Dropping Mac OS X 10.4 support in Gecko 1.9.3

classic Classic list List threaded Threaded
185 messages Options
1234 ... 10
Reply | Threaded
Open this post in threaded view
|

Dropping Mac OS X 10.4 support in Gecko 1.9.3

Mulder-7
While Mozilla may think that dropping support for Mac OS X 10.4 is a
good idea, you're dead wrong. There is no need to do this; it's a
short-sighted plan to avoid supporting well over a million users who
are still running 10.4 for various reasons; i.e., they can't afford to
upgrade their equipment; 10.5 and 10.6 drop or break features that
work perfectly well in 10.4; that upgrading would require them to
purchase new versions of software that works perfectly well under
10.4; or even that many people think both 10.5 and 10.6 suck.

I don't care what Apple's unstated plans are about Java1 and Java2
plugins; they may come to pass or not, but we won't know until it
actually happens. That's no reason to drop support for Firefox 3.5.x
or 3.6.x so soon after releasing them, assuming your delivery schedule
is accurate for the next major release (it's never been accurate).

You can't claim poverty, either. Mozilla takes in many millions of
dollars per year from Google by having their search engine as the
default in Firefox, so don't even try to use that as a justification
for keeping your software working.

If you're going to drop support for 10.4 no matter what people say,
then what's the point of asking for our feedback? When you can explain
that, you might be getting to the root of the problem: yourselves.
_______________________________________________
dev-planning mailing list
[hidden email]
https://lists.mozilla.org/listinfo/dev-planning
Reply | Threaded
Open this post in threaded view
|

Re: Dropping Mac OS X 10.4 support in Gecko 1.9.3

Robert Kaiser
Mulder wrote:
> While Mozilla may think that dropping support for Mac OS X 10.4 is a
> good idea, you're dead wrong. There is no need to do this

1) It's not about "good" vs. "bad" idea, it's about how much pain it is
to support 10.4, 10.5, 10.6 32bit and 10.6 64bit all at the same time,
and how much time we as a community can invest into developing for and
testing on all those systems - and all that in the end for a rather tiny
minority of our total users, which the whole Mac user base still is. We
all know that it's not a "good idea", the question is how much it is a
necessity for going forward overall, as we have limited resources.

2) How do you know there's "no need to do this"? Have you actually
looked at the code and found a way to easily support 10.4 in addition to
the newer systems? I'm sure Josh as our main Mac system support
developer would be very happy to see that option and its implementation
and would welcome your work on it.

Robert Kaiser
_______________________________________________
dev-planning mailing list
[hidden email]
https://lists.mozilla.org/listinfo/dev-planning
Reply | Threaded
Open this post in threaded view
|

Re: Dropping Mac OS X 10.4 support in Gecko 1.9.3

James Rohde
On Feb 5, 6:07 pm, Robert Kaiser <[hidden email]> wrote:

> 1) It's not about "good" vs. "bad" idea, it's about how much pain it is
> to support 10.4, 10.5, 10.6 32bit and 10.6 64bit all at the same time,
> and how much time we as a community can invest into developing for and
> testing on all those systems - and all that in the end for a rather tiny
> minority of our total users, which the whole Mac user base still is. We
> all know that it's not a "good idea", the question is how much it is a
> necessity for going forward overall, as we have limited resources.
>
> 2) How do you know there's "no need to do this"? Have you actually
> looked at the code and found a way to easily support 10.4 in addition to
> the newer systems? I'm sure Josh as our main Mac system support
> developer would be very happy to see that option and its implementation
> and would welcome your work on it.
>
> Robert Kaiser

Robert, your own bias appears to be showing... (Not meant as a
criticism, but the reality is, it's not "your ox that's being gored",
as you don't seem to have any personal investment on behalf of the Mac
version)

_______________________________________________
dev-planning mailing list
[hidden email]
https://lists.mozilla.org/listinfo/dev-planning
Reply | Threaded
Open this post in threaded view
|

Re: Dropping Mac OS X 10.4 support in Gecko 1.9.3

Asa Dotzler-2
In reply to this post by Mulder-7
On 2/5/2010 3:18 PM, Mulder wrote:
> While Mozilla may think that dropping support for Mac OS X 10.4 is a
> good idea, you're dead wrong. There is no need to do this; it's a
> short-sighted plan to avoid supporting well over a million users who
> are still running 10.4 for various reasons;

And you think it will still be well over 1 million users a year from now?


>
> You can't claim poverty, either. Mozilla takes in many millions of
> dollars per year from Google by having their search engine as the
> default in Firefox, so don't even try to use that as a justification
> for keeping your software working.


No one is claiming poverty. We are talking about how we best utilize our
limited resources. I can count the serious Mac platform experts we have
on one hand and splitting those resources more than is critically
necessary isn't something I'm excited about. One of those top experts is
making this proposal and I doubt he'd be making it if he thought we had
the resources to easily support 10.4 and 10.5&6.


> If you're going to drop support for 10.4 no matter what people say,
> then what's the point of asking for our feedback? When you can explain
> that, you might be getting to the root of the problem: yourselves.

Can you provide further insight than we already have into the number of
people using 10.4 or using Firefox on 10.4 or into Apple's future
support schedule or how much work is actually involved in maintaining
support for 10.4?  Those kinds of things would be useful feedback.  "I'm
on 10.4 so you're stupid for dropping it in a year" isn't valuable
feedback.

- A
_______________________________________________
dev-planning mailing list
[hidden email]
https://lists.mozilla.org/listinfo/dev-planning
Reply | Threaded
Open this post in threaded view
|

Re: Dropping Mac OS X 10.4 support in Gecko 1.9.3

PhillipJones
Asa Dotzler wrote:
> On 2/5/2010 3:18 PM, Mulder wrote:
>> While Mozilla may think that dropping support for Mac OS X 10.4 is a
>> good idea, you're dead wrong. There is no need to do this; it's a
>> short-sighted plan to avoid supporting well over a million users who
>> are still running 10.4 for various reasons;
>
> And you think it will still be well over 1 million users a year from now?

If the economy doesn't get better - yes.

-------------------snip-------------------

> - A


--
Phillip M. Jones, C.E.T.    "If it's Fixed, Don't Break it"
http://www.phillipmjones.net           http://www.vpea.org
mailto:[hidden email]
_______________________________________________
dev-planning mailing list
[hidden email]
https://lists.mozilla.org/listinfo/dev-planning
Reply | Threaded
Open this post in threaded view
|

Re: Dropping Mac OS X 10.4 support in Gecko 1.9.3

Teoli
In reply to this post by Robert Kaiser
On 06.02.10 01:07, Robert Kaiser wrote:
> it's about how much pain it is to support 10.4, 10.5, 10.6 32bit
 > and 10.6 64bit all at the same time,

I think you should also take in account the Intel/PPC support.

It means that keeping 10.4, Mozilla have to test/support:
10.4 PPC
10.4 x86
10.5 PPC
10.5 x86
10.5 x64
10.6 x86
10.6 x64

Abandoning 10.4, means 29% of OS/CPU combination less.

If concerns arise about the 10.4 user base, extend the security-fix
support of 3.6 from mid-2011 to end-2011, it should be less work (but
still work) to maintain an old branch for security rather than to
back-port anything new; especially as other products (Tb?) may still
need Gecko 1.9.3 by then.

Note that I don't know if you plan to support x64 on 10.5. I don't think
it is useful.

--
Teoli
_______________________________________________
dev-planning mailing list
[hidden email]
https://lists.mozilla.org/listinfo/dev-planning
Reply | Threaded
Open this post in threaded view
|

Re: Dropping Mac OS X 10.4 support in Gecko 1.9.3

Blair McBride (Unfocused)
In reply to this post by PhillipJones
On 6/02/2010 4:34 p.m., Phillip Jones wrote:
>> And you think it will still be well over 1 million users a year from now?
>
> If the economy doesn't get better - yes.

There may well be 1 million users using 3.5 and 3.6 in a year's time.
And those versions will continue to support 10.4. But the majority of
those on 10.4 won't upgrade to 3.next, even if it has 10.4 support. As
it is, very few 10.4 users upgraded to 3.6. Looking at the numbers, 24%
of 3.5 Mac users run 10.4; while only 12% of 3.6 Mac users run 10.4. I'd
say 6% on 3.next would be a very optimistic number - and at great
development cost.

- Blair
_______________________________________________
dev-planning mailing list
[hidden email]
https://lists.mozilla.org/listinfo/dev-planning
Reply | Threaded
Open this post in threaded view
|

Re: Dropping Mac OS X 10.4 support in Gecko 1.9.3

Mulder-7
In reply to this post by Asa Dotzler-2
On Feb 5, 9:05 pm, Asa Dotzler <[hidden email]> wrote:
> On 2/5/2010 3:18 PM, Mulder wrote:
>
> > While Mozilla may think that dropping support for Mac OS X 10.4 is a
> > good idea, you're dead wrong. There is no need to do this; it's a
> > short-sighted plan to avoid supporting well over a million users who
> > are still running 10.4 for various reasons;
>
> And you think it will still be well over 1 million users a year from now?

That's a strawman argument. Nobody knows what the situation will be a
year from now, so it's impossible for anyone to answer that question.

>
>
>
> > You can't claim poverty, either. Mozilla takes in many millions of
> > dollars per year from Google by having their search engine as the
> > default in Firefox, so don't even try to use that as a justification
> > for keeping your software working.
>
> No one is claiming poverty. We are talking about how we best utilize our
> limited resources. I can count the serious Mac platform experts we have
> on one hand and splitting those resources more than is critically
> necessary isn't something I'm excited about.

If you currently have limited resources (programmers), try using some
of the $25 million or more you collect every year from Google to hire
more Mac OS X programmers. Not only would that add to the expertise
you have, it would make your development cycle move faster.

> One of those top experts is making this proposal and I doubt he'd be making it
> if he thought we had the resources to easily support 10.4 and 10.5&6.

You might not currently have the resources you need to support all
three versions, but you have more than enough money to get those
resources; all it takes is the courage to do it and the willingness to
support your users. Alienating users by dropping support when there's
no valid technological reason only makes more users unhappy and causes
them to switch again to another browser, which is a never-ending
cycle.

Perhaps you should consider serving users better by making the browser
smaller and much faster, instead of focusing on feature creep. You
also need to make the auto-upgrade feature work for everyone, even if
they have stripped the useless Intel or PowerPC code from it for use
on their system. If that involves completely recoding everything, so
be it. As it stands, it's painfully slow to launch on PowerPC systems,
while Safari is very fast, even at version 4.0.4.

> > If you're going to drop support for 10.4 no matter what people say,
> > then what's the point of asking for our feedback? When you can explain
> > that, you might be getting to the root of the problem: yourselves.
>
> Can you provide further insight than we already have into the number of
> people using 10.4 or using Firefox on 10.4 or into Apple's future
> support schedule or how much work is actually involved in maintaining
> support for 10.4?  Those kinds of things would be useful feedback.  "I'm
> on 10.4 so you're stupid for dropping it in a year" isn't valuable
> feedback.

I can't see the future, so I have no idea what the usage numbers will
be, just as you have no idea. I'm sure you're well aware that Apple
doesn't comment on future plans and I certainly can't make them talk.
I do know that there are millions of Mac OS X 10.4 users who either
cannot or will not upgrade, for a variety of reasons, all of them
perfectly valid for their needs. I also know that Mac OS X 10.5 and
10.6 are riddled with serious bugs and had features removed without
notice; something Apple doesn't publicize or disclose to anyone before
or after they purchase the OS or a system with those versions
installed. Users are left to discover it on their own, and it's always
when they need it to work correctly; that's too late.

I've never suggested or stated that anyone was stupid for dropping
support for 10.4, so implying that I did is intellectually dishonest.
The issue is that you're making a broad assumption that the users
you'd be abandoning will switch to 10.5 or 10.6, and based on current
usage numbers. there's no evidence of that. I certainly won't, and I'm
confident most others on 10.4 won't, either. They'll use something
else, whether it's Safari, Camino, iCab, Opera, or Shiira.

So the question you need to ask yourselves before pursuing this ill-
considered idea is: "Do you want to spend the money to hire the
programmers you need to support 10.4, or do you want to give up
millions in revenue?"
_______________________________________________
dev-planning mailing list
[hidden email]
https://lists.mozilla.org/listinfo/dev-planning
Reply | Threaded
Open this post in threaded view
|

Re: Dropping Mac OS X 10.4 support in Gecko 1.9.3

L. David Baron
In reply to this post by Blair McBride (Unfocused)
On Sunday 2010-02-07 00:19 +1300, Blair McBride wrote:
> There may well be 1 million users using 3.5 and 3.6 in a year's
> time. And those versions will continue to support 10.4. But the
> majority of those on 10.4 won't upgrade to 3.next, even if it has
> 10.4 support. As it is, very few 10.4 users upgraded to 3.6. Looking
> at the numbers, 24% of 3.5 Mac users run 10.4; while only 12% of 3.6
> Mac users run 10.4. I'd say 6% on 3.next would be a very optimistic
> number - and at great development cost.

I think numbers for Firefox 3.6 users will change substantially once
we make a prompted major update offer from 3.5 to 3.6.  Please note
that the 3.6 numbers Josh posted were much smaller than the 3.5
numbers (basically 5.2% of the 3.5+3.6 users were on 3.6, and 94.8%
were on 3.5).

The 3.6 users right now are presumably largely people who are
interested enough in technology that they read sources of news that
told them about the 3.6 release, users who I'd expect are more
likely to be on the cutting edge in terms of OS versions as well.

-David

--
L. David Baron                                 http://dbaron.org/
Mozilla Corporation                       http://www.mozilla.com/
_______________________________________________
dev-planning mailing list
[hidden email]
https://lists.mozilla.org/listinfo/dev-planning
Reply | Threaded
Open this post in threaded view
|

Re: Dropping Mac OS X 10.4 support in Gecko 1.9.3

PhillipJones
In reply to this post by Blair McBride (Unfocused)
L. David Baron wrote:

> On Sunday 2010-02-07 00:19 +1300, Blair McBride wrote:
>> There may well be 1 million users using 3.5 and 3.6 in a year's
>> time. And those versions will continue to support 10.4. But the
>> majority of those on 10.4 won't upgrade to 3.next, even if it has
>> 10.4 support. As it is, very few 10.4 users upgraded to 3.6. Looking
>> at the numbers, 24% of 3.5 Mac users run 10.4; while only 12% of 3.6
>> Mac users run 10.4. I'd say 6% on 3.next would be a very optimistic
>> number - and at great development cost.
>
> I think numbers for Firefox 3.6 users will change substantially once
> we make a prompted major update offer from 3.5 to 3.6.  Please note
> that the 3.6 numbers Josh posted were much smaller than the 3.5
> numbers (basically 5.2% of the 3.5+3.6 users were on 3.6, and 94.8%
> were on 3.5).
>
> The 3.6 users right now are presumably largely people who are
> interested enough in technology that they read sources of news that
> told them about the 3.6 release, users who I'd expect are more
> likely to be on the cutting edge in terms of OS versions as well.
>
> -David
>

I am type person that updates either automatically through auto update
or going to web site and updating. In in the Past I tested nightly until
I figured what I said about bugs or suggestion were laughed at or
ignored. since users don't know anything.

--
Phillip M. Jones, C.E.T.    "If it's Fixed, Don't Break it"
http://www.phillipmjones.net           http://www.vpea.org
mailto:[hidden email]
_______________________________________________
dev-planning mailing list
[hidden email]
https://lists.mozilla.org/listinfo/dev-planning
Reply | Threaded
Open this post in threaded view
|

Re: Dropping Mac OS X 10.4 support in Gecko 1.9.3

Shawn Wilsher-3
In reply to this post by Mulder-7
On 2/6/2010 6:44 AM, Mulder wrote:
> If you currently have limited resources (programmers), try using some
> of the $25 million or more you collect every year from Google to hire
> more Mac OS X programmers. Not only would that add to the expertise
> you have, it would make your development cycle move faster.
>
>    
Have you read The Mythical Man Month?  Adding more people will not make
us move faster in the short term, and it may not in the long term either.

Besides that, you are arguing that we should spend a large sum of money
on small percentage of our users (a shrinking percentage at that).  What
would you say to all of our windows users who would then see less
resources devoted to them?
> You might not currently have the resources you need to support all
> three versions, but you have more than enough money to get those
> resources; all it takes is the courage to do it and the willingness to
> support your users. Alienating users by dropping support when there's
> no valid technological reason only makes more users unhappy and causes
> them to switch again to another browser, which is a never-ending
> cycle.
>    
I'm sorry, but Josh covered the technical reasons in his first post.  
You can certainly try to argue that they aren't issues, but simply
dismissing them is not the way to do that.

> I can't see the future, so I have no idea what the usage numbers will
> be, just as you have no idea. I'm sure you're well aware that Apple
> doesn't comment on future plans and I certainly can't make them talk.
> I do know that there are millions of Mac OS X 10.4 users who either
> cannot or will not upgrade, for a variety of reasons, all of them
> perfectly valid for their needs. I also know that Mac OS X 10.5 and
> 10.6 are riddled with serious bugs and had features removed without
> notice; something Apple doesn't publicize or disclose to anyone before
> or after they purchase the OS or a system with those versions
> installed. Users are left to discover it on their own, and it's always
> when they need it to work correctly; that's too late.
>    
What are these serious bugs and removed features you speak of?  If you
are trying to convince us that this is a serious issue, I suggest you
make actual claims instead of hand-wavy references to problems that come
across as FUD.

> So the question you need to ask yourselves before pursuing this ill-
> considered idea is: "Do you want to spend the money to hire the
> programmers you need to support 10.4, or do you want to give up
> millions in revenue?"
>    
It's not even about hiring programmers.  It's also about buying more
hardware, and supporting it.  It's also about using our resources in the
most beneficial way for our users.  You are arguing that we should spend
a disproportionate amount of our resources on a small and shrinking
(even if the number of 10.4 users stays the same, they will make up a
smaller percentage of our users over time) percentage of our user base.  
We are not a massive corporation like Apple, so we have to use our
limited resources in the best possible way for our users.

Cheers,

Shawn


_______________________________________________
dev-planning mailing list
[hidden email]
https://lists.mozilla.org/listinfo/dev-planning
Reply | Threaded
Open this post in threaded view
|

Re: Dropping Mac OS X 10.4 support in Gecko 1.9.3

Robert Kaiser
In reply to this post by James Rohde
JimR63701 wrote:
> Robert, your own bias appears to be showing... (Not meant as a
> criticism, but the reality is, it's not "your ox that's being gored",
> as you don't seem to have any personal investment on behalf of the Mac
> version)

It might, and I'm neither on a Mac nor on an even more-used OS in terms
of our numbers. Hell, I'm not even a Firefox user, but still a
Gecko-lover and Mozilla enthusiast.
In my project, 1.4 million daily users would be more than our
application has as a whole. But in Firefox world, 1.4 million daily
users on OS X 10.4 are not only less than a quarter of all Mac users,
but also an almost (!) insignificant number compared to the total daily
user number that is somewhere between 200 and 300 million, IIRC.
Of course, it's not completely insignificant or 10.4 would not work with
any current versions. It's even important enough that the just-released
Firefox 3.6 will - for its whole lifetime of probably still at least a
year - support this OS version fully, with add-ons, security updates and
everything. The same will probably be true for Thunderbird 3.1.

All the talk here is about the *next* version of Firefox and other
Mozilla applications, i.e. Firefox 3.7 (or higher), Thunderbird 3.2 (or
higher) and probably SeaMonkey 2.1 as well, all of which will not be
released before summer or fall this year, maybe even later. Note that
all future version numbers are tentative and can be changed at any time
(if so, usually to a higher number).

Robert Kaiser
_______________________________________________
dev-planning mailing list
[hidden email]
https://lists.mozilla.org/listinfo/dev-planning
Reply | Threaded
Open this post in threaded view
|

Re: Dropping Mac OS X 10.4 support in Gecko 1.9.3

Mulder-7
In reply to this post by Mulder-7
On Feb 6, 2:18 pm, Shawn Wilsher <[hidden email]> wrote:
> On 2/6/2010 6:44 AM, Mulder wrote:

>> If you currently have limited resources (programmers), try using some
> > of the $25 million or more you collect every year from Google to hire
> > more Mac OS X programmers. Not only would that add to the expertise
> > you have, it would make your development cycle move faster.
>
> Have you read The Mythical Man Month?  Adding more people will not make
> us move faster in the short term, and it may not in the long term either.
>
Nonsense. On the one hand you're arguing that you don't have the
resources, and on the other you're saying that if you did have those
resources it wouldn't help you. I'm sorry, but you can't have it both
ways.

>
> Besides that, you are arguing that we should spend a large sum of money
> on small percentage of our users (a shrinking percentage at that).  What
> would you say to all of our windows users who would then see less
> resources devoted to them?
>
>> You might not currently have the resources you need to support all
> > three versions, but you have more than enough money to get those
> > resources; all it takes is the courage to do it and the willingness to
> > support your users. Alienating users by dropping support when there's
> > no valid technological reason only makes more users unhappy and causes
> > them to switch again to another browser, which is a never-ending
> > cycle.
>
> I'm sorry, but Josh covered the technical reasons in his first post.
> You can certainly try to argue that they aren't issues, but simply
> dismissing them is not the way to do that.
>
I never said they aren't issues, nor did I dismiss them. I did say
that none of us know what Apple will do until they actually do it, so
the stated problem of compatibility with the Java2 plugin has yet to
materialize, if it ever does.

>
> > I can't see the future, so I have no idea what the usage numbers will
> > be, just as you have no idea. I'm sure you're well aware that Apple
> > doesn't comment on future plans and I certainly can't make them talk.
> > I do know that there are millions of Mac OS X 10.4 users who either
> > cannot or will not upgrade, for a variety of reasons, all of them
> > perfectly valid for their needs. I also know that Mac OS X 10.5 and
> > 10.6 are riddled with serious bugs and had features removed without
> > notice; something Apple doesn't publicize or disclose to anyone before
> > or after they purchase the OS or a system with those versions
> > installed. Users are left to discover it on their own, and it's always
> > when they need it to work correctly; that's too late.
>
> What are these serious bugs and removed features you speak of?  If you
> are trying to convince us that this is a serious issue, I suggest you
> make actual claims instead of hand-wavy references to problems that come
> across as FUD.
>
There's no FUD or hand-waving in the fact that Address Book doesn't
import all records or notes in 10.5 and 10.6; exporting Mail messages
in certain formats doesn't work; .gif animation was removed in Safari
as of 10.5; Safari's Top Sites feature is a resource hog and
accumulates gigabytes of files without notice; and iCal To-Do items
become corrupted at the drop of a hat, among many others.
>
> > So the question you need to ask yourselves before pursuing this ill-
> > considered idea is: "Do you want to spend the money to hire the
> > programmers you need to support 10.4, or do you want to give up
> > millions in revenue?"
>
> It's not even about hiring programmers.  It's also about buying more
> hardware, and supporting it.
>
You need hardware for programmers to use, but it doesn't cost $25
million.
>
> It's also about using our resources in the
> most beneficial way for our users.  You are arguing that we should spend
> a disproportionate amount of our resources on a small and shrinking
> (even if the number of 10.4 users stays the same, they will make up a
> smaller percentage of our users over time) percentage of our user base.
>
I'm not arguing that you should spend a disproportionate amount of
resources on supporting 10.4 at all. I'm arguing that you already have
the hardware to support Mac OS X 10.4, so that's what you should be
doing. In the meantime, you're trying to convince people that you're
suddenly going to lose those resources after 3.7 ships. Unless you're
planning to suddenly disgorge yourselves of that hardware, you're not
losing it, you just don't want to use it to support 10.4 users. You're
probably spending more time and resources on supporting a crappy OS
(Windoze), because Microsoft doesn't know how to make a decent OS, or
a browser that's secure and standards-compliant. You could just as
easily drop support for Firefox for Windows and see if that forces all
those users to get a new Mac. I'm betting it won't.
>
The notion that your 10.4 users are a small percentage of Firefox
users may be true, but at this point you can only infer that user base
will shrink, since we cannot see the future or forecast it with any
accuracy. The reason that the current adoption percentage for 3.6 on
Mac OS X 10.4 is so low may be as simple as the fact that those users
aren't early adopters, having learned their lesson years ago. Just
because people don't stampede to their browser to download the newest
version when it's available doesn't mean they won't update to that
version after they see what shakes out as far as bugs or extension
incompatibilities (which are often the reason for not upgrading right
away).
>
> We are not a massive corporation like Apple, so we have to use our
> limited resources in the best possible way for our users.
>
You don't have billions in cash like Apple, but you do have a lot more
money than indie software developers, so you should make your product
work on as many versions of Mac OS X as possible, especially if Apple
is still supporting those versions with security updates or browser
updates, as they are with 10.4.
>
> Cheers,
>
> Shawn
_______________________________________________
dev-planning mailing list
[hidden email]
https://lists.mozilla.org/listinfo/dev-planning
Reply | Threaded
Open this post in threaded view
|

Re: Dropping Mac OS X 10.4 support in Gecko 1.9.3

Boris Zbarsky
On 2/7/10 10:37 AM, Mulder wrote:
> Nonsense. On the one hand you're arguing that you don't have the
> resources, and on the other you're saying that if you did have those
> resources it wouldn't help you.

Wouldn't help in the timeframe involved here.

That is, if we hired more people _today_ they would not be up to speed
in time to help.

In the long term, the main issue is that amount of work that can be done
scales sublinearly with number of people doing it.  So in fact putting
more people on an issue requires "disproportionate" investment to get an
improvement, after a certain point.

> I'm not arguing that you should spend a disproportionate amount of
> resources on supporting 10.4 at all. I'm arguing that you already have
> the hardware to support Mac OS X 10.4

So you're arguing that support for 64-bit builds on 10.6, say, is less
important than support for 10.4?  The point is that the resources
currently being used for 10.4 would be repurposed for better supporting
10.5 and 10.6.  In particular, the hardware used for 10.4 and the
hardware needed for 64-bit 10.6 builds are about equivalent in terms of
number of machines and such.

> In the meantime, you're trying to convince people that you're
> suddenly going to lose those resources after 3.7 ships.

We are if we want to use them to support 10.6 64-bit builds.

> you just don't want to use it to support 10.4 users

Right, because we feel that there is a bigger user demographic that
would be better served by it.

Note that the hardware is not a huge issue, by the way; the fact that
you have to structure the code very differently to work on 10.4 and 10.6
(_especially_ 10.6 64-bit) is a much bigger problem.

> You could just as
> easily drop support for Firefox for Windows and see if that forces all
> those users to get a new Mac. I'm betting it won't.

Of course it won't.  I don't see what that has to do with the discussion
here, though.

> The notion that your 10.4 users are a small percentage of Firefox
> users may be true, but at this point you can only infer that user base
> will shrink

That seems like a good bet, since the actual number of 10.4 users is
likely to be shrinking.  It's not like people are buying many new 10.4
installs out there.

> You don't have billions in cash like Apple, but you do have a lot more
> money than indie software developers, so you should make your product
> work on as many versions of Mac OS X as possible

That's a non-sequitur, sorry.  We should make our product work as well
as it can for as many of our users as it can.  The question is what to
do when those two goals are in conflict, as here.  We can significantly
improve the user experience on 10.5 and especially 10.6 if we drop
support for 10.4 (we're talking something like 30% performance
improvement on 10.6, for example if I recall the numbers correctly,
between the newer compiler and doing 64-bit builds).  Which of those is
more important?  It really depends on one's point of view, obviously.

> especially if Apple
> is still supporting those versions with security updates or browser
> updates, as they are with 10.4.

No one's talking about dropping 10.4 support right now.  The discussion
is about dropping support for it in late spring 2011 or so at the earliest.

Sadly, that involves guessing what Apple will do.  Guessing wrong one
way means we drop support for users whose OS is still supported.
Guessing wrong the other way means we invest a lot of time into users
whose OS is no longer supported and shortchange other users.

Welcome to life.  It's full of hard choices.  What Josh is asking for
here is any information anyone might have that has not been brought up
yet (of which I have seen none in this thread so far) that will affect
the decision that has been made thus far based on the information that
has already been raised before.

-Boris
_______________________________________________
dev-planning mailing list
[hidden email]
https://lists.mozilla.org/listinfo/dev-planning
Reply | Threaded
Open this post in threaded view
|

Re: Dropping Mac OS X 10.4 support in Gecko 1.9.3

Boris Zbarsky
In reply to this post by Mulder-7
On 2/7/10 10:37 AM, Mulder wrote:
> You're probably spending more time and resources on supporting a crappy OS
> (Windoze), because Microsoft doesn't know how to make a decent OS

And just for the record, we're putting in a lot more work into our Mac
support than Windows support, on a per-user basis.  If we hired 2 more
people to do Mac stuff, we'd be putting in more work in absolute terms too.

The fact is, the changes between Mac OS versions are _huge_; supporting
multiple versions of Mac OS at once is a huge pain (supporting 10.4 and
10.6 at once is about like supporting Win98 and Win7 at once, if not
harder).

-Boris

_______________________________________________
dev-planning mailing list
[hidden email]
https://lists.mozilla.org/listinfo/dev-planning
Reply | Threaded
Open this post in threaded view
|

Re: Dropping Mac OS X 10.4 support in Gecko 1.9.3

Daniel Veditz-2
In reply to this post by PhillipJones
On 2/6/10 11:16 AM, Phillip Jones wrote:
> In in the Past I tested nightly until I figured what I said about
> bugs or suggestion were laughed at or ignored. since users don't
> know anything.

I'm sorry you had that experience -- no one should be laughing at
anyone trying to give good feedback.
_______________________________________________
dev-planning mailing list
[hidden email]
https://lists.mozilla.org/listinfo/dev-planning
Reply | Threaded
Open this post in threaded view
|

Re: Dropping Mac OS X 10.4 support in Gecko 1.9.3

Mulder-7
In reply to this post by Boris Zbarsky
On Feb 7, 10:25 am, Boris Zbarsky <[hidden email]> wrote:

> On 2/7/10 10:37 AM, Mulder wrote:
>
> > Nonsense. On the one hand you're arguing that you don't have the
> > resources, and on the other you're saying that if you did have those
> > resources it wouldn't help you.
>
> Wouldn't help in the timeframe involved here.
>
> That is, if we hired more people _today_ they would not be up to speed
> in time to help.

I find that hard to believe. I know many Mac programmers with years of
experience, having worked for Adobe, Apple, etc on major applications,
include Adobe Illustrator and iMovie. They're out there and available
for the right price.

>
> In the long term, the main issue is that amount of work that can be done
> scales sublinearly with number of people doing it.  So in fact putting
> more people on an issue requires "disproportionate" investment to get an
> improvement, after a certain point.
>
> > I'm not arguing that you should spend a disproportionate amount of
> > resources on supporting 10.4 at all. I'm arguing that you already have
> > the hardware to support Mac OS X 10.4
>
> So you're arguing that support for 64-bit builds on 10.6, say, is less
> important than support for 10.4?

Yes, they are less important. 64-bit is strictly hype, there is no
benefit to the average user.

> The point is that the resources
> currently being used for 10.4 would be repurposed for better supporting
> 10.5 and 10.6.  In particular, the hardware used for 10.4 and the
> hardware needed for 64-bit 10.6 builds are about equivalent in terms of
> number of machines and such.
>
> > In the meantime, you're trying to convince people that you're
> > suddenly going to lose those resources after 3.7 ships.
>
> We are if we want to use them to support 10.6 64-bit builds.

You don't need to support 64-bit builds, because users gain nothing by
using 64-bit applications. The only exception is a very narrow subset
of all users that use resource-intensive applications such as those
for motion graphics, RAW image processing, and even film editing, that
are designed to get a performance boost from it.

>
> > you just don't want to use it to support 10.4 users
>
> Right, because we feel that there is a bigger user demographic that
> would be better served by it.
>
> Note that the hardware is not a huge issue, by the way; the fact that
> you have to structure the code very differently to work on 10.4 and 10.6
> (_especially_ 10.6 64-bit) is a much bigger problem.
>
> > You could just as
> > easily drop support for Firefox for Windows and see if that forces all
> > those users to get a new Mac. I'm betting it won't.
>
> Of course it won't.  I don't see what that has to do with the discussion
> here, though.

It has everything to do with it, because what Mozilla plans to do is
based on the assumption that it will spur 10.4 users to switch to
10.6, which many or most of them cannot do for financial reasons, or
choose not to do because of the myriad of bugs in 10.5 and 10.6 that
have yet to be resolved. So if you think that tactic will work with
Mac users, there's no reason you shouldn't do the inverse and impose
it on Windoze users and see if they buy a Mac instead.

>
> > The notion that your 10.4 users are a small percentage of Firefox
> > users may be true, but at this point you can only infer that user base
> > will shrink
>
> That seems like a good bet, since the actual number of 10.4 users is
> likely to be shrinking.  It's not like people are buying many new 10.4
> installs out there.
>
> > You don't have billions in cash like Apple, but you do have a lot more
> > money than indie software developers, so you should make your product
> > work on as many versions of Mac OS X as possible
>
> That's a non-sequitur, sorry.  We should make our product work as well
> as it can for as many of our users as it can.  The question is what to
> do when those two goals are in conflict, as here.  We can significantly
> improve the user experience on 10.5 and especially 10.6 if we drop
> support for 10.4 (we're talking something like 30% performance
> improvement on 10.6, for example if I recall the numbers correctly,
> between the newer compiler and doing 64-bit builds).  Which of those is
> more important?  It really depends on one's point of view, obviously.
>
> > especially if Apple
> > is still supporting those versions with security updates or browser
> > updates, as they are with 10.4.
>
> No one's talking about dropping 10.4 support right now.  The discussion
> is about dropping support for it in late spring 2011 or so at the earliest.
>
> Sadly, that involves guessing what Apple will do.  Guessing wrong one
> way means we drop support for users whose OS is still supported.
> Guessing wrong the other way means we invest a lot of time into users
> whose OS is no longer supported and shortchange other users.
>
> Welcome to life.  It's full of hard choices.  What Josh is asking for
> here is any information anyone might have that has not been brought up
> yet (of which I have seen none in this thread so far) that will affect
> the decision that has been made thus far based on the information that
> has already been raised before.

Then you're acknowledging publicly that it doesn't matter what users
say, you're going to do something that you already decided to do
before asking for input. Choices may be hard, but when you ask for
user input, you should at least listen to the wishes of the users, not
your own predetermined decision. Welcome to the ranks of hypocrisy.
>
> -Boris

_______________________________________________
dev-planning mailing list
[hidden email]
https://lists.mozilla.org/listinfo/dev-planning
Reply | Threaded
Open this post in threaded view
|

Re: Dropping Mac OS X 10.4 support in Gecko 1.9.3

Brad Lassey-2

> Then you're acknowledging publicly that it doesn't matter what users
> say, you're going to do something that you already decided to do
> before asking for input. Choices may be hard, but when you ask for
> user input, you should at least listen to the wishes of the users, not
> your own predetermined decision. Welcome to the ranks of hypocrisy.
>    
There is a difference between having your wishes and opinions listened
to and having them change the proposed decision, please don't confuse
the two.

The fact of the matter is that the people who originally posted asking
for feedback have already researched the pros and cons of this decision
to the best of their ability. It is the realm of possibility that they
missed some crucial fact, but thus far such an oversight has not been
pointed out (as far as I can tell anyway).

-Brad
_______________________________________________
dev-planning mailing list
[hidden email]
https://lists.mozilla.org/listinfo/dev-planning
Reply | Threaded
Open this post in threaded view
|

Re: Dropping Mac OS X 10.4 support in Gecko 1.9.3

John J Barton
In reply to this post by Daniel Veditz-2
On 2/7/2010 9:20 AM, Daniel Veditz wrote:
> On 2/6/10 11:16 AM, Phillip Jones wrote:
>> In in the Past I tested nightly until I figured what I said about
>> bugs or suggestion were laughed at or ignored. since users don't
>> know anything.
>
> I'm sorry you had that experience -- no one should be laughing at
> anyone trying to give good feedback.

But it is common nevertheless. And 'ignored' is even more common.

jjb
_______________________________________________
dev-planning mailing list
[hidden email]
https://lists.mozilla.org/listinfo/dev-planning
Reply | Threaded
Open this post in threaded view
|

Re: Dropping Mac OS X 10.4 support in Gecko 1.9.3

Zack Weinberg-6
johnjbarton <[hidden email]> wrote:

> On 2/7/2010 9:20 AM, Daniel Veditz wrote:
> > On 2/6/10 11:16 AM, Phillip Jones wrote:
> >> In in the Past I tested nightly until I figured what I said about
> >> bugs or suggestion were laughed at or ignored. since users don't
> >> know anything.
> >
> > I'm sorry you had that experience -- no one should be laughing at
> > anyone trying to give good feedback.
>
> But it is common nevertheless. And 'ignored' is even more common.

Yup.  And it only takes one bad bugzilla (or support forum) experience
to put a user off contacting us *forever*, & they will tell all their
friends not to bother trying to give us feedback, too.

IMO a good start toward fixing this would be to change the Bugzilla
etiquette guidelines so that under no circumstances short of actual
spam is it acceptable to tell anyone to stop commenting on a bug.

zw
_______________________________________________
dev-planning mailing list
[hidden email]
https://lists.mozilla.org/listinfo/dev-planning
1234 ... 10