Proposal: Raise minimum requirements for 1.9.2 on Windows to WinXP SP3

classic Classic list List threaded Threaded
115 messages Options
1 ... 3456
Reply | Threaded
Open this post in threaded view
|

Re: Proposal: Raise minimum requirements for 1.9.2 on Windows to WinXP SP3

Robert Kaiser
Mike Connor wrote:
> Again, I think this is developer-centric, when the burden of "support"
> comes in a bunch of situations. Are we okay with not having unit
> tests/performance tests/etc on operating systems we count as supported?

We technically still support Mac OS 10.4 on 1.9.2 AFAIK, and we don't
even have tests running on 10.4 for Shiretoko, and from all I heard, the
(compatibility) differences between Tiger and Leopard tend to be higher
than between Win2k and XP SP2+.

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

Re: Proposal: Raise minimum requirements for 1.9.2 on Windows to WinXP SP3

Robert Kaiser
In reply to this post by Aakash Desai-2
Mike Connor wrote:
> What's the case for continuing to invest in Windows 2000 and pre-SP2
> versions of Windows XP?

I still haven't seen the compelling arguments of how much of an
investment supporting Win2k is. What functionality that we want to be
using is missing in Win2k? How much work do we usually spend on ensuring
Win2k works? Or is it a political matter in terms of what we feel our
users should be using? In that case, would that counter the part of the
Mozilla mission that is about choice and opportunity more or less than
it would foster the openness, innovation and personal security parts?
I think those are the kinds of questions we actually need to think about
here.

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

Re: Proposal: Raise minimum requirements for 1.9.2 on Windows to WinXP SP3

emailaddress
In reply to this post by Boris Zbarsky
On Apr 16, 11:04 am, Boris Zbarsky <[hidden email]> wrote:
> [hidden email] wrote:
> > With zero patches, the most vulnerability to an unpatched Win2k, not even SP1,
> > is the browser.
>
> You were sounding fairly reasonable until right here.  If you connect an
> unpatched Win2k system to the internet with no firewall, it'll be owned
> in a matter of minutes with no browser involved.
>

Who do you know that connects their unpatched Win2k system directly to
the internet?
Risk is about use, who would even want to exploit a system that must
be using a dial-up connection these days if it's not behind some kind
of NAT if not a proper firewall?  I suppose credit-card list
generators would still make the effort, but it's even less likely
someone with basic sophistication level that they are paying bills
online, would be using an unsecured dial-up internet access method.
Any other method by default insulates them.


> > It is not relevant how many years ago it was, only what the goals of
> > the browser are and where the time is best spent.
>
> The "years ago" thing is relevant in terms of the browser goals, and in
> particular as a rough proxy for user numbers.  If we can get better user
> numbers, of course, that's even better.  But it's hard to do better for
> user numbers a year or year and a half from now, really.  Do you have a
> time machine?

No time machine needed, you simply need to keep growth higher than
losses today and in the future.  That definitely involves keeping the
browser working on as many OS variants as possible.  I accept that
beyond businesses, fewer individuals would run Win2k, but pre-SP3 XP
is on quite a few systems, perhaps more than some people realize as
many people have a mindset that if their computer seems to be working
fine that they do not want the potential complications of installing a
service pack.  Remember, most of society has no idea what Mozilla is,
probably it's the 3rd cousin to Godzilla, but if someone says Firefox
is better they can handle clicking to install it.


>
> > The point is, we are reaching the point in history where there should
> > not be a need to perpetually change to a newer OS to do the same
> > tasks.
>
> While true, we also want to do new tasks
>

Some people do, some don't.  Since this topic is about a browser I'd
have to assume browsing is the task.


> > To take the age old phrase, "if it ain't broke don't fix it".
>
> All OSes are broke, pretty much.  We've had to work around bugs in every
> singe OS we support.  As we add new code for new low-level features
> (including silly things like process separation, accelerated graphics,
> out-of-process plug-ins), we will run into more and more things we need
> to work around, especially on the older OSes in the Windows case
> (because the newer ones tend to have more APIs that do the same thing,
> so more chance that one of them will happen to work for us).

Agreed, but you are talking about philosophical breakage and I am
talking about "does it get the job done for the user" breakage.  To
that end, a user will keep using what gets the job done and be
resistant to change for reasons of time and expense.  Similarly,
wouldn't it be great to drive a different car every week, but for most
people the time and cost is impractical.

I am not so much suggesting that official support for Win2k and pre-
SP3 XP should not be dropped, rather I am stating that we need to keep
clarity about the reasons and about the potential effect, weighing all
the goals moving forward.

>
> If those features (and whatever else is being worked on) happen to work
> on Win2k, then there won't be a problem: Firefox will just run on it.
> If not... which of the above three features would you propose dropping
> if we can't make it work on Win2k?
>

It is the burden of the OS to be backwards compatible if it claims to
be.  The target OS should be the lowest common denominator, not new
APIs.  Since you asked, I'd drop accelerated graphics, but personally
I feel the greatest hurdle to Firefox is not lack of features, it's
making it as cross platform capable as possible.  "Designed for the
latest Windows version" is an idea that tomorrow's world won't like
much, if it has to rely on MS APIs, make it as few as possible.

> > There is little sanity in changing an OS if it does what is required
>
> Fully agreed, but if it can't do what Firefox wants it to do and you
> want to run Firefox, then it's in fact not doing what is required.  The
> job of an OS is to let you run applications.

Indeed, I've written the same thing myself about OS being a means to
run apps, but is Firefox really having a mind of it's own what it
should want to do, or is it the user who chooses what is running on
their machine, that most important is whether it runs in the available
environment, not features that seem good but really aren't as
important to the average user as to those inspired to make
improvements for the sake of change?

Certainly there are some very sophisticated browser enthusiasts with
some great ideas, but they are the minority, a minority I suspect is
far lower than the % of people out there not wanting to depend on the
newer APIs because their OS isn't the latest and greatest, and looking
forward I think people will be changing their OS even less often than
in the past due to economic as well as decreasing usability
improvement reasons.

> Generally speaking, no.  Heck, _I_ won't claim to know whether my
> machine is as secure as it could be or should be, and I work on this
> stuff for a living and for fun.

That's a good point, but with security in mind people are likely to be
more secure with a more modern browser than a more modern OS.


> > Many people who aren't
> > particularly techinically inclinded, which may be the majority of
> > society, use the originally shipping OS on their PC and only end up
> > changing when hardware failure forces them to buy a new PC.
>
> Very true, but that doesn't square with the claim that they know
> anything about how secure their setup is.

They will, it seems, decide risk based upon personal experience.  If
they're getting infected with malware once every 4 years they are not
going to think, and arguably rightly so, that they are at high risk.
Security is ultimately about risk since nothing is ever truely
secure.  Could we even argue that new APIs are less secure than mature
old ones?

>
> > Through technology mankind has a hope, and a green one at that, that our
> > progress allows a system to run for it's useful lifespan which with
> > today's performance levels (even those of 5 years ago) comes close to
> > predating Win2k for the most common tasks people do on a pc.
>
> Note that hardware lifetimes for typical consumer hardware are < 5 years
> nowadays, especially for laptops.  

I'd roughly estimate that outside of tech sectors, in real life I know
as many people with desktops >= 5 years old as those < 5 years.
Certainly for laptops the age is lower, but won't people tend to want
to use the same browser on their laptop that they use on their
desktop?  Further given the performance increases in recent years vs.
the most common applications people run (like browsers), plus a
downturn in the economy and a rise in availability of information
necessary to troubleshoot PC problems,  it has become more and more
cost effective to repair a system rather than replace it.

> This brings us back to the age
> question, because if the last hardware that came with win2k was bought 5
> years ago and people upgrade OS when they get new hardware (not all, but
> most, as you say), what will the user base of Win2k be in Q3 2010?

The user base will certainly be small, smaller than it is today, but
the topic started out including newer OS such as XP SP2.  In 2010 it
would not surprise me if > 20% of PCs worldwide were not at XP SP3 OS
level or higher.  Online polls and surveys might suggest less than
this, but online polls and surveys tend to reach the enthusiasts more
than the average PC owner.    Although my home phone is on a do not
call list, I wouldn't expect someone to call and ask what operating
systems my computers are running.

> In any case, the point is that no one is proposing breaking things on
> win2k for the sake of breaking them.  The proposal is just to not spend
> core QA and developer time fixing things if they do break due to new
> feature work.  Anyone who cares to do said fixing is welcome to do it
> and contribute patches: the source is open.

I agree that the time is precious, but do feel maximizing OS support
is more important than feature growth.  Design for lowest common
denominator and let MS take care of their own bugs if Windows 2012 SP2
won't do what it's supposed to.  I admit I have not kept up with
browser development discussions but what I read more than anything is
that the features are bogging down the browser, and that people don't
want to be forced into a corner about which OS they use.  Maybe it was
good that MS had a large market share to unite the industry,
accelerating growth, but code bloat will be their undoing in our
mobile future.
_______________________________________________
dev-planning mailing list
[hidden email]
https://lists.mozilla.org/listinfo/dev-planning
Reply | Threaded
Open this post in threaded view
|

Re: Proposal: Raise minimum requirements for 1.9.2 on Windows to WinXP SP3

emailaddress
In reply to this post by Karl Tomlinson-4
On Apr 16, 8:36 pm, Karl Tomlinson <[hidden email]> wrote:

> > If you connect an unpatched Win2k system to the internet with no
> > firewall, it'll be owned in a matter of minutes with no browser
> > involved.
>
> It's not clear to me how good the "these computers are owned
> anyway" argument is.
>
> AFAIK the "minutes" figures are for a computer with an IP address
> on the internet but no firewall.
>
> Old operating systems should be behind a firewall, and I suspect
> many of their users are in this situation, even if it is just the
> network address translation provided by their modem.  This doesn't
> eliminate the security risk; it only reduces the exposure;
> the low-level network stack is still exposed.
>
> Is it still a reasonable assumption that unsupported OS's are
> owned?

One Win2k system I've had running since brand new, maybe 8 years ago
behind only a NAT, modem plus router which is a pretty typical network
topology.  It has never been exploited AFAIK, meaning in several years
of constant uptime no evidence of infection, no new processes, nothing
any scanner has ever picked up.  However, it is not used to surf the
'net, which I feel is the highest risk activity any Windows box can
encounter.  In past years I've owned, as well as encountered quite a
few win2k systems only behind a NAT with no issues beyond those caused
by the user running an app like browser or email client, warez, etc.
Logs show plenty of port scanning but no exploits.

Part of this may be due to the fact that like some browser developers,
malware authors also make presumption about the future viability of
their code and are not targeting what was patched years ago and is on
a smaller % of systems.

A moment ago I became curious as to what % of visitors to a 'site I
maintain were still running IE6.  23%, but how many new exploits are
targeting only IE6?  It's likely that capturing some of this market
will leave people safer.
_______________________________________________
dev-planning mailing list
[hidden email]
https://lists.mozilla.org/listinfo/dev-planning
Reply | Threaded
Open this post in threaded view
|

Re: Proposal: Raise minimum requirements for 1.9.2 on Windows to WinXP SP3

emailaddress
In reply to this post by Mike Connor-4
On Apr 13, 10:40 pm, Robert Accettura <[hidden email]> wrote:
> I know Win2k is somewhat limited at this point but is there any data
> regarding what percentage of Windows XP users run something less than
> SP3?  I presume most impact will be corporations that are slow to roll
> out upgrades as SP3 has been out for a while now.


I've heard it argued that if people were running an alternate browser
and email client, that even SP2 wasn't very necessary if they had good
security practices in place besides the OS SP level.
_______________________________________________
dev-planning mailing list
[hidden email]
https://lists.mozilla.org/listinfo/dev-planning
Reply | Threaded
Open this post in threaded view
|

Re: Proposal: Raise minimum requirements for 1.9.2 on Windows to WinXP SP3

Robert Blair-2
In reply to this post by emailaddress
On 4/16/2009 11:18 PM, [hidden email] wrote:

> On Apr 16, 8:36 pm, Karl Tomlinson <[hidden email]> wrote:
>>> If you connect an unpatched Win2k system to the internet with no
>>> firewall, it'll be owned in a matter of minutes with no browser
>>> involved.
>> It's not clear to me how good the "these computers are owned
>> anyway" argument is.
>>
>> AFAIK the "minutes" figures are for a computer with an IP address
>> on the internet but no firewall.
>>
>> Old operating systems should be behind a firewall, and I suspect
>> many of their users are in this situation, even if it is just the
>> network address translation provided by their modem.  This doesn't
>> eliminate the security risk; it only reduces the exposure;
>> the low-level network stack is still exposed.
>>
>> Is it still a reasonable assumption that unsupported OS's are
>> owned?
>
> One Win2k system I've had running since brand new, maybe 8 years ago
> behind only a NAT, modem plus router which is a pretty typical network
> topology.  It has never been exploited AFAIK, meaning in several years
> of constant uptime no evidence of infection, no new processes, nothing
> any scanner has ever picked up.  However, it is not used to surf the
> 'net, which I feel is the highest risk activity any Windows box can
> encounter.  In past years I've owned, as well as encountered quite a
> few win2k systems only behind a NAT with no issues beyond those caused
> by the user running an app like browser or email client, warez, etc.
> Logs show plenty of port scanning but no exploits.
>
> Part of this may be due to the fact that like some browser developers,
> malware authors also make presumption about the future viability of
> their code and are not targeting what was patched years ago and is on
> a smaller % of systems.
>
> A moment ago I became curious as to what % of visitors to a 'site I
> maintain were still running IE6.  23%, but how many new exploits are
> targeting only IE6?  It's likely that capturing some of this market
> will leave people safer.

In a two-week collection of hits to 18 of my Web pages last month, 39%
of the IE hits were from IE 6.  2% were from IE 5.  Divide those in half
to get the proportion of hits to total recognizable browser hits since
all versions of IE accounted for only half.

Also, 12% of the Gecko hits involved versions 1.8.x.  Since Gecko
accounted for about 40% of the recognizable browsers, that's about 5% of
the total.

--
David E. Ross
<http://www.rossde.com/>

Go to Mozdev at <http://www.mozdev.org/> for quick access to
extensions for Firefox, Thunderbird, SeaMonkey, and other
Mozilla-related applications.  You can access Mozdev much
more quickly than you can Mozilla Add-Ons.
_______________________________________________
dev-planning mailing list
[hidden email]
https://lists.mozilla.org/listinfo/dev-planning
Reply | Threaded
Open this post in threaded view
|

Re: Proposal: Raise minimum requirements for 1.9.2 on Windows to WinXP SP3

Serge Gautherie-2
In reply to this post by Robert Kaiser
Robert Kaiser wrote:

> We technically still support Mac OS 10.4 on 1.9.2 AFAIK, and we don't
> even have tests running on 10.4 for Shiretoko, and from all I heard, the
> (compatibility) differences between Tiger and Leopard tend to be higher
> than between Win2k and XP SP2+.

(SeaMonkey 2.0 (Gecko 1.9.1) does have a MacOSX 10.4 box ;-))

Afaik, there is no W2K boxes (of any kind), especially unit tests.
*Neil has a "WNT 3.51" SeaMonkey 1.1 (Gecko 1.8.1) build box.
*KaiRo has a "WXP 5.1" SeaMonkey 1.1 (Gecko 1.8.1) build box.
*All tests since Gecko 1.9.0 are on "W2K3 5.2".
  *(And none on WXP nor WVista for that matter.)
Then making W2K a tier-2 is (just) making this official :-|

NB: I might be willing to care for a W2K box (which could probaly build
multiple applications/versions), if there was (even a slow VM) one...

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

Re: Proposal: Raise minimum requirements for 1.9.2 on Windows to WinXP SP3

Mike Shaver
On Fri, Apr 17, 2009 at 8:57 AM, Serge Gautherie <[hidden email]> wrote:
> NB: I might be willing to care for a W2K box (which could probaly build
> multiple applications/versions), if there was (even a slow VM) one...

If you want to be a W2K steward, that'd be great -- please file a bug
to get a VM allocated for that purpose, and cc: me!

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

Re: Proposal: Raise minimum requirements for 1.9.2 on Windows to WinXP SP3

Boris Zbarsky
In reply to this post by emailaddress
[hidden email] wrote:
> Who do you know that connects their unpatched Win2k system directly to
> the internet?

No one, but I don't know anyone who's running Win2k at all.

>>> It is not relevant how many years ago it was, only what the goals of
>>> the browser are and where the time is best spent.
>> The "years ago" thing is relevant in terms of the browser goals, and in
>> particular as a rough proxy for user numbers.  If we can get better user
>> numbers, of course, that's even better.  But it's hard to do better for
>> user numbers a year or year and a half from now, really.  Do you have a
>> time machine?
>
> No time machine needed, you simply need to keep growth higher than
> losses today and in the future.

You misunderstood the question.  The question is: "Who, if anyone, will
be using Win2k 18 months from now, and how many of them will there be?"
  It's not a question about browser market share but OS marketshar.

> That definitely involves keeping the
> browser working on as many OS variants as possible.

That need not be true.  If dropping OS A allows more effort to be put
into working better on OS B, and OS B has 90% of the market while OS A
has 0.01%, then loss of all users of OS B might easily be offset by a
slight gain in market share amogst users of OS A.

That's assuming the goal is only maximization of market share, which
it's not.

> I accept that beyond businesses, fewer individuals would run Win2k, but pre-SP3 XP
> is on quite a few systems

And pre-SP2? I think it's become pretty clear that SP2 is going to stay
supported for the time being.

>> While true, we also want to do new tasks
>
> Some people do, some don't.  Since this topic is about a browser I'd
> have to assume browsing is the task.

Yes, but that's a moving target.  Browsing used to be just text, then
text+images, then text+images+plug-ins+javascript, then
text+images+plug-ins+javascript+css+dom, and now I welcome you to take a
long hard look at the set of existing and proposed specifications for
what a "browser" needs to be able to do.

> Agreed, but you are talking about philosophical breakage and I am
> talking about "does it get the job done for the user" breakage.

We're talking about the same breakage.  If the browser ends up in an
infinite loop due to an OS bug (most recent one I ran into), that's not
getting the job done for the user.

> To that end, a user will keep using what gets the job done and be
> resistant to change for reasons of time and expense.

Of course.  The problem is that the "job" of browsing keeps changing.

> Similarly, wouldn't it be great to drive a different car every week, but for most
> people the time and cost is impractical.

Yet if the roads got changed every week so you needed different
wheels... you might be kinda stuck with that.  Or keep using the old one
and have a bad ride.

> I am not so much suggesting that official support for Win2k and pre-
> SP3 XP should not be dropped, rather I am stating that we need to keep
> clarity about the reasons and about the potential effect, weighing all
> the goals moving forward.

I think everyone agrees on that, sure.

>> If those features (and whatever else is being worked on) happen to work
>> on Win2k, then there won't be a problem: Firefox will just run on it.
>> If not... which of the above three features would you propose dropping
>> if we can't make it work on Win2k?
>
> It is the burden of the OS to be backwards compatible if it claims to
> be.

That's a non-sequitur to the above.

> The target OS should be the lowest common denominator, not new
> APIs.

If the lowest common denominator couldn't get the job done, I don't
think so.

The lowest common denominator right now is a microkernel that does
nothing at all other than process management; we'd need to write our own
filesystem to work on something like that.  Clearly we're not targeting
that, nor should we be.

> Since you asked, I'd drop accelerated graphics

Then you get people complaining about the browser using 100% CPU when
the competition uses 2%.

> but personally I feel the greatest hurdle to Firefox is not lack of features, it's
> making it as cross platform capable as possible.

Since it's significantly more cross-platform than either of its two main
competitors (Safari and IE), I would dearly love to hear the reasoning
behind your feeling here.  Assuming there's reasoning, not just feeling.

> "Designed for the
> latest Windows version" is an idea that tomorrow's world won't like
> much, if it has to rely on MS APIs, make it as few as possible.

If you rely on as few as possible, you don't interact well with the OS
and make the user's life a living hell.  Do you really not want Firefox
to get helper app information from the OS?  Do you really want it to not
get the location of the temporary directory from the OS?  Do you really
want it to not use the OS-native look?  To make the user manually do
virus-scanning on downloads?

We use pretty few OS APIs as it is, and they're not added lightly,
because we do aim to be cross-platform.  But "as few as possible" is not
a hard target, and doesn't guarantee anything about working on all OSes.

For example, we assume you use an OS that can support Unicode filenames.
  If you don't (Win9x), you lose.  We feel this is acceptable for a
significantly better user experience for non-English-speaking users.

> Indeed, I've written the same thing myself about OS being a means to
> run apps, but is Firefox really having a mind of it's own what it
> should want to do

Firefox is trying to browse the web, quickly and securely.  Period.
This is getting more complicated to do from day to day.

> or is it the user who chooses what is running on
> their machine, that most important is whether it runs in the available
> environment

Of course.  And the user is also welcome to not run the bleeding edge
version of Firefox, right?

> not features that seem good but really aren't as
> important to the average user as to those inspired to make
> improvements for the sake of change?

Every single feature I listed is directly in either the "quickly" or
"securely" column of "browse the web".  Not that I didn't even list any
of the "needed to browse, period" features; those are a lot more
complicated to explain the details of.

> Certainly there are some very sophisticated browser enthusiasts with
> some great ideas, but they are the minority, a minority I suspect is
> far lower than the % of people out there not wanting to depend on the
> newer APIs because their OS isn't the latest and greatest

Most people have no idea what their browser _should_ do, but notice it
immediately when sites break.

 > and looking forward I think people will be changing their OS even
less often than
> in the past due to economic as well as decreasing usability
> improvement reasons.

That has little to do with this discussion other than having a possible
bearing on the projection of Win2k use in 18 months, because we're
talking about the situation now, not "looking forward".

> That's a good point, but with security in mind people are likely to be
> more secure with a more modern browser than a more modern OS.

Assuming their OS is correctly firewalled (or at least NATted), yes.
That's a big assumption, though.

> They will, it seems, decide risk based upon personal experience.  If
> they're getting infected with malware once every 4 years they are not
> going to think, and arguably rightly so, that they are at high risk.

And if they're getting infected every week they probably won't even
notice.  So sorry, this doesn't fly.

>> Note that hardware lifetimes for typical consumer hardware are < 5 years
>> nowadays, especially for laptops.  
>
> I'd roughly estimate that outside of tech sectors, in real life I know
> as many people with desktops >= 5 years old as those < 5 years.

I do too, but every single person I know with a desktop >= 5 years old
bought a high-quality desktop (Apple, Micron, IBM) back then, for
significantly more money than the typical desktop (Compaq, say) was
selling for.  And the people I know have a general tendency to value
quality over price that doesn't seem to be true for the general population.

> Certainly for laptops the age is lower, but won't people tend to want
> to use the same browser on their laptop that they use on their
> desktop?

Yes, but more and more people just don't have a desktop, period.

> The user base will certainly be small, smaller than it is today, but
> the topic started out including newer OS such as XP SP2.

We've moved on from there several days ago.  I'm not sure why we're
still talking about SP2 here.  Let's stick to the current proposal, not
to the initial brainstorming?

> I agree that the time is precious, but do feel maximizing OS support
> is more important than feature growth.

The market disagrees, as far as I can tell.

 > I admit I have not kept up with
> browser development discussions but what I read more than anything is
> that the features are bogging down the browser

And the problem is that websites are using the features, so not
providing them is not an option.

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

Re: Proposal: Raise minimum requirements for 1.9.2 on Windows to WinXP SP3

Serge Gautherie-2
In reply to this post by Serge Gautherie-2
Mike Shaver wrote:

> If you want to be a W2K steward, that'd be great -- please file a bug
> to get a VM allocated for that purpose, and cc: me!

I filed
Bug 488847 - Allocate a Windows 2000 VM, for tier-2 (unit test) monitoring
;-)

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

Re: Proposal: Raise minimum requirements for 1.9.2 on Windows to WinXP SP3

Gervase Markham
In reply to this post by Mike Connor-4
On 15/04/09 16:51, Aakash Desai wrote:
> I'd suggest you actually talk to school principals and school board
> administrators about this situation and I'll bet that they'll come
> right back at you and tell you that there's much more important
> things (i.e. broken ceiling panels, lighting fixtures, 20 year-old
> textbooks, TEACHER'S SALARIES, etc.) that are much higher in the
> priority list than upgrading their non-broken machines over to Win XP
> SP3 (that they still can't afford to pay for).

Thousands of people all over the world are working hard, and making the
fruit of their labours freely available to you, so you don't have to
upgrade to Windows XP. Or pay for almost any software ever again, for
that matter. Use Linux. Ah, but Linux is only free if your time has no
value, right? True. But if you have no money _and_ no time to improve
your school, then you are pretty much stuffed anyway.

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

Re: Proposal: Raise minimum requirements for 1.9.2 on Windows to WinXP SP3

Gervase Markham
In reply to this post by pawel.jewstafjew
On 15/04/09 19:14, [hidden email] wrote:
> My employer is a sizeable semiconductor company (10000+ employees)
> The company sticks to win2K as desktops and 2003 as servers.
> The end-of-life for personal users is not going to influence them in
> any way
> (they have a customised support agreement anyway)

Super. How would your company like to dedicate a person to fixing
Win2K-only issues in Firefox, and keeping it working? After all, the
browser must be a key piece of software for most of your desktop users.

Gerv

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

Re: Proposal: Raise minimum requirements for 1.9.2 on Windows to WinXP SP3

mateuszcedro
In reply to this post by Robert Kaiser
I think if Mozilla will stop supporting windows NT5.x series, more
people will use other browsers.

In Poland, FF had about 41%, IE about 50.5%.
In Poland, about 89% of people had Windows XP. Only about 7% had
Vista.
(stats from feb09)

It isn't good proposition for people which using XP.
_______________________________________________
dev-planning mailing list
[hidden email]
https://lists.mozilla.org/listinfo/dev-planning
Reply | Threaded
Open this post in threaded view
|

Re: Proposal: Raise minimum requirements for 1.9.2 on Windows to WinXP SP3

Michael Lefevre-2
On 18/04/2009 12:25, [hidden email] wrote:
> I think if Mozilla will stop supporting windows NT5.x series, more
> people will use other browsers.

But "5.x series" is not being discussed - only 5.0. The current proposal
is that Firefox after 3.5 will not support Windows 2000, or Windows XP
SP1.  So there is no issue for Windows XP with service pack 2 or 3.

> In Poland, FF had about 41%, IE about 50.5%.
> In Poland, about 89% of people had Windows XP. Only about 7% had
> Vista.
> (stats from feb09)

But what proportion have Windows 2000 or Windows XP without SP 2 or 3?
I guess a lot less...

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

Re: Proposal: Raise minimum requirements for 1.9.2 on Windows to WinXP SP3

dank-3
In reply to this post by LJHeal
On Apr 15, 6:01 am, [hidden email] wrote:
> I need Windows 2000 for a scanner that gets funky with XP.  It would
> take too much time to develop or discover another solution.  Third-
> party security software and firewall appear to make that argument for
> dropping support a pretty transparent one.  One other essential
> application works with XP but not with WINE.

Which app is that?
- Dan
_______________________________________________
dev-planning mailing list
[hidden email]
https://lists.mozilla.org/listinfo/dev-planning
1 ... 3456