Re: [Usability] Special Open Source Issue of Interactions Magazine

[Alan Horkan <horkana maths tcd ie>]

On Tue, 16 Jan 2007, Lennart Borgman wrote:

> Date: Tue, 16 Jan 2007 03:32:00 +0100
> From: Lennart Borgman <lennart borgman 073 student lu se>
> To: j s arnowitz acm org
> Cc: usability gnome org
> Subject: Re: [Usability] Special Open Source Issue of Interactions
>     Magazine
>
> Hi Jonathan,
>
> I am not going to contribute because I am not an expert in this area.

When did anyone in open source* ever let that stop them?  ;)

You can learn a lot from a few good books.

One of the biggest tricks in open source development is breaking down the
problem into smaller pieces.  The Human Interface Guidelines have been
essential in allowing users to express to developers how much they really
do care about usability, and consistency in particular. The buzzword here
might be heuristic analysis, because with a few general rules you can get
a lot done.

The bigger holistic analysis requires a much deeper understanding of a
project and the tasks being solved and although these problems cannot be
so easily broken up and tackled by many people they can be tackled by a
few serious designers with enough time.  (Also paper prototyping or making
mockups using RAD tools and small surveys can go a long way, huge
resources not required.)

> But could you please also look at accessibility at the same time.
> Accessibility is difficult and it seems to me that it often get
> overriden by some personal usability style. This is very unfortunate.

I've found that good usability means you end up learning a lot about
accessibility, and a bit of everything else actually.  Understand the
constraints and be able to justify the trade-offs and then you can at
least say the usability is good for your target audience.  We really are
aiming for mass market, the best answer for the most people possible so
the last thing we would want is to intentionally exlcude users.

Understanding accessibility (a11y) helps you justify a lot decisions and
things to avoid like the careful use of colour and texture.  Working on
documentation teaches you the importance writing style, and
interenationalisation teaches you more about your own language.  Both of
these are reasons why avoiding acronyms is something I always make a point
about (I should explained above that RAD stands for Rapid Application
Development, and URL is never as clear as saying link or location or
address or whatever is most appropriate for the situation and actually
translatable.  Also the space constraints of translatable strings mean
that trying to be excessively terse in English is akin to a premature
optimisation.)

If you care about accessibility it often helps if you can turn it back
around into a techincal discussion which developers will be more
interested in hearing.  If you can make your system work for a pointer you
can make it work well for pen driven devices too (so many portable
devices, but multimedia and artists too).
Dont talk to me about right click.

> Microsoft has been a proponent for accessibility and sometimes they have
> listen well and done it good. Now I am sitting here with XP (I need it,
> unfortunately and my latest test of GNU/Linux led to a reinstall of
> everything - those small little details...). What I see is that
> Microsoft has recently done the same mistake. The latest Windows Media
> Player is more good looking than accessible. Bah! Why do they hire that
> kind of kids that can't keep enough details in their head to do both
> things - both a nice and accessible GUI.

The aren't the only ones who dont use the standard widgets, witness the
frequent reinvention of skinning and themes.  Microsoft ends up competing
against itself with the Office team coming up with whole new widgets every
major release and making things inconsistent again.  Apple try to keep
things together but some of their products jump ahead with new and unusual
widget stylings.

> Well, that is the story in my opinion. It seems like when things gets
> complicated then accessibility is forgotten. It simply makes it simpler.
>
> That is my belief of course. That the developers or perhaps rather the
> managers are not good enough. It could be worse. It could be that they
> do not care. It could be that the time lines are so short that they do
> not care about accessibility. And that is worse!
>
> Antoher of my favorites when it comes to usability is my frustration
> about the continous wheel inventing. One time when I installed GNU/Linux
> everything went very, very fine -- until I should login and start using
> it. I had no mouse on that system. I tried Ctrl-Esc, the Window Keys,
> Ctrl-Alt-Del and everything else I learned from my MS Windows
> experience. Nothing of it worked.

Funny you should mention that because if I recall correctly one of the
stated reasons for not copying those keybindings is *accessibility* or
more specifically predictability. [1]

Gnome uses
Alt+F1 for the main menu
Alt+F2 for the run dialog
...

There was a discussion recently about how we might make it easier for
users who wanted to quickly setup those keybindings and use the Super key
(aka Windows key) in the ways they already learnt but as far as I know no
one is following up on that.

(Incidentally Ctrl+Esc used to work way back when there was just one panel
and the Gnome main menu was in the bottom left corner.)

> to let the user use their experience. Even if they have experiences from
> the big EVIL MS. Do not bannish the users for their experience. That is
> just not the way forward!

Given the majority of existing users are familiar with Microsoft
technologies I certainly favour an embrace and extend approach but that is
easier siad than done.  Ideally though it is projects like OLPC targetting
the even bigger audience of totally new users which have the potential to
make revolutionary changes.

-- 
Alan

* Open Source and Freedom Software et cetera ...

[1] vaguely relevant links
http://mail.gnome.org/archives/usability/2006-November/msg00013.html
http://mail.gnome.org/archives/usability/2006-October/msg00009.html
http://www.gnome.org/learn/access-guide/latest/keynav-1.html
http://mail.gnome.org/archives/usability/2006-August/msg00023.html