On Wednesday, June 17, 2015 at 11:52:37 PM UTC-5, Ryan Kelly wrote:
> On 18/06/2015 13:56, B Galliart wrote:
> > You claim Pocket(TM) being integrated into the core of Firefox to be 
> > pre-installed is justified because "the master goal requires Mozilla to 
> > attract and retain users."  As such, Firefox is by extension a Pocket(TM) 
> > application.
> 
> Asserting this as fact does not make it so.

The only place in the Terms of Service which seems to define the term 
"Pocket(TM) application" is here:

"The pocket software application, supporting files and accompanying 
documentation (referred to collectively herein as the 'Pocket application') is 
provided solely for your personal, noncommercial use."

My understanding is pktApi.js is a supporting file provided by Read It Later, 
Inc.  Based on the wording, the Terms of Service is claiming Firefox now fits 
what it defines to be a "Pocket(TM) application" and as such the user is 
requested to read and agree to Pocket(TM)'s ToS and Privacy Policy "before 
[they] install."

> "Firefox is now a pocket application" is IMHO a completely unreasonable
> conclusion to jump to, unless one is deliberately looking for gotchas
> with which to hate on this integration.

I am completely open to an employee of Pocket(TM) providing clarification on 
the definition and have even requested it from them.  So far, I have heard 
nothing back.

Feel free to explain in your own words what exactly the Pocket(TM) Terms of 
Service is defining and how it applies now that Pocket(TM) is a non-optional 
install part of the core of Firefox.

It is true that I am deliberately looking for gotchas.  I believe the Mozilla 
Foundation's campaign for The Web We Want put obligations on us to look for 
gotchas.  If I am misunderstanding the Terms of Service in a way that you can 
have a lawyer explain on this mailing list, then great.  If this is an out of 
date Terms of Service that includes language that will no longer apply, then 
that is also great.  But until then, I believe the "gotchas" are still there.

Another gotchas in that definition is the restriction for personal and 
noncommercial use.  I believe this is the first time Firefox has directly 
integrated a feature with such terms.  Even if a company's own data regulation 
policies doesn't conflict with Pocket(TM)'s Privacy Policy, that may still be 
in violation of the ToS by attempting to roll out Pocket(TM) use to all 
employees.  This is yet another reason why allowing for a pocket-compatible 
privately run server is needed and why it is problematic that a 
private/undocumented "/v3/firefox/*" API namespace exists.

This is not out of hate.  The gotchas appear regardless of if you love or hate 
the integration.  It is just the nature of how the ToS is currently written.

> As Gerv already pointed out earlier in the thread, all of the pocket
> code shipping with firefox is open-source and is clearly licensed for
> use without agreeing to any terms-of-service.  It's declared right here
> at the top of the source file:
> 
> 
> https://hg.mozilla.org/mozilla-central/file/a3f280b6f8d5/browser/components/pocket/pktApi.js

Yes, Gervase Markham did say the above.  Later he also said the following:

"Right. I take back this implied criticism of your position. I'm seeking 
clarification on this. I'm pretty sure the sections you mention don't 
apply (and are not intended to apply) to the Pocket code in Firefox, but 
I entirely agree it's very unclear."

"I will come back to this group when I hear more."

I asked Pocket(TM) for clarification back on June 6th and I am assuming he also 
asked around June 10th.  If there is a misunderstanding, I believe Pocket(TM) 
could easily clear it up by responding to the pending requests.  So far, I have 
not heard anything and over a week has passed.
 
> > According to Pocket(TM)'s Terms of Service, merely by installing a 
> > Pocket(TM) application the user should have both read and agreed to the 
> > Pocket(TM) Terms of Service and Privacy Policy.  According to Pocket(TM) 
> > Privacy Policy, the users are a data point to be sold in the form of 
> > aggregated information.
>
> I'm not a lawyer, but I've shipped code under the watchful eye of
> Mozilla's laywers, and ToS acceptance is something they take very
> seriously.  My team recently had to remove a feature from an (unrelated)
> project because we were not appropriately surfacing agreement to the
> terms of a third-party service.

That is good to know.  Is there anyplace as part of Mozilla's effort to be 
transparent where Mozilla's lawyers provide the details or synopsis of the 
Terms of Service documents the reviewed?

Also, how does Mozilla lawyers take into account the impact of a new currently 
unpublished Privacy Policy that goes into effect 30 days after publication?  
Are they continually monitoring and reviewing the documents as they are 
updated?  Are they able to always accomplish a full review in less than 30 
days?  What is Mozilla's stated policy/method for giving notice to the impacted 
users when a change no longer adheres to Mozilla's mission?
 
> I am in broad agreement that there's a lot about this integration that
> could have been communicated better, coordinated better, and generally
> made more transparent.

I glad we are both in agreement on this.  I wish you could see that this is not 
a result of hate of integration of feature but instead out of respect for 
Mozilla's previously stated mission.

> But to suggest that this integration suddenly springs unrelated and
> unsurfaced ToS on our unsuspecting users seems like a fanciful overreach
> to me.

Not only do I claim that this integration based on the wording of Pocket(TM)'s 
ToS springs itself on unsuspecting users, it also springs itself on 
unsuspecting developers as well.

Putting pktApi.js under an open source license is a good first step but is not 
a reasonable replacement for a statement of intent regarding treating the API 
and service as open.  Currently, the closest to a statement of intent is the 
Terms of Service which *revoke* rights instead of giving them.  There are at 
least two major problems with this in terms of how the USA legal system works:

(1) The court finding in Oracle vs Google show it is possible to put a specific 
implementation of an API under open source while still treating the API itself 
as being under additional restrictions.  The fact "/v3/firefox/*" API namespace 
is undocumented and stated to be private indicates that attempting to create a 
pocket-protocol compatible server may still result in legal action for 
violating Pocket(TM)'s intellectual property.  So far, only an implementation 
of the client has been put under the open source.  The API is not stated to be 
an open standard for free use.

(2) The lack of being able to create a private or test API server service bring 
us to another of USA worst laws, the Computer Fraud and Abuse Act.  Currently, 
the only Pocket(TM) API service available is the *production* one provided 
under Pocket(TM)'s Terms of Service.  If a developer does anything not 
explicitly permitted by the Terms of Service and even unintentionally 
disrupting service, the developer can be charged criminally.  Also, under the 
CFAA, it is up to District Attorney, not up to the company (in this case 
Pocket(TM)) on if charges are to be filed.  The fact that the CFAA takes the 
concept of a ToS being related to civil damages and make it a criminal offense 
make it that much more problematic when companies use the strictest restrictive 
language possible.  In the case of the Pocket(TM) "LICENSE RESTRICTIONS" 
section, the language is extremely damning.  It leaves no room for error in 
modifying the code which uses the service.

I still understand that you probably still believe I am saying this out of same 
hate of Pocket(TM).  I doubt I can change your mind about that.  I don't blame 
Pocket(TM) for the actions of Oracle and I don't blame Pocket(TM) for the CFAA. 
 But it is still part of the world Americans live in.  If Pocket(TM) is part of 
a transparent "Web We Want" then is it really an act of "hate" to expect better 
clarification?!

There has to be some common ground that can be reached where looking for 
gotchas is acceptable instead of "hate."  Everything about the Web We Want open 
letter was statements of common web "gotchas."  Was Mozilla operating on the 
basis of hate then?
_______________________________________________
governance mailing list
[email protected]
https://lists.mozilla.org/listinfo/governance

Reply via email to