Measuring success/failure in the installation
ubuntu at kitterman.com
Tue May 17 17:07:40 UTC 2011
On Tuesday, May 17, 2011 11:37:04 AM Evan Dandrea wrote:
> On Tue, May 17, 2011 at 2:31 PM, Scott Kitterman <ubuntu at kitterman.com>
> > You knew you'd get this response eventually ...
> But of course :)
> > While I believe you are trying to solve an important problem, I don't
> > think this is the right way to go about it. I do not think a design the
> > phones home by default is appropriate for Ubuntu. I don't think that
> > opt-out via pre- seeding is acceptable even if on by default is
> > determined to be acceptable as it limits this to expert users. Privacy
> > should be for everyone.
> I think there is a large overlap between expert users and those who
> believe that the anonymous sending of a simple pass/fail response can
> be considered a privacy issue. This especially at a time when you're
> likely to be giving up more interesting information just by sending
> this email. Chances are you use GMail, Windows, OS X, Chrome, Chrome
> OS, or Firefox, all of which send data back to their creators. I
> believe that users' willingness to use these applications without
> complaint shows a general acceptance of parting with such small
> amounts of insignificant data, which again enriches their experience.
> But if we are to force an extreme privacy view on them, should we also
> modify Firefox to not check against the plugin blacklist unless the
> user first checks a box on the preferences page or clicks through a
I don't think 'everyone else does it' is reason for us to do it. Also this is
a bit different. I know that navigation data goes back to Google when I use my
Android phone not to get lost, but sending that data benefits me by improving
the experience of how navigation works on my phone (I do wish this was more
opt-in/fine grained, but that's a different story). This is engineering data
that is meant to benefit Ubuntu, not the user.
> I think the overlap between users who read and consider every bit of
> text in the installer and people who are otherwise not bothered about
> sending this data to be quite small. I especially do not want us to
> move in the direction of having every bit of measurable data that
> these users generate and send out to the Internet wrapped in a
> checkbox, as a means of forcing a particular view on them and on the
> Ubuntu project by ensuring that a sufficiently small number of people
> ever see such a feature, neutering it to such an extent that it
> becomes useless.
I don't think it needs to be particulary detailed and I do think opt-in is
very important. I can understand wanting the data to be as broad based as
possible, but I think it's important to invite participation in engineering
data collection, not assume it's OK. I'd also like to avoid the /. headlines
about "Ubuntu installs phone home - Canonical the next Microsoft".
Ubuntu was meant to bring the Linux desktop to the masses and be easy to use.
I don't think designing this kind of choice to be hard to use is consistent
with that vision.
> I also believe the small set of data we would get in such a neutering
> would largely come from Ubuntu enthusiasts, and I wonder if the data
> would skew towards a certain set of choices (maybe they often use
> advanced partitioning, for example).
Perhaps. The problem is, of course, it could skew the other way as those are
also the ones under your theory would be most likely to have privacy concerns
and opt out.
> > I would recommend instead one more checkbox in the existing setup that
> > invites the user to provide installation success information to help
> > improve Ubuntu and it should be unchecked by default.
> I disagree, for the above reasons. Putting this behind a checkbox
> that isn't checked by default is a sure-fire way to have it ignored by
> the largest of percentages.
Alternately, it may be a way to give more people a sense of participation in
the Ubuntu project that may be a gateway to more involvement in other ways.
> > This is something that is a significant change that should be reviewed
> > and approved by the Tech Board and possibly the CC.
> I believe it's the domain of the Tech Board, if it comes to that.
Probably. My recommendation is that you take it there before implementation
rather than have someone else take it there afterwards.
More information about the ubuntu-devel