[Usability] Re: [Desktop_architects] Printing dialog and GNOME
- From: Christopher Blizzard <blizzard redhat com>
- To: Linus Torvalds <torvalds osdl org>
- Cc: usability gnome org, Jeff Waugh <jdub perkypants org>, desktop_architects lists osdl org
- Subject: [Usability] Re: [Desktop_architects] Printing dialog and GNOME
- Date: Tue, 13 Dec 2005 01:45:12 -0500
I'll jump in here, too, because I've done a lot of thinking about this
lately, trying to make it possible to explain to people where projects
like GNOME and the Firefox project have gone. I suspect that in your
mind that they have just "gone crazy" but I promise there is a method to
the madness that you see.
First, I want to throw out the word "Usability." I hate it. Hate hate
hate. Because it describes something that doesn't exist in the real
world. Why? Because as you have discovered it's entirely subjective.
That is, it means different things to different people. For me it means
something that doesn't get in my way, is visually elegant and easy to
teach others to use. For you it means something that you can configure
to your specific needs, probably based on how you've done things in the
past and makes you the most productive. Two different people, two
different targets and for one it's "usable" and for the other one it's a
shitpile. That's why I think that we need to chuck that word right out
Instead, let's use the word "design" because it describes not the end
result - what you and I feel about the experience - but instead it
describes the process that we use to make decisions. During that
process a lot of things happen. These include choosing which users we
want to target, how an interface should be presented to a user based on
who they are, what their experience should be, what a system is capable
of doing and other things. Each of that adds up to software that's a
reflection of a target audience. But you don't actually give a shit
about this, because you're focused on one particular fact: people
removed crap that was core to the way that you got work done.
And this brings us to one of the fundamental tenants of design: that you
have to make tradeoffs based on the users who you are targeting. Havoc
eluded in other email to this, but largely in the context of a strategy
for the desktop. For people like you this means that sometimes stuff
gets removed that you care about. But for someone like me, who cares
about getting the default experience right for a large base of users,
this is a tradeoff that I'm willing to make because it increases the
pool of available users. Even if it means that someone like yourself
can't stand to use it without making changes to the way you work.
Which is the lead up to the next statement logical question: if not you,
then who are we targeting? I think the answer there is reasonably well
understood in the GNOME design communities but misunderstood outside of
that core group. At this point we're interested in corporate users
(office, productivity, mobile users), fixed function users (people who
do only one or two things) and some subset of hackers. But I guess not
hackers who want to configure everything themselves. This leads up to a
few design rules that I see at work in GNOME and Firefox:
1. Shit should just work.
One of the best demos I've seen so far was when we managed to get to the
point where if you had a desktop application up and running with the
print dialog _already open_ and you plugged a printer into the back of
the computer, it just appeared in the dialog. No confusing
configuration dialogs, no searching the interface for the right model
type, it just freaking showed up. That took an incredible amount of
work to get right and we could have left 70% of it up to the user to
figure out, but we wanted to do better than that. I've got better
things to do with my time (like replying to angry email from kernel
hackers, hi!) than to mess with printer settings.
The vision of this kind of experience is what drives NetworkManager as
well, but we're not quite there yet.
2. Shit should be easy to figure the fuck out.
This is the hardest one to solve because "easy to figure out" means
different things to different people. Hence, death to usability, hello
design! This means finding that delicate balance between how people
expect things to work (why doesn't middle click maximize my window?) and
how they might learn something new (hence the "Computer" icon on my
desktop that lets me find the local network and filesystem.) This is
the hardest part of design, really, requiring some kind of balancing act
between good guesses, solid research and an eye to history.
3. This club is not all-inclusive.
Yep, someone is going to get pissed off if you design something and make
a decision about how something is going to work. This isn't easy and is
the cause for email like this, but if we're not pissing someone off
we're not making someone else happy.
So let's bring this back to your original assertion about GNOME and KDE
and probably the point of this email: we don't think that people are
idiots, we just think that in 90% of the cases people have better things
to do than learn the low level details of the desktop and how to
configure confusing key combinations. Software should "just work" for
our users, express clearly what's going on on the system, and it should
be accessible to a wide audience. This doesn't mean that GNOME sucks
and KDE rules - it just means that it's not right for _you_. And for
the record I feel the same way about KDE. I think that it's fine, but
just not for me. And lucky for you, there are still a lot of desktop
options out there that fit your bill.
So I hope that this email gives you a little better perspective on where
we're coming from. We're not treating people like idiots, we're just
trying to make an operating system that tries to make computers not get
in the way but instead enable a large number of people to get useful
On Mon, 2005-12-12 at 19:35 -0800, Linus Torvalds wrote:
> On Tue, 13 Dec 2005, Jeff Waugh wrote:
> > That's definitely not a point of view of the GNOME Project - we're focused
> > on making Free Software appropriate for users who are smart (we don't talk
> > about 'dumb users'), but just don't care about computing technology. We're
> > just like every other Free Software project - fixing stuff requires the work
> > and attention of people who care about the problem at hand.
> No. I've talked to people, and often your "fixes" are actually removing
> capabilities that you had, because they were "too confusing to the user".
> That's _not_ like any other open source project I know about. Gnome seems
> to be developed by interface nazis, where consistently the excuse for not
> doign something is not "it's too complicated to do", but "it would confuse
> The current example of "intentionally not listed in the printing dialog,
> the usability team of GNOME was against listing these options." is clearly
> not the exception, but the rule.
> Jeff, if the explanation had been "exposing PPD features is too hard, we
> need developer manpower", I'd have understood. THAT is what open source
> projects tend to say. Not "powerful interfaces might confuse users and not
> look nice".
> If this was a one-off, I'd buy it. But I've heard it too damn many times.
> And only ever from Gnome.
> The reason I don't use Gnome: every single other window manager I know of
> is very powerfully extensible, where you can switch actions to different
> mouse buttons. Guess which one is not, because it might confuse the poor
> users? Here's a hint: it's not the small and fast one.
> And when I tell people that, they tend to nod, and have some story of
> their own why they had a feature they used to use, but it was removed
> because it might have been confusing.
> Same with the file dialog. Apparently it's too "confusing" to let users
> just type the filename. So gnome forces you to do the icon selection
> thing, never mind that it's a million times slower.
> Desktop_architects mailing list
> Desktop_architects lists osdl org
] [Thread Prev