Re: The future of gnome-pim - and Balsa, too?

[David Pettersson <dave se linux org>]

On 13 Aug 2001 13:49:48 +0200, Toralf Lund wrote:

(Please, no flame war ;)

> and its developers seem unable to give any good
> arguments against this.

Do they need one? Evolution is to be for GNOME what Outlook is for
Windows. I am quite sure they know that. I suspect that very few
companies would even consider switching to GNU/Linux and GNOME if they
didn't know that their investment (in time and training) couldn't be
used.

Evolution is here for the corporate users and the home users that don't
want to fiddle about with Balsa and friends. We can't make hackers out
of people that aren't or don't want to be.

> This touches on another concern I have about GNOME: I think that too much time is spent on
> emulating MS Windows behaviour. Why? Aren't we all using Linux/Unix because
> Windows is crap???

No, we are using GNU/Linux, Unix and whatever other operating system we
choose because it is better suited for the task and/or superior to the
alternatives. Well, at least that is what I hope the majority of people
are thinking.

Regarding the adoption of Windows looks and functionality - well, until
recently, there hade not been any usability tests of GNOME AFAIK (Sun
did one as you might recall - and the results weren't too good IIRC).
Microsoft has surely spent millions of dollars in HCI (Human Computer
Interaction) studies to make their software usable.

But that doesn't make a difference. If average user John Doe has to
choose between Windows and GNOME with usability in concern - I'm quite
sure he'd choose Windows. Why? Because it is easier to use.

Let's compare the Windows Calculator with the GNOME calculator. First
time you use the Windows Calculator it shows a standard desktop
calculator. It has the four basic operators, memory, percentage and
square root. Just like any other off-the-self calculator you can buy in
the closest office supply store. Users will be familiar to the user
interface, and will know just how to use it _immediately_.

Now take a look at the GNOME calculator. Same operation - start it up
for the first time. What happens? The user is presented with almost
twice as many buttons, and the average users doesn't know what to do
with all of them if he just wants to do some summing up. It is
confusing, and although the calculator isn't any more difficult to use,
it appears to be and frightens the user. 

The GNOME calculator (which is a exact copy if the X calculator xcalc)
gives hints of being a scientific calculator, but it doesn't evalute
expressions as one (ie, 2 + 4 / 3 != 2). At least you can switch and get
the expected behaviour in the Windows calculator.

All of these small things add up. If we want GNOME to be successful, we
have to do something about them. I am currently working on a calculator
for GNOME that is intended to replace the default one (if people approve
of it). And it looks just like the calculator in Windows. And it isn't
because I tried to mimic it - no I grabbed the closest catalogue that
had calculators and I looked at them. And guess what, they all look like
Microsofts calculator.

GNU/Linux and GNOME already has the technical power. It is stable, bugs
get fixed fast etc. Now we have to work on usability. Fixing things like
the calculator and having applications like Evolution really help. But I
don't think we should ever let usability come before robustness etc. If
we do - I think we would be in for some real trouble.

> Having said that, I really appreciate the fact that the different types of
> functionality in Evolution are split into different reusable components.
> Maybe some of the good bits ought to be integrated into Balsa?

The design of Evolution is very nice, and is split up into components
just as you say. If just want the mail part, take the mailer component,
put it a shell (window with menus) and voila you have Balsa (more or
less).

I'll finish of with what I started - we can't make hackers from people
that aren't and don't want to be. Usability is important.

Sincerely,
-- 
David Pettersson
dave se linux org