Comment 463 for bug 269656

On Friday 19 September 2008 8:39:41 pm kafpauzo wrote:
> @ Chip Bennet:
> However I don't agree with you that "non-free service" and "freedom" are
> suitable terms for services. I think "free" causes confusion rather than
> clarity. It sounds like you mean "free as in the GPL", to which the
> necessary reply must be "the GPL doesn't apply". We do need a term, but
> "free" is too confusing.

Perhaps it is too early to get caught up in terminology. I think that's part
of the problem: the environment is so new that we don't even know the proper
terminology.

That said, I think there's a big difference between "if you don't agree to the
use terms, just don't use" (passive end-user action) and "if you don't agree
to the use terms, disable" (active end-user action) - especially considering
that, in this example (Firefox anti-phishing services), most end-users will
not know that their assent is required to use the services. (To that end, I
understand why Mozilla wanted to expose the end user to the use terms.)

The real problem arises when something that squarely falls into our
understanding of "free software" (Firefox web browser) gets tangled up with
something that still does not (anti-phishing services).

 Free software by definition imposes no use restrictions on end users. Firefox
sans anti-phishing services imposes no use restrictions on end users. Firefox
*with* anti-phishing services *does* impose use restrictions on end users.

Therein lies the problem. And whether the services are called free/non-free,
or something else entirely, the end result is that Firefox itself is in a
state that really can't be considered to be fully consistent with the
definition of free software (at least, as far as I can thus reason).

> "Enabling those services is - at most - as simple as checking a
> configuration check box."
>
> People tend to interpret the defaults as a very strong recommendation.

I'm not sure I agree with that assertion. *Many* defaults are matters of
personal preference.

Perhaps it is true to say that most people do not change configuration
defaults for a given application, whether due to ignorance or apathy;
however, I'm not sure that most people would see those configuration defaults
as strongly recommended, "best practice", etc.

> When people are uncertain about the consequences of touching a setting,
> many will see the default as a recommendation that you should disobey
> only if you have a really compelling reason, and only if you have
> thought through all the consequences with great care.

That is exactly why Ubuntu could pop up a notice of its own (similar to the
notice that comes up when installing/enabling MP3 codecs) on a user's first
run of Firefox, explaining that the anti-phishing services are disabled by
default, why they are disabled, the benefits the services provide to the
user, and how to enable them (perhaps even offering to do so with a simple
button-click).

> As a consequence, the people who need this service will generally be
> afraid to touch the setting, and will only very rarely turn it on.
> Meanwhile those who don't need the service can turn it off very easily,
> for them it's quite trivial.

This stance appears to be the one shared by our SABDFL. I cannot disagree -
unless some means exists to educate those users (as in the above example).

Regardless, the bottom line really is this: with the anti-phishing services
enabled, and therefore the restrictions with which those services encumber
the end user, can Firefox still be considered to meet the definition of free
software?

If that answer is yes, then there is no issue.

If, however, the answer is no, then *regardless* of how beneficial those
services are to end users, Firefox cannot and should not be allowed to stay
in the Main repository.

If the answer is no, then the disposition of Firefox really becomes one of
principle: will Canonical be willing either to violate or to modify its
requirements for software in the Main repository for the sake of Firefox (and
its anti-phishing services)?

If the answer is no, then for many in the Ubuntu community, the disposition of
Firefox will be indicative of Canonical's willingness to demonstrate the
courage of their convictions with respect to free software.

(And if anyone should think that statement to be sensational, I would refer
you to the degree of emotion inherent in many of the preceding comments.)