[ubuntu-art] Apport icons -- merely some ideas

[Troy James Sobotka]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Donn wrote:
>> From a practical integration element -- it also must _readily_ fit
>> into the the "Human" scheme.  At the very least, it should be based
>> on the "Human" warning type symbols.  (Yet another reason we need DESIGN
>> guidelines for Human.)
> Have you a link for me? I have found the artwork links on the wiki to be 
> pretty confusing. Then again, I've been distracted.
> 

Well there aren't any ;).  This is largely due to a number of very real
issues in the design pattern.  There probably won't be from what I can tell.

What I can state with utmost clarity is that there is no 'crossover'
between the K* X* U* buntu's.  This means that the icon 'styles',
artwork semantics, etc., are all completely different across the
different platforms.  "Human" is relevant only to "Ubuntu" and it is the
brainchild of sabdfl himself -- with no formal design pattern to get
there.  Factor in the KDE / GNOME / XYZ dynamic, and you have another
layer of ugly.

Currently, if you wish to develop in the "Human" style, you will need to
use your package manager to leech the Human icon set and dowse into the
'scalable' variety.

These are all loadable by Inkscape even though they were developed in
Adobe Illustrator from what I can tell.  From there, you can pull
gradients, examine the linework, etc., and hopefully use your eye to
figure out a 'style'.  For some icons, there is a good amount of
consistency, for others, not quite... ;)

>> It doesn't stop at this however, we could very well include the vast
>> spectrum of different users in this -- power users, motor impaired,
>> etc., all require different graphical renderings and layouts.
> In terms of these "bands" of users they are like a locale within a major 
> Locale. 
> 
>> At the very least, we need to bind icons to locales and offer fallbacks
>> as per the standard icon scheming in GNOME.
> I fully agree that icons -- all artwork really -- should be locale-based.
> 

I just had a rather long discussion with Matt Zimmerman on this
(Canonical and Ubuntu's CTO for those that don't know who he is) and the
net answer was "not in Ubuntu's scope".  In other words, this becomes an
'upstream' issue -- either GNOME or KDE.  In that, the issue is complex.

> What would you say to the idea of a website that allows icons to be rated per 
> locale, per app/purpose? Those with the highest votes over time should be 
> used in preference to any one designer's "big idea".

Let's assume that we had a structure whereby we had a clearly outlined
design pattern (along the lines of
http://msdn2.microsoft.com/en-us/library/aa511258.aspx or
http://developer.apple.com/documentation/UserExperience/Conceptual/OSXHIGuidelines/index.html
), I don't know if voting would result in the optimal version of
'expressive and communicative' design.  It would very likely be a
combination of preference with design vision.  It _would_ be democratic,
but we can all imagine what a fully 'democratic' elephant might look
like ;).  Of course, totalitarianism and dictatorship doesn't work
either.  SOMEWHERE in between that polemical dichotomy is what is
required.  Unfortunately, this opens up the bikeshed...

The best course of action right now appears to be:
 1) Learn how the rest of Ubuntu works - the core technology pushers.
In particular Bzr/LP/Malone.
 2) Learn who those folks are who can help / offer opinions / etc.  Not
all ideas are as simple as they might appear to someone on the outside.
 3) In conjunction with 1 and 2, _demonstrate_ to the best of your
knowledge and ability, how something could work.  Python is your friend.
 4) Do all the boring and very 'un-artistic' things like continuing to
form documents and explanations for the new folks.  It is boring, but if
everyone requires a personal IRC message or a mail to the list to get up
to speed, it only slows progress.  I would encourage everyone to lend in
and try to streamline the wiki area as much as possible.

For the time being, this is the best course of action.

Credibility is hard to come by.  We will need it if we are to achieve
_any_ degree of control.  There are many other political factors, but
ultimately _credibility_ is our most important facet.

Sincerely,
TJS


-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.3 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFFzhL2ar0EasPEHjQRAjhYAJ9AiKYuonu5IYxsf+xp91G9cFBpRwCg0ZOO
DFciLTScj3zXyz/rxcyytj8=
=3f5u
-----END PGP SIGNATURE-----