Re: The target user and consequences

[Gleef <gleef capital net>]

On Fri, 7 Aug 1998, Stephan Pfab wrote:
> The target user of GNOME will be neither
>   a Windows user, a Mac user, an Amiga user, me (sadly),
>   Miguel, a coder ...

Agreed.  Where target users come from is less important than where they
are going to.  Our target users are people who want a consistant, easy to
use, powerful desktop environment.

In delivering this, we should also be able to keep GNOME Free and highly
customizable, as per the GNOME Manifesto.  Striking a good balance between
all these goals is, I feel, achievable.


> GNOME will be (more or less) different from every other GUI.
> That means that everybody will have to learn to use GNOME efficiently.

It can be hoped that making GNOME "easy to use" will encourage it to be
"easy to learn".  I know they don't necessarily follow, and if
forced to choose between the two, I would much rather GNOME be "easy to
use", since the users will only have to learn it once, but have to use it
every day.


> Accepting that GNOME is different we should make it clear to the user 
> too. If we follow any other GUI too closely he will
> not tolerate the differences. The first impression must
> indicate clearly that this application follows the GNOME model
> and not any other.

To enhance this point, I certainly feel after a Style Guide is in good
shape, someone should work up a GNOME User Guide.  This would tell a user
how to navigate the interface described in the Style Guide. Remember, the
Style Guide is for developers to read, in order to better support the
users, we will still need a guide for the Users.


> After we create this expecation we must follow through.
> Whatever we decide to be top level GNOME behavior we must
> enforce it. As many people said consitency is the key.
> Consistency with ourself not with any other GUIs or
> expectations.

Agreed.


> The consequences:
> 
>   We should not follow the Windows look or feel or menu order ...
>   (substitute Amiga/Next/Mac)
>   if there is no good reason to do so.

Agreed.


>   We need visual clues to brand applications as GNOME applications.
>   (I love the foot).
>   Some phantasic GUI elements to distinguish us from the rest,
>   would be great. (Not everything new, but something)
>   (colorreaction, ...)

Visual clues to brand GNOME apps would be good, but not necessary.  The
foot does go a long way towards this, but it can introduce logistical
problems that could hurt the totality of the GNOME look and feel (eg.
scaling the foot to match the font).  

It should seriously be considered, however.  I would like to see the
GNOMEprint work.


>   We need to enforce inner consistency.

Agreed.


>   (this will never allow emacs to be a real GNOME application :-( )

I fail to see how this follows.  As long as the inner consistency we
enforce does not directly interfere with Emacs, there should be no
problem.  Keep in mind that Emacs is one of the most customizable programs
ever written.  It should be possible to work a GNOME mode into Emacs.


>   We need to think very hard about what is required from a GNOME app.
>   The defaults we choose will be used 90% of the time, regardless
>   of themabilty (see emacs). But complete restriction will throw
>   people off (see KDE).

Agreed.  However, I'm not clear what point you're trying to make regarding
KDE (I'm not as familiar with KDE as I should be).


>   The defaults should give room to expansion, (see emacs key binding).

Strongly agreed.


>   (binding "alt" to menu open seams like a waste of an modifier).

I think the keyboard binding discussion is far from over.  I haven't given
any of my proposals regarding this yet, since it is likely to generate
more volume than the File menu debate did.


>   We need to come up with this standard pretty fast.

Not really, there's a current Style Guide, highly visible on the GNOME web
site.  If developers would actually follow it, they won't be too far
astray from what we eventually come up with.


>   (The coders will hate the GUI proposel people
>    if they have to change their programs dramatically.)

They will gripe, then they will modify their programs, then everyone will
be happy.  I think most GNOME developers realize that good Style Guide
compliance is a Good Think (TM).


>   We have to accept compromises.
>   (I will sacrifice emacs, the program I use most of the time.
>    What do you sacrifice ? Menu entries ? ...)

Of course compromise is absolutely necessary.  However, with a little
effort you might be surprised how accomodating something like this can be.

 
> lets propose not flame

Strongly agreed.

> 
> Stephan
> 
> -- 
> Will verify for food

:-)


-Gleef