Enrico Ros proposing a SoC project for improving the GUI

[Leo Franchi]

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

On 18 Mar 2009, at 09:01, Nikolaj Hald Nielsen wrote:

> I have some strong reservations about this. A "let's make the gui
> rock" project sound nice, but the scope of it is really staggering! I
> have done a significant amount of work on the current GUI, and as a
> quick scroll through my old blog entries will reveal, I went through 4
> or 5 different ideas for looks before finally settling on the current
> one based on a mockup by Nuno. Now, I will be the first to admit that
> the current look could be much better, but it was the best anyone
> could come up with at the tie that did have half the devs hating it
> with a passion.
>
> My point is basically that I dont believe having a single person try
> to define and implement a new look will work as a GSoC project ( or
> any other project for that matter ). There is a reason that we are
> currently setting up our lines of communication with the Oxygen
> artists to help us, not only create the actual svgs and whatnot that
> will be used in the app, but also, and perhaps more importantly, come
> up with an overall vision of how Amarok should look. We are also
> starting to work with more usability people in order to ensure that we
> don't overlook this point.

The main thing that strikes me about this idea is that it is so very  
vague at the moment. "Make amarok look awesome" does really sounds  
like a sweet project, but as we all know Amarok is a complex project  
and it is really quite an undertaking.

I strongly agree with Nikolaj on the fact that we have begun to work  
with artists and usability people and are trying to move forward  
improving the look of Amarok. I think it is important to continue our  
collaboration and develop our UI iteratively rather than  
revolutionarily. There is no doubt that it needs work, I will be the  
first to say that the CV in particular needs way more time than a few  
spare hours a week here or there from 1 developer.

Anyway, I think working on a more concrete proposal or idea, rather  
than just carte blanche to toy with amarok, would be a much better idea.

leo

(on a purely technical note, rewriting the GUI as a QGV + OpenGL +  
kinetic could be cool, but i honestly don't see it as a job for 1  
person. Besides, if we re-use current widgets you would end up with  
qgraphicsproxy widgets which absolutely suck as well.)

> Furthermore while exploratory work in this regard can get you so far,
> what we really need with regards to the overall look right now, is a
> strong sense of where to go.
>
> Even if Enrico actually manages to do all the work and create a new
> look in the time frame of GSoC, I know from bitter experience, that if
> not done right, with the participation of enough artists ( Something
> that has really been one of the main issues so far ) that it is likely
> that half the devs will hate it.
>
> So, in essence, what I am trying to say I guess, is that the process
> of defining a new vision for the look of Amarok has already started,
> albeit slowly. Because of being dependent on so many people, it is
> unlikely to make for a good GSoC project in my opinion. It is also, in
> my opinion, unlikely that this work will have progressed far enough
> that there is anything concrete to start implementing by the time GSoC
> starts.
>
> That said, it is very possible that we could define a GSoC project to
> support this process somehow.
>
> - Nikolaj
>
>
> On Wed, Mar 18, 2009 at 9:37 AM, Bart Cerneels  
> <bart.cerneels at kde.org> wrote:
>> On Wed, Mar 18, 2009 at 9:09 AM, Mark Kretschmann <kretschmann at kde.org 
>> > wrote:
>>> On Wed, Mar 18, 2009 at 1:03 AM, Enrico Ros <enrico.ros at gmail.com>  
>>> wrote:
>>>> On Tuesday 17 March 2009 20:33:43 Mark Kretschmann wrote:
>>>>> I assume that you have by now read our discussion about your ideas
>>>>> here on this list. Could you please comment on them, and decide  
>>>>> if you
>>>>> would like to make an official SoC proposal?
>>>>>
>>>>> While there still is bit time left until the deadline, it would be
>>>>> better to discuss this earlier, so that we could reach a  
>>>>> conclusion
>>>>> about this without hurrying.
>>>>
>>>> Hello Mark, hello All,
>>>>  yes, I've read the discussion and I've been waiting for all the  
>>>> pieces to
>>>> fall in place before committing to the Summer of code.
>>>>
>>>>  About the "still unnamed" Summer of Code project: what I want is  
>>>> to make
>>>> Amarok's interface look prettier and be faster essentially. Easy  
>>>> enough if you
>>>> have a magic wand.
>>>> But instead of proposing my take on the job (top-level  
>>>> qgraphicsview container
>>>> + using OpenGL where available + fx (w/shaders) + qt kinetic -  
>>>> (minus) plasma,
>>>> etc..) I'd like to know what do you guys think about a "let's  
>>>> make the gui
>>>> rock" kind of project. So this is still an "evaluating if and  
>>>> how..." kind of
>>>> mail.
>>>>
>>>> We should decide:
>>>>  - which are the criteria for judging the gui? how the current is  
>>>> rated?
>>>>  - do we feel the need to have "the best gui" we can have?
>>>>  - who should decide how to make it? (coders, designers, usabs,  
>>>> etc..)
>>>>
>>>> My personal opinions:
>>>>  When looking at today's Amarok (and I saw it made lots of  
>>>> progress in the
>>>> last months) I see a big gray area with many different type of  
>>>> gray, different
>>>> icon sizes, and the central area is corrupted [1]. What should  
>>>> itlook like?
>>>> Like the mockups floating around [2], or better: the UI should be  
>>>> like a
>>>> plugin; in this way the current gui could be kept as the default  
>>>> gui
>>>> implementation and new guis could be developed in parallel (even  
>>>> switched at
>>>> runtime).
>>>>  If the SOC happens for me I'd like to experiment with a new gui,  
>>>> like the
>>>> [2]. In this case I'll do a call to anybody wanting to contribute  
>>>> to make such
>>>> a thing, collect all the ideas, do the design in team and  
>>>> implement as much as
>>>> I can to have nearly the same functionality as the default gui.
>>>>
>>>> I'm open to all the kind of suggestions expecially criticism,  
>>>> since it helps
>>>> to shape my SOC proposal better ;-)
>>>>
>>>> Enrico
>>>>

- ---
Leo Franchi				(650) 704 3680
Tufts University 2010

lfranchi at kde.org
leonardo.franchi at tufts.edu

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.10 (Darwin)

iEYEARECAAYFAknAvIoACgkQDIxsnB9isr/o+gCfRxRTvaHpHGUkvznnV6SRNkRP
RzUAn2CaipwhVg6tkKu9wMriR+rvu7qS
=MbAN
-----END PGP SIGNATURE-----