[Wolves] Novells Gnome?

Shane M. Coughlan shane at shaneland.co.uk
Sat May 13 23:53:56 BST 2006


-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA256

David Morley wrote:
> Shane have you ever thought about politics as a career, you have
> successfully skirted around the question without answering it?  My
> query was whether you thought it was the way to go not what gnome may
> or may not do?

No need to be aggressive.

It's a difficult topic, and not one that lends itself to easy answers.
Further to this, the whole substance of my opinion could be described as
"we need more research."  That's what I said on the GNOME usability
list, and that's what I said in my post here.  It's important to note
that Novell are doing research for their market, but that this market
definition will not necessarily suit the whole GNOME user base.

Alan Horkan kicked it off the discussion in the usability list, and I
suggested that greater reference to Semiotics might be useful in
determining screen objects and their optimal placement.  Matthew Paul
Thomas asked me for some references regarding this, and I answered:

==

Kress, Gunther & Theo van Leeuwen (1998): 'Front Pages: (The Critical)
Analysis of Newspaper Layout'. In Allan Bell & Peter Garrett (Eds.):
Approaches to Media Discourse. Oxford: Blackwell, pp. 186-219

Kress and van Leeuwen engage specifically with the analysis of newspaper
front pages. They consider that the layout composition of a front page
involves three signifying systems: Information Value, Salience and
Framing. Information Value involves the concepts of Given and New, Real
and Ideal, Centre and Margin. Given is positioned on the left and
represents already known situations. New is on the right and gives us
unknown information. Real is positioned at the bottom and it presents us
with more specific information whereas Ideal is positioned at the top
and gives us a more generalised picture of the event. Centre is either
an important information element and is framed by a margin containing
other elements, or is not clear but there is a conception of it.

This work is based on Roland Barthes visual semiotics.

Barthes, Roland, "Le message photographique", in Communications,  1,
1961, pp. . Also in Barthes, Roland, L´obvie et l´obtus. Paris: Seuil
1982, pp. 9-24.

Barthes, Roland, "Rhétorique de l´image", in Communications, 4, 1964,
pp. 40-51. Also in Barthes, Roland, L´obvie et l´obtus. Paris: Seuil
1982, pp. 25-42

Of course Semiotics is much wider in scope than newspaper analysis.  A
very good place to start with regards Semiotic reading is
http://www.aber.ac.uk/media/Documents/S4B/semiotic.html

There an area of Semiotics that applies specifically to computing.
Outline available here: http://en.wikipedia.org/wiki/Computational_semiotics

>> >> ...
>> >> - From the perspective of traditional semiotics it would make sense to
>> >> have ideological data like application names and/or documentation
>> >> names at the top, but to have predictable active objects on the left
>> >> of the screen. The middle and right of the screen should contain new
>> >> information like dynamic user data (desktop?).
>> >> ...
> >
> > I double dog dare you to define "ideological data" with reference to
> > human-computer interfaces.

OK  :)

As above "Ideal is positioned at the top and gives us a more generalised
picture of the event."  Ideological can be connected with general
ideology of some sort - like a logo - or it can be understood to provide
the framing context for our interaction with the information.  When it
comes to human-computer interfaces we might understand this to be the
place where (a) identification logos can be placed, and (b) where
general information pertaining to the current information event can occur.

However, you are absolutely right with your underlying assertion that
textual and visual Semiotic analysis does not automatically translate
into multimedia analysis in the context of computing environments.  I
don't suggest that we can make such a link directly and uncritically.
However, there is the possibility of learning much from pre-existing
research regarding human perceptive methods and existing social encoding.

==

A later message continued with reference to the need to define things
more clearly before real progress can be made, especially with regards
defining the target audience.

==

As a follow-up, I believe that it could be very useful to ask Novell et
al why they have made certain decisions with regards screen layout.
Their decisions would not have been taken lightly, and their process
might enlighten us.

As Roberto said "first step we should know what was considered
problematic during its testing."  That should apply not only to the
usability tests mentioned at the start of this thread, but also with
regards the testing other friendly parties might have undertaken.

IMHO Alexandre is quite right in suggesting that we need a lot more work
on defining our samples before any judgments can be made.  There is a
real danger of too much speculation, and not enough definition.  There
is also the possibility that the HIG will have to be revised to take
into account user testing.

Alan, you mentioned before that GNOME is currently lacking user
profiles.  Our audiences are not entirely clear.  Perhaps this is really
going to be the determining issue with regards making large-scale
usability decisions.  That being said, GNOME appears to have done a
great job of providing for most users.  It's adoption is very widespread
precisely because many people find utility for it.  It's possible
therefore that real-world usage patterns might inform the creation of
user profiles.

==

There you have it.

Roberto has a clear suggestion about his personal preference for the screen:

==

I believe that's because usually menus are drop down, not push up: it's
way easier (at least for me) to push up the mouse cursor to a Fitts' Law
complying target on an upper menu and then slide down to the desired
item than doing the reverse and be forced to move the mouse up (at that
point, besides, you'll probably have to lift it from the mousepad to
leave room for the upward movement: again can't tell if this is a
generally acknowledged law, but this is what I noticed observing my and
other people's mouse usage).

Whatever the other OSes "compatibility" desired, menu layout should be
considered carefully because IMHO the present layout presents several
good points. I wonder if it would be possible to retain it while
improving its usability. Of course, as a first step we should know what
was considered problematic during its testing.

==

Fair enough.  I don't disagree with him.  I don't agree either.  It
sounds like a sensible comment with merit.  I would like to see some
research done so that we can get something solid guiding this matter.

Shane

- --
Shane Martin Coughlan
e: shane at shaneland.co.uk
m: +447773180107
w: www.shaneland.co.uk
- ---
Projects:
http://mobility.opendawn.com	http://gem.opendawn.com
http://enigmail.mozdev.org	http://www.winpt.org
- ---
Organisations:
http://www.fsfeurope.org	http://www.fsf.org
http://www.labour.org.uk	http://www.opensourceacademy.gov.uk
- ---
OpenPGP: http://www.shaneland.co.uk/personalpages/shane/files/publickey.asc
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.4-svn4127: (MingW32)

iQCVAwUBRGZjc9wG3M95JPpzAQgocgP/a8cCuCYzS13Ruy8JGr64+gsD0VHC9Pc1
A8fp0dqOXKTt1wcJWwmCGNjJ4jbbGEy2Xny7CVK3JfnLbZ4/CkUhgt7aAuGbYKiq
M9gFHhwDvnPID37MfuUnd6MU4GnmiKxzd8U/rvKNWJFOVOiLZkFKLoNbDCNn31Ay
LRyTlQ9TRbs=
=/5bk
-----END PGP SIGNATURE-----




More information about the Wolves mailing list