patches to ESR.
We've decided
<https://groups.google.com/d/msg/mozilla.dev.platform/aT7hy7YDdqA/dPmpfrDOR1wJ>that
the cost of keeping 10.5 support for an additional release is too high for
the benefit received.

- Kyle