Remove Pocket integration

RESOLVED INVALID

Status

()

Firefox
Untriaged
RESOLVED INVALID
2 years ago
5 months ago

People

(Reporter: tucker.mcknight, Unassigned)

Tracking

40 Branch
Points:
---

Firefox Tracking Flags

(Not tracked)

Details

(URL)

(Reporter)

Description

2 years ago
User Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.10; rv:40.0) Gecko/20100101 Firefox/40.0
Build ID: 20150603004005

Steps to reproduce:

Mozilla's recent integration with Pocket, a proprietary third-party service, is a mistake.

It is very exciting to see the ways in which Firefox continues to improve. And it's even more exciting to see the ways that Mozilla advances it's stated mission outside of the Firefox browser with new developments like Firefox Accounts. Pocket now allows you to log in on their site using your Firefox Account; being able to authenticate with a trusted third party like Mozilla is a huge win for online privacy advocates and the Mozilla community. However, adding Pocket as a built-in feature to Firefox should not have been done.

This is particularly surprising since it was Firefox that made browser extensions mainstream. Pocket should have been an extension (in fact, a Pocket extension used to exist). It could have even been bundled with the browser. This distinction is important, since extensions can be removed entirely, whereas currently Pocket can only be disabled. 

The user experience of disabling Pocket is not good, either. It needs to be disabled in about:config, which is not at all user friendly, and therefore not in line with Mozilla's mission. In the past, Mozilla has been very good about showing the user what new features have been added to the interface and explaining any privacy implications that may come with them. That is why I was so surprised when the Pocket icon suddenly appeared in Firefox Developer Edition a couple days ago. It is so unlike Mozilla to introduce something like that, I ran a virus scan and checked what programs had been installed recently -- I assumed it had been put there in the same way that IE users used to get the Ask Toolbar installed. 

It may also not be clear to some users that, even when signing in with your Firefox account, you are still giving your email address to a third party whose privacy policy is different than Mozilla's. Many users would not assume this, since it is a feature that is bundled with the browser.

Mozilla's recent blog post about the Pocket feature is titled "Firefox Puts You in Control of Your Online Life" (https://blog.mozilla.org/blog/2015/06/02/firefox-puts-you-in-control-of-your-online-life/). Had this been coming from a startup, that post would be humorously ironic. But given how much people care about Mozilla and it's stated mission, it is more painful than funny.

Firefox should continue to add new features that benefit its users, but those features must be done in accordance with Mozilla's core values. This feature should've been done as an extension, which allows for greater user choice and avoids bloat. Most importantly, there was very little public discussion about this inclusion of a proprietary, third-party service. It's a huge departure from Mozilla's commitment to transparency. The existence of the Pocket code in Firefox is a bug in the browser, and it does not adhere to Mozilla's core mission.
Comment hidden (me-too)
Bugzilla is not for discussion of product decisions. Please use Governance or input, https://groups.google.com/forum/#!forum/mozilla.governance
https://input.mozilla.org/feedback

Updated

2 years ago
Status: UNCONFIRMED → RESOLVED
Last Resolved: 2 years ago
Resolution: --- → INVALID
If I'm not wrong, Pocket can be disabled with the customizable UI (hamburger menu -> customize, drag the icon out)

Comment 4

2 years ago
@Tyler: What is a "product decision", and how does it differ from feature requests and bug reports in a way that it is not suitable for the bugtracker?

@manishearth: The button is just hidden with this, but the corresponding code not properly disabled. Given that Pocket integration is open-source, there is IMO no difference for the user.

--

How will the inclusion of Pocket affect the "reading list" feature Fennec currently has, and the "reading mode" Firefox and Fennec have?
Pocket is just a bunch of API calls. Firefox UI code is lazy loaded. Put those two together, and yes, Pocket code is effectively "disabled". It will cause no extra baggage until viewed.
Tucker, who filed this bug, has moved this conversation to the mozilla.governance list:

https://groups.google.com/d/msg/mozilla.governance/2PYq2w8tejs/i_IindFDxxgJ
(Reporter)

Comment 7

2 years ago
Yes. Thanks Tyler; I wasn't aware of the governance list.

Comment 8

2 years ago
(In reply to Manish Goregaokar [:manishearth] from comment #3)
> If I'm not wrong, Pocket can be disabled with the customizable UI (hamburger
> menu -> customize, drag the icon out)

It can be disabled in about:config, but for example the new Reader view has a mandatory Pocket icon which doesn't go away even if you disable it in about:config.
Comment hidden (me-too)
(In reply to Jon D from comment #9)
> This needs to be reopened and acted on ASAP. This Pocket bundleware is
> unacceptable.

This bug will remain closed until something is worked out on the associated email thread. 

https://groups.google.com/forum/#!msg/mozilla.governance/2PYq2w8tejs/i_IindFDxxgJ

Updated

2 years ago
Duplicate of this bug: 1172218

Comment 12

2 years ago
I genuinely like firefox pocket only thing I want to be changed that it should ask me if I want firefox pocket to work in privet browsing mode
ATTN: Pocket removal instructions are located at the following URL:
https://support.mozilla.org/en-US/kb/disable-pocket-firefox

For anyone who comes across this bug looking for removal instructions, please refer to the link above.

Comment 14

2 years ago
(In reply to Mark Schmidt (:marksc) from comment #13)
> ATTN: Pocket removal instructions are located at the following URL:
> https://support.mozilla.org/en-US/kb/disable-pocket-firefox
> 
> For anyone who comes across this bug looking for removal instructions,
> please refer to the link above.

ATTN: At this point in time instructions above only remove Pocket button from toolbar. Everything else still remains in Firefox.

It is very different from what this bug is requesting.
(In reply to mxxcon from comment #14)
>ATTN: At this point in time instructions above only remove Pocket button from toolbar. Everything else still remains in Firefox.

1171569 was filed last week to take care of this.


(In reply to mxxcon from comment #14)
> It is very different from what this bug is requesting.

While true, I have reason to believe many people are coming across this bug while looking for the removal instructions available here: https://support.mozilla.org/en-US/kb/disable-pocket-firefox
See Also: → bug 1171569

Comment 16

2 years ago
Very disappointing that Firefox is going down the bloatware road of integrating non-core components of a browser that are well served by add-ons.

Comment 17

2 years ago
I did not ask for and do not want Pocket integration with Firefox.  
I want to completely and permanently remove Pocket from Firefox.
Please stop adding this kind of unwanted functionality unless it includes the means to completely remove it.

I want Mozilla to stop adding unwanted functionality, such as Pocket and Hello, to Firefox.
Instead, the hundreds of millions of dollars Mozilla makes from Firefox should be devoted to eliminating slow performance and crashes.  

No functionality that can be provided by a plugin should be baked into Firefox itself.  Doing so removes user choice.
Comment hidden (me-too)

Comment 19

2 years ago
I assume this "integration" helps fund Mozilla development.   

If so, I think we'd understand that this is just one of the harsh realities of life.

If Not, please allow removal of bookmark pocket.icon and other junk.
You already can.

https://support.mozilla.org/en-US/kb/learn-more-about-the-design-of-new-firefox#w_an-easy-to-discover-customization-interface

Features are lazy loaded so if you remove the icon this way the feature will have zero impact on your browser performance and it will be like it used to be.

According to some employees elsewhere, money was not involved.

Comment 21

2 years ago
That didn't remove it from the bookmark dropdown list.  (I did exit all windows and restart.)  BUT in the process I deleted so many icons, I just did a Restore Defaults and it's gone.  Restore defaults did restore hello, but not pocket & removed pocket from bookmarks.  This may really be a bug from FF's point of view.

Thanks for the tip, that didn't work, but did in the end.  I'll see if i can duplicate on another system.  I smell a science project.

Comment 22

2 years ago
(In reply to nosmo.king from comment #21)
> That didn't remove it from the bookmark dropdown list.  (I did exit all
> windows and restart.)  BUT in the process I deleted so many icons, I just
> did a Restore Defaults and it's gone.  Restore defaults did restore hello,
> but not pocket & removed pocket from bookmarks.  This may really be a bug
> from FF's point of view.
> 
> Thanks for the tip, that didn't work, but did in the end.  I'll see if i can
> duplicate on another system.  I smell a science project.

My problem appears to be caused by setting 
browser.toolbarbuttons.introduced.pocket-button;false

When I simply remove pocket from the icon bar it is also removed from the bookmark dropdown. 

If you set the toolbarbuttons like I did following some random advice on the web, you can remove pocket from the icon bar, but the bookmark icon will remain.  I set back to true and all is fine.  

(I did set browser.pocket.enabled;false, which removes it from the pool of menu options, so it won't be restored w/ defaults.)
Comment hidden (me-too)
Comment hidden (me-too)

Comment 25

2 years ago
> ATTN: Pocket removal instructions are located at the following URL:

Mark, this may be a nit-pick but cuts to the heart of the complaint people have about this default: those instructions are not *removal* instructions, they're instructions on how to *disable* the feature.
> Mark, this may be a nit-pick but cuts to the heart of the complaint people
> have about this default: those instructions are not *removal* instructions,
> they're instructions on how to *disable* the feature.

Given the way Firefox is designed, there is practically no difference. Disabling it makes it have zero overhead.

Comment 27

2 years ago
> Given the way Firefox is designed, there is practically no difference. Disabling it makes it have zero 
> overhead.

That's true, but I don't think it's the resource overhead that is causing people to complain.
> That's true, but I don't think it's the resource overhead that is causing
> people to complain.

Then this feature is no worse than search integration. An enabled-by-default feature that only sends data to a third party service if you use it.

Comment 29

2 years ago
(In reply to Manish Goregaokar [:manishearth] from comment #28)
> Then this feature is no worse than search integration. An enabled-by-default
> feature that only sends data to a third party service if you use it.

To tell the truth, despite some users just being hostile about this integration, there are some real issues with it.
For example, there are several ambiguities in Pocket(tm) ToS applicability that neither Mozilla (in the discussion thread that can be accessed by the URL in this bug description), neither Read It Later, Inc. (yet ignoring my e-mails regarding these issues to them) care to clarify. Moreover, some mozillians (who I believe aren't directly related to this integration at any level) were trolling in that thread, being aggressive or assuring us with nothing but their words that there's no problems at all (which is actually against common sense).
You can read the thread or I could summarize those ambiguities (of which I'm aware of) for you, if that'll make any difference.

Comment 30

2 years ago
> Then this feature is no worse than search integration. An enabled-by-default
> feature that only sends data to a third party service if you use it.

That's kinda ridicuolous statement considering the ToS.

Mozilla: be nice. be free (as in beer, as in speech)
> That's kinda ridicuolous statement considering the ToS.
> 
> Mozilla: be nice. be free (as in beer, as in speech)

The Pocket ToS doesn't say anything about the button, only about what happens when you use the feature. Google also has a ToS. Yahoo also has a ToS. Please point out what portion of the Pocket ToS is different from these other third party integrations.

Comment 32

2 years ago
(In reply to Manish Goregaokar [:manishearth] from comment #31)
> The Pocket ToS doesn't say anything about the button, only about what
> happens when you use the feature.

Or does it? It seems that you've never read Pocket(tm) ToS. Let me quote a part of the first paragraph of the last version of Pocket(tm) ToS (which were posted May 28, 2014):

> By installing the Pocket™ application, visiting our website or installing or using any of the Pocket Technologies, you are accepting these terms of service.

I'm referring to the part about "installing" and "any of the Pocket Technologies".
Can you prove that the button (and so-called "Pocket Marks") doesn't count as "installed Pocket Technologies"? This is an ambiguity here.
Pocket registered logo (or the code from ) can easily be a part of so-called "Pocket Technologies" (whatever that is). And that logo in pre-installed in our browsers. You can't argue with that.
> It seems that you've never read Pocket(tm) ToS

I have, please don't assume.

The button is open source code in Firefox with no licensing/copyright header other than MPL, I don't think that it can be counted as Pocket Technologies since there is no assertion of ownership there.

The logo, perhaps. That ambiguity exists; hadn't occurred to me. That should indeed be clarified by Mozilla legal. It mostly seems like a technicality that can be fixed with some better wording in the Firefox/Pocket ToSes.


IANAL, but I don't think Read It Later can force you to accept an arbitrarily powerful ToS just by wording on the page. If this was possible, I could similarly write a ToS that says "By installing Google Chrome you agree to this ToS" and then place arbitrary restrictions on the ToS. So the applicability of a auto-accepted Pocket ToS would only be to the extent that one *uses* Pocket; and the Pocket integration code in Firefox would fall under the regular Firefox ToS/Privacy Policy.

And this does seem to be the case, because the first time you try *using* the Pocket addon you get this[1] prompt, which mentions that continuing will constitute acceptance of the ToS.

FWIW I don't think anyone can force you to accept a ToS in the first place without mentioning explicitly (not on some webpage somewhere) that you're agreeing to it.

So I don't see anything malicious or out of place here. It's a bit ambiguous and could be clarified, but most of this fearmongering about "Pocket could change their ToS and become malicious" is unnecessary given that you use Firefox under the Firefox ToS and you're not prompted about the Pocket ToS until you actually click the button and go through a bunch of steps. So "common sense" as you say dictates that there's no additional danger to the user here. Ambiguity, yes. Danger, no.

 [1]: http://i.stack.imgur.com/vHkwt.png

Comment 34

2 years ago
Multiple Vulnerabilities in Pocket
https://www.gnu.gl/blog/Posts/multiple-vulnerabilities-in-pocket/

Comment 35

2 years ago
ertymail: Pretty unrelated to this bug report as any software has issues. Thanks for your understanding.

Comment 36

2 years ago
(In reply to Andre Klapper from comment #35)
> ertymail: Pretty unrelated to this bug report as any software has issues.

Not all issues are security vulnerabilities of that gravity.



Ticket 1185393 is a duplicate of this one.

Tucker, I do not see why you call this issue a bug.

Comment 37

2 years ago
Will this be fixed by bug 1215694?

Comment 38

2 years ago
Solving issue 1215694 would fix this one, since the report is a duplicate (expect for the fact that ticket 1215694 requests a specific solution).
Comment hidden (me-too)
Please see comment 2

Comment 41

a year ago
Comment 39 has been hidden. It may be flaming, but it does not merely repeat Description.

Tyler, why did you mark this as INVALID?

Comment 42

a year ago
(In reply to Filipus Klutiero from comment #41)
> Tyler, why did you mark this as INVALID?

See bug 1172126 comment 2, as also pointed out in comment 40 (right above your comment).

Comment 43

a year ago
(In reply to Andre Klapper from comment #42)
> (In reply to Filipus Klutiero from comment #41)
> > Tyler, why did you mark this as INVALID?
> 
> See bug 1172126 comment 2, as also pointed out in comment 40 (right above
> your comment).

Comment 2 does not describe any invalidity, it merely discusses the purpose of Bugzilla and requests a contribution from the reporter. Please let Tyler answer for himself.


I have requested a change narrower than that requested here in ticket 1245340.
(In reply to Filipus Klutiero from comment #43)
> Comment 2 does not describe any invalidity, it merely discusses the purpose
> of Bugzilla and requests a contribution from the reporter.

It clearly describes why the bug was closed.

> Please let Tyler answer for himself.

Please don't make passive-aggressive demands in bugs.

> I have requested a change narrower than that requested here in ticket
> 1245340.

I've closed that too. Please see comment 2 for the appropriate way to voice your concerns about this feature - bugzilla isn't it.
please use Governance or Input for further discussion about removing pocket integration.

https://groups.google.com/forum/#!forum/mozilla.governance
https://input.mozilla.org/feedback
Restrict Comments: true

Updated

a year ago
Duplicate of this bug: 1272905
You need to log in before you can comment on or make changes to this bug.