incubator-flex-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Cortlandt Winters <>
Subject Re: [IDEAS] Flex: New user interface design
Date Sat, 25 Feb 2012 21:41:35 GMT
Nice work Jude

I very much disagree with David here, though I totally understand the
desire. It takes much too much time to do a good skin or theme.

Note that there isn't a single complete spark skin available for public
purchase yet and the spark architecture has been out for a couple of years
now. Removing the themes would essentially make it impossible to do small

I like the idea of trying to do a skin that follows the android spec and to
look for other ui specs to hit with an out of the box skin. Its a shame to
lose the linux folk because I think Linux could really use flash to provide
a decent ui to all those nice command line tools and the linux graphic
designers offer a lot to us.

One problem with the current skins is that they all are based on subtle
shade variations of a single color. Most user interfaces want to have 2 or
3 key colors in a hierarchy of emphasis that goes with a companies brand

I would very much like to design a few skins that use css to create
conceptual categories or user cases like so

Theme A: one primary saturated color with two complementary saturated
colors for occasional highlights.

Theme B: 2 primary colors with equal weight on a light or dark ground

Theme C: Strictly Minimalist, no color or graphics unless strictly

Each theme would have its css blocks defined with a level of conceptual
rules for content such as Content area, light text on dark background, or
dark text on light background, so that one doesn't run into the problem
where you end up with light text on a light background or dark text on a
dark background. Too many skins/themes don't pay attention to that.

The goal is that a developer could fill out a half a dozen properties and
create an nice looking skin that follows a companies existing brand
guidelines or a native platforms look and feel.

I'm not trying to put design agencies out of business or anything, just to
cover the 2 or 3 most common cases so that teams of 1 or 2 developers can
be maximally productive if they don't have a design budget.

I know that our flex user group (in Albany ny) had a lot of users that were
individual developers working on a project by themselves with no graphics
budget and I believe that the spark architecture hit this class of
developer very hard.

I myself was hit very hard on a project a year or so ago when a client was
disturbed that the spark scrollbar looked exactly opposite from the way
that he thought scrollbars always looked. He was a lifetime windows user
and noted that it was very disturbing that the track was dark and the thumb
was light. Used to the flexibility of the mx components I casually
mentioned that it was no problem and that I would get it to look exactly
like he thought it should look only to find myself doing about 6 hours of
free work because we didn't have any budget for it to override a half a
dozen classes. Now that I've done a lot more spark skining it wouldn't take
so long, probably half of the time was spent trying to figure out what I
was missing, but it was a non-optimal experience.

Also sometimes it's not lazyness, many programmers are simply hopeless when
it comes to graphic design or user interface design, by offering some solid
themes for these folk we might see a bunch of "flex looking" apps, but they
would at least be good looking flex looking apps and not a hopeless shamble
of clashing colors and textures with no consistency and breaking basic user
interface rules.

It's a lot of work to create skins, but it's kind of fun work and just the
sort of thing that we should be able to get folk to help with. I know that
I can commit to doing at least one or two, though not for a few months.

One thing that would help the development of skins would be a simple
"kitchen sink" app, that uses each of the components in it's typical
configuration and has a bunch of canned content to test light and dark
sections. Like with the css zen garden. The as fusion folk, like scale
nine, had a contest to fill out the kitchen sink app with designs and got
some good results. Without a kitchen sink app many of the good graphic
designers would get bogged down with implementation details.

Anyway those are my ideas, thanks for listening.

On Sat, Feb 25, 2012 at 2:52 PM, Nicholas Kwiatkowski <>wrote:

> My thoughts are -- it depends on the application.
> The existing themes are great if you are doing a lot of forms based
> 'screens'.  I don't have a designer on my team, and I most likely never
> will get one.  My apps are used by professions that like good usability and
> a good clean layout -- that is one thing some of the existing themes give.
> Separating them out and making them easily accessible is not a bad idea,
> but removing them -- I'd vote against hat.
> -Nick
> On Sat, Feb 25, 2012 at 1:11 PM, David Francis Buhler <
> > wrote:
> > My own personal preference would be to remove the current themes (both MX
> > and Spark) from the SDK. This includes the Cobalt theme, Zen theme, etc.
> I
> > would stick with the "Wireframe" theme for Spark controls. In doing so,
> we
> > remove the obvious visual impression of a "Flex Application", and
> > encourage the use of Flex/AIR apps that look like part of their native
> > environment (FaceBook, Android, iOS, Windows 8, etc.). Most developers
> take
> > short-cuts and use one of the existing themes when building a product for
> > their client, and unintentionally give the impression of a hodgepodge of
> > technologies that prevent the impression of product cohesion. Removing
> the
> > Cobalt theme, Zen theme, and other themes would discourage this practice
> of
> > use-what-i-found.
> >
> > If companies have designers, they're better off with a tool like Martin
> > suggests then they are with existing Themes. Moreover, the existing
> themes
> > confuse designers (with the MX and Spark namespaces, the inability to
> > understand each and every style property, or the overwhelming number of
> > properties available). If companies don't have designers, they're better
> > off sticking with Wireframe theme until they do.
> >
> > Incidentally, I'd love to see a tool that generates themes from a
> > user-defined base color, with the palette generation of complimentary,
> > monochromatic or triad colors, similar to Kuler.
> >
> > [1]
> >
> > On Sat, Feb 25, 2012 at 12:20 PM, Martin Heidegger <
> > >wrote:
> >
> > > On 23/02/2012 17:59, Haykel BEN JEMIA wrote:
> > >
> > >> I think the first step should be to create new skins for the current
> > Spark
> > >> components. Designers could make designs for them (ideally using a
> tool
> > >> that enables export to FXG like Illustrator) and then developers can
> > >> create
> > >> skins out of them.
> > >>
> > >> Haykel
> > >>
> > >
> > > Doesn't necessarily be a design for flex: Here i found some nice
> > > inspiration[1]. A *fictional*
> > > Windows 8 user interface - very nice :)
> > >
> > > yours
> > > Martin.
> > >
> > > [1]**2/24/2822891/windows-desktop-**
> > > ui-concept<
> >>
> > >
> >

  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message