Re: The Spirit of Free Software, or The Reality

[Ian Jackson <ijackson@chiark.greenend.org.uk>]

Ben Finney writes ("Re: The Spirit of Free Software, or The Reality"):
> So the above seems to argue either that search engine icons are
> sufficiently important that we can violate the Social Contract, or I've
> misunderstood. I'd like to know exactly where that misunderstanding is.

You are arguing from the Social Contract.  This is the Debian
equivalent of godwinating the conversation.  But I will try anyway.


The point of having ethical principles is to do good in the world.

We can disagree about what good is, of course.  But our users are not
harmed, and their freedom is not diminished, if we ship nonmodifable
icons for proprietary search services.[1]  There is no significant risk
that anyone would think that these icons are modifiable.

As I wrote before, in this case, pickiness about the modifiability of
the icons /is/ harming our users (not very much, but still).

You haven't come up with a counterargument to these points, which I
made in an earlier mail.


I have also made the point that we make an exception for licence
texts.  Obviously the situations aren't entirely parallel, but this
demonstrates that the absolutist position you are arguing for is both
contrary to our existing practice, and impractical.  If you are saying
that this principle of modifiability is entirely absolute and we have
to make no exceptions at all at all at all, you have to address that
point too.

If we are prepared to make exceptions, no matter how narrow, then
the question is: on what basis might we make an exception, and should
we make one in this case ?

I am happy that we should use our documented principles and aims to
guide our actions, but if applying the letter of the law undermines
our values, we should go with what is right rather than what is
written down.


One problem is that the principle that we should protect our users'
privacy isn't written down in our foundation documents, even though
it's clear that most of us (probably, an overwhelming majority) think
it important.  If it _were_ written down then it would be more obvious
that there is a conflict between different principles here.

As someone who has come to think that reference to foundation
documents to illuminate these kind of problems is not normally
helpful, I'm not particularly bothered that the foundation documents
lack a commitment to our users' privacy.

But if this bothers you then I would support a GR to improve this.  If
you are going to clean this up then you should probably also deal with
the fact that they also lack a commitment to our users' security, and
you should consider whether it would be useful for these documents to
use words and phrases like `autonomy' and `in practice'.


I'd like to thank Mike Hommey again for all his hard work and his
toleration for this kind of conversation.  I support his intentions as
he has just laid out.

Ian.


[1] To be clear, I mean that the users' freedoms are not diminished,
nor the users harmed, by the nonmodifiability of the icons.  An
argument could be made that the very presence of these search engine
configurations is a problem, but if that is the case it doesn't depend
very much on what icon is shown.  The obvious counterargument is that
respecting the user's autonomy - including not putting barriers in
front of their choice to use a proprietary service - is part of
upholding the user's freedom.