Large market share loss since end of may (statcounter)

[Benoit Jacob]

Hi,

According to statcounter, Firefox has been losing market share very
fast since the end of May, whereas it was making gains in the previous
months:

http://gs.statcounter.com/#browser-ww-daily-20120101-20120701

Do you think that this is just noise or meaningful? If meaningful,
what could be possible reasons? Has an increase in crashes been
observed?

Thanks,
Benoit

[Asa Dotzler]

According to Net Applications, Firefox has gained share in the last month.

There's truth and lie in all of these stats packages. We're probably
seeing less usage in some geographies and user types and more in others.

We're also seeing some of our traditional summer dip (look back to the
last 5 or 6 years and you'll see we always dip in the summer.)

We have lost some share for sure over the last year but I don't worry
too much about any one month swings.

As for an increase in crashes, we are seeing a big spike in Flash
crashes thanks to the recent Adobe update that tried to put Flash in a
sandbox. There's a fix on the way for that.

- A

[Matt Brubeck]

Statcounter also showed IE actually gaining share in June, while it had
been falling steeply since last year, and Chrome almost flat while it
had been rising steadily. I suspect that seasonal or random factors
played some part in these fluctuations.

Flash 11.3 (released in the first week of June) did cause high-volume
crashes and other significant problems for a very large number of
Firefox users, which might have contributed too.

[David H. Lipman]

From: "Asa Dotzler" <a...@mozilla.org>

Your myopic idiocy also plays a part. You don't see the trees through the
forest. Why don't you step down and let someone with vision step up to the
plate.


--
Dave
Multi-AV Scanning Tool - http://multi-av.thespykiller.co.uk
http://www.pctipp.ch/downloads/dl/35905.asp

[Benoit Jacob]

Daniel, this email was off-topic, offensive, and unwarranted. This
thread is about a market share loss that started at a precise time at
the end of May. It wasn't intended for generic ranting. The summer
season and Flash crashiness reasons given so far by Asa and Matt seem
like satisfactory answers.

Benoit

2012/7/2 David H. Lipman <DLipman~nospam~@verizon.net>:

> _______________________________________________
> dev-apps-firefox mailing list
> dev-apps...@lists.mozilla.org
> https://lists.mozilla.org/listinfo/dev-apps-firefox

[Gavin Sharp]

On Mon, Jul 2, 2012 at 9:26 PM, David H. Lipman
<DLipman~nospam~@verizon.net> wrote:
> (snipped)

Messages like this one will not be tolerated on this mailing list.

Gavin (with my list moderator hat on)

[Omega X]

On 7/3/2012 4:40 AM, Benoit Jacob wrote:
> Daniel, this email was off-topic, offensive, and unwarranted. This
> thread is about a market share loss that started at a precise time at
> the end of May. It wasn't intended for generic ranting. The summer
> season and Flash crashiness reasons given so far by Asa and Matt seem
> like satisfactory answers.
>
> Benoit
>

Despite the harshness of the post, he does have a point. Mozilla has
been very poor in the PR department. Most articles about Firefox these
days are often negative and the comments to those articles are as
equally negative. That does have an impact on marketshare as that same
word of mouth that Firefox has benefited from in the past is now working
against the brand.

[Gen Kanai]

There has been negative PR for some time, but that is starting to
change. That Firefox won Lifehacker's most recent browser comparison is
just one data point but the PR around the most recent release is
certainly much better than releases in the past year or so.

http://lifehacker.com/5917714/browser-speed-tests-chrome-19-firefox-13-internet-explorer-9-and-opera-1164?tag=firefox

It may take a lot of time for users who've left Firefox to return. There
may be many different reasons why they would return. We need to keep our
focus on our own issues and be civil in our own discussions.

Look at the competition as well. The search engine provider in Mountain
View who also distributes a browser has had some pretty bad bugs
recently (such as their browser crashing Mac Book Air computers, and
also breaking the combined search/url bar.)

--
Gen Kanai

[Mart Rootamm]

Agree with Benoit, Asa and Matt.

In addition to crashes with Flash, I would also add triskaidekaphobia :>

-Mart.

2012/7/3 Benoit Jacob <jacob.b...@gmail.com>

> Daniel, this email was off-topic, offensive, and unwarranted. This
> thread is about a market share loss that started at a precise time at
> the end of May. It wasn't intended for generic ranting. The summer
> season and Flash crashiness reasons given so far by Asa and Matt seem
> like satisfactory answers.
>
> Benoit
>

> 2012/7/2 David H. Lipman <DLipman~nospam~@verizon.net>:

[Andrew Joakimsen]

I have come to the conclusion that firefox is a browser designed for
idiots and there will be sacrifices to its integrity due to this.
Therefore I have not installed any new copies and ensured updates are
disabled on any existing installations. I do not wish to use or
propagate the use of a browser that has in its core philosophy the
need to protect the user from themselves.

On Mon, Jul 2, 2012 at 6:24 PM, Benoit Jacob <jacob.b...@gmail.com> wrote:
> Hi,
>
> According to statcounter, Firefox has been losing market share very
> fast since the end of May, whereas it was making gains in the previous
> months:
>
> http://gs.statcounter.com/#browser-ww-daily-20120101-20120701
>
> Do you think that this is just noise or meaningful? If meaningful,
> what could be possible reasons? Has an increase in crashes been
> observed?
>
> Thanks,
> Benoit

[Robert Kaiser]

Mart Rootamm schrieb:

> Agree with Benoit, Asa and Matt.
>
> In addition to crashes with Flash, I would also add triskaidekaphobia :>

I think metathesiophobia is more likely to influence a larger number of
people, given what I hear on several channels. A share of people fear
that our rapid releases change things all the time (not that those
people have a lot of other choices - except one from a Redmond, WA,
company that seems to have incredibly long support cycles).

Robert kaiser

[Gian-Carlo Pascutto]

On 5/07/2012 15:05, Robert Kaiser wrote:

> A share of people fear

> that our rapid releases change things all the time.

I can't help but wonder why you used the word "fear" in the above
sentence. Do you think it's unwarranted or incorrect?

Jono just said something about that subject, too:
http://evilbrainjono.net/blog#1094

--
GCP

[Robert Kaiser]

Gian-Carlo Pascutto schrieb:

> On 5/07/2012 15:05, Robert Kaiser wrote:
>
>> A share of people fear
>> that our rapid releases change things all the time.
>
> I can't help but wonder why you used the word "fear" in the above
> sentence. Do you think it's unwarranted or incorrect?

For one thing, I think irrational fear of change is bad. For the other,
I used the word metathesiophobia, which is "fear of change". And I was
referring to the other -phobia being talked about in the previous post.

Robert Kaiser

[Andrew Joakimsen]

I don't think its irrational if the browser you have known and liked
for many years suddenly starts to degrade and become unusable to you.
Why would you want what the developer calls the new and improved
version if constantly it is becoming worse for you the user?

[Chris Ilias]

On 12-07-06 3:03 PM, Andrew Joakimsen wrote:
> I don't think its irrational if the browser you have known and liked
> for many years suddenly starts to degrade and become unusable to you.
> Why would you want what the developer calls the new and improved
> version if constantly it is becoming worse for you the user?

Let's not let this thread turn into a "What's your problem with Firefox"
thread. Benoit noticed a dip in Firefox market share on statcounter for
June (whereas it was gaining market share in previous months), and he asked:
* is that consistent with other market share reports?
* is there data that suggests a reason?
* does crash-stats.mozilla.com indicate an increase of crashes?

Notice how that's data-driven. :)


If you want to give feedback, that's great, but I think it should be
separate from this, and detailed, so developers can take action on it.

[Benoit Jacob]

The trend continues:
http://gs.statcounter.com/#browser-ww-daily-20120101-20120802

AFAICS, the only remaining plausible explanation at this point is the Flash
11.3 crashes, which seem to account for the +50% increase in the number of
crash reports that we get:
http://people.mozilla.org/~bjacob/gfx_features_stats/#num-reports

I know that people are all over it already, but I must say, I would be
interested in knowing what the plan is!

Benoit

PS. Thanks Chris for restoring sanity here, and let's not turn this into a
flamefest again.

2012/7/6 Chris Ilias <nm...@ilias.ca>

> ______________________________**_________________
> dev-apps-firefox mailing list
> dev-apps-firefox@lists.**mozilla.org <dev-apps...@lists.mozilla.org>
> https://lists.mozilla.org/**listinfo/dev-apps-firefox<https://lists.mozilla.org/listinfo/dev-apps-firefox>
>

[Eric Shtivelberg]

I think it's just sad to see a great browser like Firefox having it's users
leave it for other browsers some of which I think are worse, but this isn't
that discussion.

I personally don't think it's only because of stability going up and down
with every release and other stuff like features dissappearing, I think we
once had a great website back when firefox 3 was just about to release that
had buttons and different stuff to let people to share firefox on their
website and with their friends.

Maybe the website is still up but I just don't see those buttons anymore.

PS: I don't see what other ppl see in other browsers, firefox is far
better, uses less memory and is faster :)


Regards,
Eric

On Sat, Aug 4, 2012 at 4:25 AM, Benoit Jacob <jacob.b...@gmail.com>wrote:

> The trend continues:
> http://gs.statcounter.com/#browser-ww-daily-20120101-20120802
>
> AFAICS, the only remaining plausible explanation at this point is the Flash
> 11.3 crashes, which seem to account for the +50% increase in the number of
> crash reports that we get:
> http://people.mozilla.org/~bjacob/gfx_features_stats/#num-reports
>
> I know that people are all over it already, but I must say, I would be
> interested in knowing what the plan is!
>
> Benoit
>
> PS. Thanks Chris for restoring sanity here, and let's not turn this into a
> flamefest again.
>
> 2012/7/6 Chris Ilias <nm...@ilias.ca>
>

> > ______________________________**_________________
> > dev-apps-firefox mailing list
> > dev-apps-firefox@lists.**mozilla.org <dev-apps...@lists.mozilla.org
> >
> > https://lists.mozilla.org/**listinfo/dev-apps-firefox<
> https://lists.mozilla.org/listinfo/dev-apps-firefox>
> >

[Mart Rootamm]

Eric Shtivelberg wrote:
> I think it's just sad to see a grmeat browser like Firefox having it's users

> leave it for other browsers some of which I think are worse, but this isn't
> that discussion.

But look at the upside — There is no word yet about Safari 6.0 for Windows :>
(Of course, (non-)appearances can be deceptive, too...)

Personally, there have so far been two gripes with Firefox that I've
taken some issue with:
* One was the disabling of on-demand loading of pinned tabs after
session restore (between versions 9–11);
* The other is the current brouhaha over Adobe's Flash crashing the
plugin container process, which is really not the fault of Mozilla.
(more below)

The rest of this writeup has become somewhat tl;dr, I hope you won't mind.

Sometimes it's not users leaving Firefox, but some of them starting to
use Chrome as their very first browser. Well, Chrome coming around is
a good thing, because this gives people more choice as to which
browser they want to use, as Chrome and Firefox both possess unique
and attractive features that meet their users' different needs.

HTML5 video
The current situation with Flash crashing the plugin container in
Firefox is coincidentally a good cause for moving to HTML5 audio and
HTML5 video, specifically Ogg Vorbis, Ogg Theora, and WebM, which are
free and especially license-free formats.

YouTube's work in converting most of its videos to WebM reduces the
immediate requirement and sometimes unpleasant chore of installing
Flash on Linux, thereby increasing adoption of Linux, as other sites
will hopefully follow suit in adopting free formats.

If we exclude the Summer low and the current Flash issue, then the
next reason behind a reduction in Firefox usage could be the choice of
format in sites using HTML5 video — most users tend to choose the
browser that plays back whatever their favoured media site offers,
with variations (mobile/desktop) of Chrome being in a rather
advantageous situation right now, as it has built-in support for
Flash.

Yet the situation with HTML5 video seems to be split right now along
the lines of which HTML5 codecs are supported by which groups of
browsers: Safari and IE vs. Opera, Chrome, Firefox and its
derivatives.

The choice of YouTube and DailyMotion to offer videos in license-free
formats is highly commendable. Now, if YouTube could actually stream
high-profile events using HTML5/WebM in addition to Flash...

Desktop to mobile/tablet
Yet another reason in reduction of Firefox market share could just as
well be the transition of people's major computing devices from
desktops (including notebooks) to hand-helds (smartphones, tablets),
nearly all of which currently have WebKit as their main rendering
engine (in the form of either Safari or Chrome). I do not know if
there has been a separate browser market share comparison for just
desktop computers, because I understand that general tallies have
usually encompassed both desktop and mobile spaces, with mobile being
the separate segment.

Ultimately, as Mozilla and then Firefox were first introduced, it was
hoped that the browser market would eventually take the shape that it
of recent times has started to form (at least worldwide) — in that no
one browser would completely rule the market to be in its singularity
the one to hold back innovation, and the one to pose itself in
unintended consequence a widespread vector for malicious attacks.

So, in conclusion, the situation, in my humble and perhaps
half-informed opinion, is quite a bit more mixed with regard to what
may be the possible causes of Firefox browser market share reduction
this Spring and Summer.

-M.

2012/8/4, Eric Shtivelberg <shed...@gmail.com>:
> I think it's just sad to see a grmeat browser like Firefox having it's users

>> > ______________________________**_________________
>> > dev-apps-firefox mailing list
>> > dev-apps-firefox@lists.**mozilla.org
>> > <dev-apps...@lists.mozilla.org
>> >
>> > https://lists.mozilla.org/**listinfo/dev-apps-firefox<
>> https://lists.mozilla.org/listinfo/dev-apps-firefox>
>> >

[Benoit Jacob]

I don't want to keep reviving this topic forever, but two things make
me do it again today:

1. the current market share loss can now be called our single largest
market share loss ever over a 3 month period: almost 3 percentage
points in 3 months.

http://gs.statcounter.com/#browser-ww-monthly-200901-201208

2. Today I heard otherwise well-informed people at Mozilla claiming
that our market share had now stabilized, which makes me think that
this data is not yet publicized enough. Maybe I am on the wrong
channel?

Again, this market share loss matches the recent 2x increase in crash
reports we receive, which matches the Flash 11.3 crash issues. I would
be really interested in knowing how we're doing on this front. Let me
know if another channel is more appropriate for that.

Benoit

2012/8/3 Benoit Jacob <jacob.b...@gmail.com>:

> The trend continues:
> http://gs.statcounter.com/#browser-ww-daily-20120101-20120802
>
> AFAICS, the only remaining plausible explanation at this point is the Flash
> 11.3 crashes, which seem to account for the +50% increase in the number of
> crash reports that we get:
> http://people.mozilla.org/~bjacob/gfx_features_stats/#num-reports
>
> I know that people are all over it already, but I must say, I would be
> interested in knowing what the plan is!
>
> Benoit
>
> PS. Thanks Chris for restoring sanity here, and let's not turn this into a
> flamefest again.
>
>
> 2012/7/6 Chris Ilias <nm...@ilias.ca>
>>

[Martijn]

On Wed, Aug 29, 2012 at 7:27 PM, Benoit Jacob <jacob.b...@gmail.com> wrote:
> I don't want to keep reviving this topic forever, but two things make
> me do it again today:
>
> 1. the current market share loss can now be called our single largest
> market share loss ever over a 3 month period: almost 3 percentage
> points in 3 months.
>
> http://gs.statcounter.com/#browser-ww-monthly-200901-201208
>
> 2. Today I heard otherwise well-informed people at Mozilla claiming
> that our market share had now stabilized, which makes me think that
> this data is not yet publicized enough. Maybe I am on the wrong
> channel?
>
> Again, this market share loss matches the recent 2x increase in crash
> reports we receive, which matches the Flash 11.3 crash issues. I would
> be really interested in knowing how we're doing on this front. Let me
> know if another channel is more appropriate for that.

Flash 11.4 has been released recently, which seems to have fixed all
the crashes I was seeing with Flash plugins.

Regards,
Martijn

> Benoit
>
> 2012/8/3 Benoit Jacob <jacob.b...@gmail.com>:
>> The trend continues:
>> http://gs.statcounter.com/#browser-ww-daily-20120101-20120802
>>
>> AFAICS, the only remaining plausible explanation at this point is the Flash
>> 11.3 crashes, which seem to account for the +50% increase in the number of
>> crash reports that we get:
>> http://people.mozilla.org/~bjacob/gfx_features_stats/#num-reports
>>
>> I know that people are all over it already, but I must say, I would be
>> interested in knowing what the plan is!
>>
>> Benoit
>>
>> PS. Thanks Chris for restoring sanity here, and let's not turn this into a
>> flamefest again.
>>
>>
>> 2012/7/6 Chris Ilias <nm...@ilias.ca>
>>>

>>> _______________________________________________
>>> dev-apps-firefox mailing list
>>> dev-apps...@lists.mozilla.org
>>> https://lists.mozilla.org/listinfo/dev-apps-firefox
>>
>>
> _______________________________________________
> dev-apps-firefox mailing list
> dev-apps...@lists.mozilla.org
> https://lists.mozilla.org/listinfo/dev-apps-firefox

--
Martijn Wargers - Help Mozilla!
http://quality.mozilla.org/
http://wiki.mozilla.org/Mozilla_QA_Community
irc://irc.mozilla.org/qa - /nick mw22

[Boris Zbarsky]

On 8/29/12 1:27 PM, Benoit Jacob wrote:
> 2. Today I heard otherwise well-informed people at Mozilla claiming
> that our market share had now stabilized, which makes me think that
> this data is not yet publicized enough.

I think it depends on which set of metrics you're looking at...

-Boris

[Benoit Jacob]

2012/8/29 Boris Zbarsky <bzba...@mit.edu>:

In fact, I saw someone very well informed looking exactly at the same
StatCounter graph, except it was only on the January 2012 -> July 2012
interval, so it didn't include the August data and didn't include
older data for context; and that person concluded that our market
share was "flat" according to this graph. Goes to show how misleading
graphs can be. Computing actual slope and comparing to the worst
slopes we had in the past, would have avoided this.

Benoit

>
> -Boris

[Jeremy Morton]

I think Firefox may be losing marketshare because it's trying to become
Chrome. I asked this question elsewhere and got no response - if
Firefox becomes Chrome, why shouldn't people just go ahead and use
Chrome itself?

I say Firefox should be working hard to maintain a separate identity.
The user interface is a very visible part of that user identity. You
aren't going to keep users who really want Chrome - they'll switch to
Chrome. This means keeping tabs-on-bottom, large toolbar icons, and
generally keeping the UI as customizable as possible. Take this away
and Firefox is basically a slower version of Chrome.

--
Best regards,
Jeremy Morton (Jez)

[Mart Rootamm]

>
> This means keeping tabs-on-bottom, large toolbar icons, and generally
> keeping the UI as customizable as possible.
>

I actually like the tabs-on-top feature, because it allows me to save
screen real estate, although I understand people who like keeping
tabs-on-bottom (of the top/toolbar side of the browser).

Firefox has many of the features that are still unique compared to other
browsers.

What I suspect is that some stats might be skewed, as some of them don't
make a distinction between desktop and mobile browsers, and Firefox is
primarily a desktop browser, while Google Chrome is almost everywhere in
mobile.

Maybe if Firefox on mobile takes off, with Firefox OS being included in
some smartphones, then the statistic could change.

If we look at this popular image —
https://en.wikipedia.org/wiki/File:Countries_by_most_used_web_browser.svg
— then we can see that Chrome is popular not even by regions, but by
language groups and cultural affiliation:
* Roman languages (Latin America, Spain, Portugal, and Italy);
* Indic languages (South Asia: India, Pakistan, Nepal);
* Slavic languages (Russia, Ukraine and several post-Soviet territories way
east);

Chrome is also used in Turkey and on much of the Arabian peninsula.

Firefox seems to be preferred along similar time zones in a North-South
direction: Much of Europe and Africa. Interestingly, also Indonesia and
some South-East Asian countries.

I'd attribute the expansion of Chrome mostly to new users and maybe also
those who live in an environment where they don't require language skills
beyound their first or primary (or official) langauge. These are regions
and countries with large populations.

Internet Explorer probably fares similarly, because it's mostly used in
English-speaking countries, and China and Japan.

This could be half-attributed to current cultural influences, in as much as
which geographic region of the world has influence over which cluster of
countries.

-Mart.

2012/8/29 Jeremy Morton <m...@game-point.net>

> On 02/07/2012 23:24, Benoit Jacob wrote:
>
>> Hi,
>>
>> According to statcounter, Firefox has been losing market share very
>> fast since the end of May, whereas it was making gains in the previous
>> months:
>>

>> http://gs.statcounter.com/#**browser-ww-daily-20120101-**20120701<http://gs.statcounter.com/#browser-ww-daily-20120101-20120701>

>>
>> Do you think that this is just noise or meaningful? If meaningful,
>> what could be possible reasons? Has an increase in crashes been
>> observed?
>>
>>
> I think Firefox may be losing marketshare because it's trying to become
> Chrome. I asked this question elsewhere and got no response - if Firefox
> becomes Chrome, why shouldn't people just go ahead and use Chrome itself?
>
> I say Firefox should be working hard to maintain a separate identity. The
> user interface is a very visible part of that user identity. You aren't
> going to keep users who really want Chrome - they'll switch to Chrome.
> This means keeping tabs-on-bottom, large toolbar icons, and generally
> keeping the UI as customizable as possible. Take this away and Firefox is
> basically a slower version of Chrome.
>
> --
> Best regards,
> Jeremy Morton (Jez)
>

[Robert Kaiser]

Martijn schrieb:

> Flash 11.4 has been released recently, which seems to have fixed all
> the crashes I was seeing with Flash plugins.

There seem to have been some pretty good improvements in crashes there,
but it's probably still about twice as crashy as 11.2 was (though our
current data is very preliminary due to slow 11.4 adoption).
Hangs of the Flash plugin (i.e. 45s of no response, followed by us
killing the plugin) seem to stay stable at roughly four times the amount
we had when 11.2 was still the current version.

We in the stability group are watching this very closely and people on
our side are continuing to work closely with Adobe to help them improve
even further.

Robert Kaiser

[Dao]

On 29.08.2012 23:40, Robert Kaiser wrote:
> There seem to have been some pretty good improvements in crashes there,
> but it's probably still about twice as crashy as 11.2 was (though our
> current data is very preliminary due to slow 11.4 adoption).
> Hangs of the Flash plugin (i.e. 45s of no response, followed by us
> killing the plugin) seem to stay stable at roughly four times the amount
> we had when 11.2 was still the current version.
>
> We in the stability group are watching this very closely and people on
> our side are continuing to work closely with Adobe to help them improve
> even further.

Adobe improving stuff seems to take too long. We should have been more
proactive two months ago. It's still unclear to me why this hasn't happened.

[Boris Zbarsky]

On 8/29/12 6:02 PM, Dao wrote:
> We should have been more proactive two months ago.

Is there a concrete suggestion for what we should have done and didn't?

-Boris

[Dao]

We should have disabled the apparently-broken protected mode.

[Benjamin Smedberg]

Mozilla specifically asked Adobe on several occasions to disable the
protected mode feature of Flash because of the ongoing performance and
stability regressions. Adobe has declined to do that.

We have recommended to users on our support site who are experiencing
problems to downgrade the Flash ESR releases, or to edit the
configuration file manually. But it is not right nor safe for Firefox to
modify another application's configuration files (absent security
exploits caused by protected mode itself).

--BDS

[Dao]

On 30.08.2012 15:23, Benjamin Smedberg wrote:
> On 8/29/2012 6:26 PM, Dao wrote:
>> On 30.08.2012 00:22, Boris Zbarsky wrote:
>>> On 8/29/12 6:02 PM, Dao wrote:
>>>> We should have been more proactive two months ago.
>>>
>>> Is there a concrete suggestion for what we should have done and didn't?
>>
>> We should have disabled the apparently-broken protected mode.
>
> Mozilla specifically asked Adobe on several occasions to disable the
> protected mode feature of Flash because of the ongoing performance and
> stability regressions. Adobe has declined to do that.

And that's it, end of story? Are we going to try the same strategy if
Adobe releases Flash 11.5 next month with a new half-broken
Firefox-specific feature?

> We have recommended to users on our support site who are experiencing
> problems to downgrade the Flash ESR releases, or to edit the
> configuration file manually.

This only reaches a fraction of affected users, it doesn't scale. As
suspected much earlier and more or less verified by now, lots of users
just abandon Firefox rather than consulting SUMO.

> But it is not right nor safe for Firefox to
> modify another application's configuration files (absent security
> exploits caused by protected mode itself).

I don't think it's cut and dried like this, especially since Protected
Mode specifically targets Firefox.

[Robert Kaiser]

Dao schrieb:

According to Adobe, that's for one a configuration they do not test
internally at all, and for the other a significant security risk, as all
"Protected Mode" is there for is (proactive) security.

Still, I'd be happy if it would be deactivated, it would spare us a ton
of crashes and even more hangs - though I would see the probably
resulting level (roughly what 11.2 had) still too high for the long run.

Right now, I think one thing we really need to push on is finding
alternatives so that Flash can completely die ASAP.

Robert Kaiser

[John Jensen]

On 8/31/2012 9:38 AM, Robert Kaiser wrote:

> Right now, I think one thing we really need to push on is finding
> alternatives so that Flash can completely die ASAP.

Yes, agreed, but 99.9% of users use Flash [1] and it is on the homepages
of 38% of the top 35,000 websites [2]. It will take some time for it to
die and we need to think about better ways of bridging the gap.

[1]
http://blog.chromium.org/2012/07/npapi-plug-ins-in-windows-8-metro-mode.html
[2] http://httparchive.org/interesting.php#flash

John

--
John Jensen
Product Strategist
Mozilla Corporation

[Robert Kaiser]

John Jensen schrieb:

> On 8/31/2012 9:38 AM, Robert Kaiser wrote:
>
>> Right now, I think one thing we really need to push on is finding
>> alternatives so that Flash can completely die ASAP.
>
> Yes, agreed, but 99.9% of users use Flash [1] and it is on the homepages
> of 38% of the top 35,000 websites [2]. It will take some time for it to
> die and we need to think about better ways of bridging the gap.

Agreed. That's why I said we should push on finding alternatives, and
didn't say we should kill it right now. There's surely a long way to go
until Flash can die, but the more we can help sites use alternatives,
the lower its overall usage will be in terms of usage hours, and the
fewer crashes and hangs we'll see. And that in the end will make our
users happy.

At the same time, we'll of course continue to work with Adobe so they
fix as many issues as possible in further updates. We had some success
with this already, as seen in slowly (way too slowly for my taste, but
at least it's something) decreasing amounts of issues in later 11.3
updates as well as in the 11.4 release. Benjamin Smedberg has done a ton
of work there (and some people in our QA team have also been quite
helpful in finding STR), and he's working on more things so we can help
Adobe get better info esp. on the "hangs" (i.e. when the Flash process
doesn't react for 45s and we kill it).

Robert Kaiser

[Benoit Jacob]

Another month, another update.

There is still no sign of decrease in the overall number of crash
reports that we get,
http://people.mozilla.org/~bjacob/gfx_features_stats/#num-reports
Although I haven't checked again if the proportion of Flash-related
crashes has changed.

In september, according to statscounter, we lost another 0.45
percentage point, which is another large loss, although smaller than
the ones we had had for the previous 3 months (almost 1 percentage
point per month).
http://gs.statcounter.com/#browser-ww-monthly-201201-201209

Benoit


2012/9/1 Robert Kaiser <ka...@kairo.at>:

[Henri Sivonen]

On Tue, Oct 2, 2012 at 5:31 AM, Benoit Jacob <jacob.b...@gmail.com> wrote:
> There is still no sign of decrease in the overall number of crash
> reports that we get,
> http://people.mozilla.org/~bjacob/gfx_features_stats/#num-reports

Have non-latest crashy versions of Flash Player (particularly 11.3.x)
been softblocked to get users off the crashiest versions?

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

[Mart Rootamm]

Henri's observation is apt.

http://helpx.adobe.com/flash-player/kb/archived-flash-player-versions.html
This ^ has it that the latest 11.3 release is 11.3.300.271 (released on
14.08.2012) and all 11.3 versions before it should be blocked.

_If_ the whole 11.3 branch is not okay, then the latest 11.4 should be
suggested — with earlier point releases also soft-blocked, if 11.4 has been
known to exhibit crashiness.

I might add, that since Firefox 13 or 14 I've observed a degradation of
performance with plugin-container.exe when playing Flash videos and YouTube
music. The symptom is that playback becomes modestly or moderately jerky,
which is annoying.

To resolve it, I must set the process priority for plugin container to
either Above Normal or even High in Process Explorer every time I start
Firefox.

-Mart.

2012/10/2 Henri Sivonen <hsiv...@iki.fi>

[Henri Sivonen]

On Tue, Oct 2, 2012 at 9:44 AM, Henri Sivonen <hsiv...@iki.fi> wrote:
> On Tue, Oct 2, 2012 at 5:31 AM, Benoit Jacob <jacob.b...@gmail.com> wrote:
>> There is still no sign of decrease in the overall number of crash
>> reports that we get,
>> http://people.mozilla.org/~bjacob/gfx_features_stats/#num-reports
>
> Have non-latest crashy versions of Flash Player (particularly 11.3.x)
> been softblocked to get users off the crashiest versions?

What about non-latest RealPlayer for
https://support.mozilla.org/en-US/kb/flash-113-doesnt-load-video-firefox

[Benjamin Smedberg]

On 10/2/2012 7:09 AM, Henri Sivonen wrote:
> What about non-latest RealPlayer for
> https://support.mozilla.org/en-US/kb/flash-113-doesnt-load-video-firefox

What about it? The product in question is not a browser plugin and it's
not a DLL loaded into a Mozilla process.

--BDS

[Benjamin Smedberg]

On 10/2/2012 2:44 AM, Henri Sivonen wrote:
> Have non-latest crashy versions of Flash Player (particularly 11.3.x)
> been softblocked to get users off the crashiest versions?

No. The current plugin blocking is bad enough that it would be a worse
user experience to block those plugins than to keep them running. We aim
to block all the insecure versions with the release of Firefox 17 with
its improved blocklisting including click-to-play plugin blocking and
better upgrade UI.

--BDS

[Henri Sivonen]

On Tue, Oct 2, 2012 at 4:34 PM, Benjamin Smedberg <benj...@smedbergs.us> wrote:
> On 10/2/2012 2:44 AM, Henri Sivonen wrote:
>>
>> Have non-latest crashy versions of Flash Player (particularly 11.3.x) been
>> softblocked to get users off the crashiest versions?
>
> No. The current plugin blocking is bad enough that it would be a worse user
> experience to block those plugins than to keep them running.

At least having a softblock in place would inform users that they
could rather update Flash than blame the bad experience on Firefox and
switch to another browser.

> We aim to block
> all the insecure versions with the release of Firefox 17 with its improved
> blocklisting including click-to-play plugin blocking and better upgrade UI.

Nice!

Interesting. How can it interfere with Flash Player then? I thought
Real had a wrapper NPAPI plug-in that took over .swf and hosted the
real Flash Player inside of the wrapper plug-in.

[Benjamin Smedberg]

On 10/2/2012 10:28 AM, Henri Sivonen wrote:
> Interesting. How can it interfere with Flash Player then? I thought
> Real had a wrapper NPAPI plug-in that took over .swf and hosted the
> real Flash Player inside of the wrapper plug-in.

No. It is a DLL that injects itself in processes with the Flash DLL and
takes them over, but not as an NPAPI plugin. And since it was trying to
take over the FlashPlayerPlugin.exe process, even our DLL blocklist is
not involved.

--BDS

[Robert Kaiser]

Benoit Jacob schrieb:

> There is still no sign of decrease in the overall number of crash
> reports that we get,
> http://people.mozilla.org/~bjacob/gfx_features_stats/#num-reports
> Although I haven't checked again if the proportion of Flash-related
> crashes has changed.

Yes, Flash hasn't become significantly better recently. Compared to the
awfully bad versions in June, they improved slightly in July, but not at
all since then. Some people on our side are working with them to
diagnose the problems, but they are very slow or unable to find out
about those themselves, apparently - or to fix them. Compared to Flash
11.2, current Flash is still 3-4 times worse in terms of crashes and ~4
times worse in terms of hangs on Windows Vista and higher (XP, without
protected mode, is at about the same level).

Robert Kaiser

[Mart Rootamm]

Many people are going to install the latest Flash Player tonight
(11.4.402.278), because video.pbs.org wants it:
http://video.pbs.org/video/2284525928

Linux users are SOL, because Adobe haven't released Flash 11.4 for Linux :/

-Mart.

2012/10/3 Robert Kaiser <ka...@kairo.at>

> Benoit Jacob schrieb:
>
> There is still no sign of decrease in the overall number of crash
>> reports that we get,

>> http://people.mozilla.org/~**bjacob/gfx_features_stats/#**num-reports<http://people.mozilla.org/~bjacob/gfx_features_stats/#num-reports>

>> Although I haven't checked again if the proportion of Flash-related
>> crashes has changed.
>>
>
> Yes, Flash hasn't become significantly better recently. Compared to the
> awfully bad versions in June, they improved slightly in July, but not at
> all since then. Some people on our side are working with them to diagnose
> the problems, but they are very slow or unable to find out about those
> themselves, apparently - or to fix them. Compared to Flash 11.2, current
> Flash is still 3-4 times worse in terms of crashes and ~4 times worse in
> terms of hangs on Windows Vista and higher (XP, without protected mode, is
> at about the same level).
>
>
> Robert Kaiser