FF 62.0.3 - Attention! Telemetry Coverage 7.0 installed without permission

[Centauri39]

What on earth is going on at Mozilla these days?

In the FF settings, section "privacy and security", they say, they'd ask
for permission before data is going to be sent to Mozilla.
But then they are installing this add-on (this one is(!) sending data)
without asking me for my permission or even notifiying me?
Why are they lying to us?

And when I disabled it, it wasn't moved to the disabled-section of my
add-ons, no, it was gone!
So, I had to look for it in the FF profil and delete it from the
subfolder features.

On a 2nd machine, in the same folder, I found a "fxmonitor.xpi".
more spyware in FF?
Of course I deleted that one, too.

[Jeff Layman]

See my post on 11 Oct "Allowed web sites for add-on installation" and
following thread, particularly my reply to WaltS48.

--

Jeff

[WaltS48]

On 10/14/18 6:12 AM, Centauri39 wrote:
> What on earth is going on at Mozilla these days?
>

You are a 1%er!

"Mozilla created the Telemetry Coverage system add-on and distributed
it to 1% of the Firefox population. The add-on is automatically
installed and designed to inform Mozilla whether Telemetry is enabled in
the browser."

<https://www.ghacks.net/2018/09/21/mozilla-wants-to-estimate-firefoxs-telemetry-off-population/>

It appears to be a System add-on, listed under Firefox Features in the
about:support page.

Here is my list for version 62.0.3.

Activity Stream 2018.08.22.1219-93becf29
Application Update Service Helper 2.0
Firefox Monitor 2.3
Firefox Screenshots 33.0.0
Form Autofill 1.0
Photon onboarding 1.0
Pocket 1.0.5
Telemetry coverage 7.0
Web Compat 2.0
WebCompat Reporter 1.0.0

In my Nightly 64.0a1 I only have a few.

Firefox Screenshots 33.0.0
Form Autofill 1.0
Web Compat 2.0.1
WebCompat Reporter 1.1.0

--
OS: Ubuntu Linux 18.04LTS - Gnome Desktop
https://www.thunderbird.net/en-US/get-involved/

[WaltS48]

On 10/14/18 6:12 AM, Centauri39 wrote:

More information for you.

<https://firefox-source-docs.mozilla.org/toolkit/mozapps/extensions/addon-manager/SystemAddons.html>

Firefox Monitor is a service.

"Firefox Monitor arms you with tools to keep your personal information
safe. Find out what hackers already know about you and learn how to stay
a step ahead of them."

<https://monitor.firefox.com/>

[Centauri39]

No, I'm not a liar.

The number of users is absolutely irrelevant.
The point is all this happened silently without notifying me or asking
me for permission.
Sorry. but this is something you usually find within malware.

[Christian Riechers]

>From what I've seen so far Firefox Monitor is a website.
Why would that need to be an add-on?

[Centauri39]

So why is it happening secretly?
Sorry, but I(!) am the one to decide what to install on my(!) machine
and I want to be notified before(!) something is going on.

[WaltS48]

Not sure.

As you can see from my other reply, my 62.0.3 has Firefox Monitor 2.3
installed in the Firefox Features section of Troubleshooting
Information. My Nightly 64.0a1 does not.

Maybe it was a test.

[WaltS48]

Never said that you were. You and I were chosen to be in that 1% of the
population.

Shame I'm not helping Mozilla since I don't use the release version
unless it is to check a users problem.

>
> The number of users is absolutely irrelevant.
> The point is all this happened silently without notifying me or asking
> me for permission.
> Sorry. but this is something you usually find within malware.
>

I think they did. You just have to keep up with blogs and tech news
sites and learn things.

[Chris Ilias]

Don't worry; if you have disabled "Allow Firefox to send technical and
interaction data to Mozilla", that data is not being sent. :)

There is no setting to ask for permission before sending data. Are you
are referring to the "Warn you when websites try to install add-ons"
setting?

Assuming that's the setting you're referring to, I'll try to clear this up.
What you're seeing are system add-ons, which are not add-ons installed
from websites. They are features included in Firefox, such as
screenshots and Pocket, that are installed as add-ons, so they can be
updated separately from Firefox.

I'm not sure if they are included in the exceptions list, but as Jeff
Layman pointed out, you can check the exceptions list for sites set to
"Allow".

Just to be clear, sending data and installing a feature are two
different things, and so long as you have disabled "Allow Firefox to
send technical and interaction data to Mozilla", that data is not being
sent.

--
<http://ilias.ca/links>
Mailing list/Newsgroup moderator

[Wolf K]

On 2018-10-14 11:36, Centauri39 wrote:
> Am 14.10.2018 um 15:11 schrieb WaltS48:
>> On 10/14/18 6:12 AM, Centauri39 wrote:
>>> What on earth is going on at Mozilla these days?
>>>
>>
>> You are a 1%er!

[...]

> No, I'm not a liar.

WaltS48 called you a 1%-er, ie, one of the 1% to whom Moz sent the
telemetry add-on. Not a "liar".

> The number of users is absolutely irrelevant.
> The point is all this happened silently without notifying me or asking
> me for permission.
> Sorry. but this is something you usually find within malware.

I have mixed feelings about this. OTOH, Moz didn't ask permission. OTOH,
they are merely asking for the answer to a question, not actual data.
After all, telemetry is turned off by default, and Moz asks you about it
when you first install FF.

I don't know how many people actually allow telemetry (I don't), but I
suspect Moz is worried that it's not a representative sample. The
repeated outcries when Moz changes some feature of FF certainly suggest
the telemetry data which prompted the changes may be skewed. A random
sample of 1% of users (a very high proportion with such a huge
user-base) will give them a better idea of whether or not the telemetry
numbers are good enough for a representative sample.

However, since they ask permission for telemetry, the sample is
self-selected, which means it's not random, which means that it's
unknown how representative the sample is.

Getting good statistics is hard, sometimes very hard. Thus, trying to
determine what could be a good change to FF is also hard. The number of
downloaded extensions are IMO a better indicator of what FF users want.
Well-behaved add-ons are IMO a better architecture than loading the
program with features most of which aren't wanted by most users, even
though each of them is wanted by a fairly large proportion of users.

Best,


--
Wolf K
kirkwood40.blogspot.com
It's called an "opinion" because it's not a fact.

[Jeff Layman]

On 15/10/18 04:42, Wolf K wrote:

> Getting good statistics is hard, sometimes very hard. Thus, trying to
> determine what could be a good change to FF is also hard. The number of
> downloaded extensions are IMO a better indicator of what FF users want.
> Well-behaved add-ons are IMO a better architecture than loading the
> program with features most of which aren't wanted by most users, even
> though each of them is wanted by a fairly large proportion of users.

It might be considered that a survey would be useful to find out whether
users are or are not satisfied with FF, but even that seems to have
issues the way Mozilla handles them. In my earlier post "Allowed web
sites for add-on installation" I did not mention that after taking some
notes I went back to FF to close it, only to find it had automatically
opened a tab for a FF survey - "Thank you for taking this short survey.
Your responses are important for making
Firefox better. Please answer all questions to the best of your ability. "

The survey had five questions:
How would you rate your experience with Firefox in the past 30 days?*
How strongly do you agree or disagree with the following statement:
Firefox interrupts my web browsing too often?*
How strongly do you agree or disagree with the following statements?*
Firefox makes it easy to find sites I have seen before
Firefox makes it easy to find new content on the internet
Firefox makes it easy to search the web

Viewing the page source showed that it ran to almost 600 lines of code.
I saved it as a text file, which is 146.6kB in size. But I don't
understand what all that code is for (some was for different languages).
Why so much for a simple five-question survey?

--

Jeff

[Wolf K]

This is a simplified answer:

Before any content is added, the page must be described. That requires
code to inform the browser (and via the browser the OS) what that page
is supposed to look like. Every detail must be coded, eg, the size of
the answer box, its position, its background and foreground colours, and
so on. This must be done for every element of the web page. IOW, merely
presenting an _empty_ web page takes a chunk of code. In addition you
need code for those different languages; to present the user interface
(web page) as a fillable form; and to record and transmit that form when
filled.

For comparison look at the size of a simple text file of just one
sentence as a Word document (text.doc or text.docx) and as a plain-text
document (text.txt).

I don't know whether you've ever done any coding. I have. I can tell you
that, for example, computing the contents of a few thousand spreadsheet
cells takes a good deal less computing power than presenting a graphical
user interface such as a browser window on a screen. On a typical
personal-use machine, most of the computing is done to present and
maintain that user interface, so that you can have a "seamless computing
experience".

[Jeff Layman]

No, I've unfortunately not done any coding. I take your point that the
pages can look very involved, and do need a lot of code. I think the
problem I have is that the operation of the webpage is opaque to me, and
I don't know what is going on "in the background". I was once again
annoyed that Mozilla chose to open a page on a new tab without asking.
All other organisations I've seen use a popup or a link on an
already-open page to *ask* whether or not I would like to take part in a
survey.

--

Jeff