Good Rant on lack of Good GUI Design Software

23 Jan 2008 - 11:22am
6 years ago
21 replies
1346 reads
Jim Jeffers
2007

Just thought I'd share:

http://www.jasonsantamaria.com/archive/2008/01/23/mucking_up_the_fireworks.php

Jason Santa Maria posted an excellent rant on his blog about the lack
of a solid design application for creating GUIs. Most of us jump
through hoops to accomplish our designs in traditional art programs
such as Photoshop or Illustrator. Does anyone here share similar
frustrations or have any lesser known solutions they use as an
alternative?

---------------------------------------
Jim Jeffers
"A trustworthy individual."
http://donttrustthisguy.com

Comments

23 Jan 2008 - 1:34pm
Andrei Herasimchuk
2004

On Jan 23, 2008, at 9:22 AM, Jim Jeffers wrote:

> Jason Santa Maria posted an excellent rant on his blog about the lack
> of a solid design application for creating GUIs. Most of us jump
> through hoops to accomplish our designs in traditional art programs
> such as Photoshop or Illustrator. Does anyone here share similar
> frustrations or have any lesser known solutions they use as an
> alternative?

I'm always frustrated, in case that wasn't obvious. 8^)

There were many times at Adobe that I attempted to try and get
something like this started. But the answer from executives was
always the same: There isn't enough market opportunity to justify the
product. Meaning, if you could only find 20,000 people who needed the
product, and sold it for $500, your gross would be $10M. That's
nothing for software products these days. It needs to be in the $50M
to $100M range.

That may no longer be true. Let's hope that is the case. Are there a
100,000 designers out there that define themselves as interface
designers or interaction+visual designers or digital designers that
would pay $500 for such a product? Again, let's hope so. Otherwise,
getting it to happen is going to be a long uphill battle.

Also, I would frame the product's main direction and market
opportunity as a "screen design" application that requires interface
and interaction features. Screen design is the broader term I think
that can apply to people needing to design traditional software or
browser based interfaces, with those that design kiosks, atm
machines, and the new brand of displays that will be built into more
general appliances that have digital components, like internet
refrigerators and such. In other words, the tool would need to work
elegantly for the person who does both the visual and interaction
design, or as a shared tool between the person doing the visuals and
the one doing the interaction. It really can't be one or the other,
and I think that's partly why Jason is complaining, the tools don't
blend the right set of features for the kind of work screen design
requires. (And he's right.)

Sidenote: Microsoft is attempting to blur the lines here by making
Expression this interface design product with a broader reach for
both interface designers along with general design work, from my
understanding, I haven't looked at it in the last 9 months and need
to. I think the problem they are going to have is tying it too much
to Microsoft technologies, amongst other things. Although the team
behind Expression is one of the few teams on the planet that could
actually pull it off since the driving force behind the team are ex-
Adobe folks who know a more than most about the product space.

--
Andrei Herasimchuk

Principal, Involution Studios
innovating the digital world

e. andrei at involutionstudios.com
c. +1 408 306 6422

23 Jan 2008 - 3:12pm
Dave Malouf
2005

I'm a current user of Expression Blend and I think it is REALLY close
... about 60% of the way there. that missing 40% is painful. I have
had good contacts with that team and I think that if we give them to
3.5 or 4.0 they might get it right.

What I see wrong with their current direction is that they are trying
to solve the Designer > Developer workflow problem instead of trying
to solve the designer problem and THEN the designer > developer
workflow problem. Which in my mind, if you solve the designer
problem, there will be less issues between the two.

If Blend is not a good designer tool, then you haven't solved the
workflow issue b/c designers won't use it.

I think adding the additional role of "integrator" as I think they
are now calling it, is a HUGE mistake.

>From an MS perspective though, there are 20million MS developers, and
MAYBE 100,000 interactive Designers (not developers). You do the math.
;)

And how many of those designers are going to choose MS over Adobe?

-- dave

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Posted from the new ixda.org
http://www.ixda.org/discuss?post=25002

23 Jan 2008 - 3:26pm
SemanticWill
2007

I hear you Dave --

A while back we talked a bit about this - and I was using Expression for
some design/interaction prototyping - but you you are right - from a
designer's perspective it was still very very painful (caveat - I haven't
used it since September). The usability and flow was just not there for a
true interface/interaction designer to feel comfortable enough giving up the
old tools - even if the new tool would eventually hold the promise of making
a much better integration between designer and developer.
And -- on the other front - as good as Nathan's WPF Unleashed book was - it
was not enough for an IxD to be able to do simple data-binding without much
gnashing of teeth and breaking of bones.

That said - in a year or two it will be a formidable tool/

On Wed, 23 Jan 2008 13:12:15, dave malouf <dave at ixda.org> wrote:

> I'm a current user of Expression Blend and I think it is REALLY close
> ... about 60% of the way there. that missing 40% is painful. I have
> had good contacts with that team and I think that if we give them to
> 3.5 or 4.0 they might get it right.
>
> What I see wrong with their current direction is that they are trying
> to solve the Designer > Developer workflow problem instead of trying
> to solve the designer problem and THEN the designer > developer
> workflow problem. Which in my mind, if you solve the designer
> problem, there will be less issues between the two.
>
> If Blend is not a good designer tool, then you haven't solved the
> workflow issue b/c designers won't use it.
>
> I think adding the additional role of "integrator" as I think they
> are now calling it, is a HUGE mistake.
>
> >From an MS perspective though, there are 20million MS developers, and
> MAYBE 100,000 interactive Designers (not developers). You do the math.
> ;)
>
> And how many of those designers are going to choose MS over Adobe?
>
> -- dave
>
>
> . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
> Posted from the new ixda.org
> http://www.ixda.org/discuss?post=25002
>
>
> ________________________________________________________________
> *Come to IxDA Interaction08 | Savannah*
> February 8-10, 2008 in Savannah, GA, USA
> Register today: http://interaction08.ixda.org/
>
> ________________________________________________________________
> Welcome to the Interaction Design Association (IxDA)!
> To post to this list ....... discuss at ixda.org
> Unsubscribe ................ http://www.ixda.org/unsubscribe
> List Guidelines ............ http://www.ixda.org/guidelines
> List Help .................. http://www.ixda.org/help
>

--
~ will

"Where you innovate, how you innovate,
and what you innovate are design problems"
-------------------------------------------------------
will evans
user experience architect
wkevans4 at gmail.com
-------------------------------------------------------

23 Jan 2008 - 4:46pm
russwilson
2005

Adobe is doing the same thing, but in the other direction with Flex/AIR
and Thermo. Because Adobe has a MUCH better undrestanding of
the designer, I would be prone to align with their efforts. That is, I
would
rather work with their products as they come up to speed on understanding
the developer and the design>developer workflow than the other way around...

Russ

On Jan 23, 2008 3:26 PM, W Evans <wkevans4 at gmail.com> wrote:

> I hear you Dave --
>
> A while back we talked a bit about this - and I was using Expression for
> some design/interaction prototyping - but you you are right - from a
> designer's perspective it was still very very painful (caveat - I haven't
> used it since September). The usability and flow was just not there for a
> true interface/interaction designer to feel comfortable enough giving up
> the
> old tools - even if the new tool would eventually hold the promise of
> making
> a much better integration between designer and developer.
> And -- on the other front - as good as Nathan's WPF Unleashed book was -
> it
> was not enough for an IxD to be able to do simple data-binding without
> much
> gnashing of teeth and breaking of bones.
>
> That said - in a year or two it will be a formidable tool/
>
> On Wed, 23 Jan 2008 13:12:15, dave malouf <dave at ixda.org> wrote:
>
> > I'm a current user of Expression Blend and I think it is REALLY close
> > ... about 60% of the way there. that missing 40% is painful. I have
> > had good contacts with that team and I think that if we give them to
> > 3.5 or 4.0 they might get it right.
> >
> > What I see wrong with their current direction is that they are trying
> > to solve the Designer > Developer workflow problem instead of trying
> > to solve the designer problem and THEN the designer > developer
> > workflow problem. Which in my mind, if you solve the designer
> > problem, there will be less issues between the two.
> >
> > If Blend is not a good designer tool, then you haven't solved the
> > workflow issue b/c designers won't use it.
> >
> > I think adding the additional role of "integrator" as I think they
> > are now calling it, is a HUGE mistake.
> >
> > >From an MS perspective though, there are 20million MS developers, and
> > MAYBE 100,000 interactive Designers (not developers). You do the math.
> > ;)
> >
> > And how many of those designers are going to choose MS over Adobe?
> >
> > -- dave
> >
> >
> > . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
> > Posted from the new ixda.org
> > http://www.ixda.org/discuss?post=25002
> >
> >
> > ________________________________________________________________
> > *Come to IxDA Interaction08 | Savannah*
> > February 8-10, 2008 in Savannah, GA, USA
> > Register today: http://interaction08.ixda.org/
> >
> > ________________________________________________________________
> > Welcome to the Interaction Design Association (IxDA)!
> > To post to this list ....... discuss at ixda.org
> > Unsubscribe ................ http://www.ixda.org/unsubscribe
> > List Guidelines ............ http://www.ixda.org/guidelines
> > List Help .................. http://www.ixda.org/help
> >
>
>
>
> --
> ~ will
>
> "Where you innovate, how you innovate,
> and what you innovate are design problems"
> -------------------------------------------------------
> will evans
> user experience architect
> wkevans4 at gmail.com
> -------------------------------------------------------
> ________________________________________________________________
> *Come to IxDA Interaction08 | Savannah*
> February 8-10, 2008 in Savannah, GA, USA
> Register today: http://interaction08.ixda.org/
>
> ________________________________________________________________
> Welcome to the Interaction Design Association (IxDA)!
> To post to this list ....... discuss at ixda.org
> Unsubscribe ................ http://www.ixda.org/unsubscribe
> List Guidelines ............ http://www.ixda.org/guidelines
> List Help .................. http://www.ixda.org/help
>

--
Russell Wilson
Vice President of Product Design, NetQoS
Blog: http://www.dexodesign.com

23 Jan 2008 - 6:56pm
Dave Malouf
2005

The Thermo stuff is definitely interesting, but thinks like a graphic
designer, not like an interactive designer. Expression in their tutorials
anyway is thinking similarly.

Basic assumption that is false that both are making is that I'm going to
make a finalized high fidelity graphic UI as a flat screen to start and THEN
add interactive elements. This goes against the very way IxD's have been
working.

What is the core element that most (sorry Andrei) IxD's work with?
Wireframes!!!!
What we want more than anything is a way to sketch, not to draw,
interactions as easily as we can wireframe.

THEN! build the styles on top of that!

Now, this is where Expression CAN do a good job, but it isn't marketed well
that way AND it requires WAY too much collaboration with a developer (call
them an integrator all you want, but they are a developer (wolf) in sheep's
clothing ("designer sensibility" - G-d! I love that line). W/o this
developer, you have to be an Andrei to get anything really interactive done
well. Simple click-throughs require you to open up (dare I say it) Visual
Studio and work in C# if you want to do anything production level or Visual
Basic for the boring stuff. At least Flash has a robust internal scripting
language (based on ECMSCript that is REALLY easy to use).

But here's the good part and how I'm working right now (mileage may
vary)...
Do your wireframes (Don't do visual comps to start ... put that part off for
as long as possible)
With your developer confirm a user control architecture that makes sense
from their point of view. an understanding of object-based or oriented
programming theory is helpful here b/c re-use, scalability, and other issues
will arise.
After you do this level of breaking out your screens, then create your user
controls.
Throw this all over the wall. It will be totally usable at the production
level.
LOTS will be missing at this point.
* interactions
* transition animations
* visual design
If your screens have many states, you put all your states in the same
control and use visibility to hide the exception states. They'll be there
for your developers to work with.
While they are busy trying to "integrate" everything; you and your team work
on the visual design. Get buy in!
Then when you are done and they are done, you'll have a raw click through,
but nothing you would want to show.

Ok, Here's the good part.
Just create resource dictionaries and templates using your new visual
design. Blend is really good at this stuff.
You hand the dictionaries and templates over to the developers and all they
have to do is add the files and the references in their solution and voila
(after a few prayers) you should have at least 60-75% of your presentation
layer designed.

Now the bad part.
You have to tweak w/ your "integrator" and then work on your transition
animations (fades, movements, slides, etc.).
You can create dummy examples of what you want in Expression, but according
to "my guy" who is writing the Unleashed book for 2.0, you really need to
code this stuff with a mix of XAML and C# to get it right, which means you
are back at his mercy again.

I have read that folks at frog are having their designers do usable
(production ready) clickable prototypes ONLY in Blend. "my guy" and everyone
else I've spoken with at MS has told me, it is not the point of Blend to do
that and it can't be done. THIS is the #1 flaw of the product and until it
can be resolved, the D > D workflow will never be solved. End of story.

There are no miracles here.

Why use it over Flash?

1) I want a desktop app.
2) I want to use .NET3 (3.5)
3) I want to build in Silverlight
4) My developers are .NET gurus and I want to build on that experience
5) My integration team requires a .NET framework, and WPF offers amazing
features.

There are amazing features in WPF that are worth exploring that is outside
of the Blend tool. It is worth checking out.

On Jan 23, 2008 5:46 PM, Russell Wilson <russ.wilson at gmail.com> wrote:

> Adobe is doing the same thing, but in the other direction with Flex/AIR
> and Thermo. Because Adobe has a MUCH better undrestanding of
> the designer, I would be prone to align with their efforts. That is, I
> would
> rather work with their products as they come up to speed on understanding
> the developer and the design>developer workflow than the other way
> around...
>
> Russ
>
>
>
>
>
> On Jan 23, 2008 3:26 PM, W Evans <wkevans4 at gmail.com> wrote:
>
> > I hear you Dave --
> >
> > A while back we talked a bit about this - and I was using Expression for
> > some design/interaction prototyping - but you you are right - from a
> > designer's perspective it was still very very painful (caveat - I
> > haven't
> > used it since September). The usability and flow was just not there for
> > a
> > true interface/interaction designer to feel comfortable enough giving up
> > the
> > old tools - even if the new tool would eventually hold the promise of
> > making
> > a much better integration between designer and developer.
> > And -- on the other front - as good as Nathan's WPF Unleashed book was
> > - it
> > was not enough for an IxD to be able to do simple data-binding without
> > much
> > gnashing of teeth and breaking of bones.
> >
> > That said - in a year or two it will be a formidable tool/
> >
> > On Wed, 23 Jan 2008 13:12:15, dave malouf <dave at ixda.org> wrote:
> >
> > > I'm a current user of Expression Blend and I think it is REALLY close
> > > ... about 60% of the way there. that missing 40% is painful. I have
> > > had good contacts with that team and I think that if we give them to
> > > 3.5 or 4.0 they might get it right.
> > >
> > > What I see wrong with their current direction is that they are trying
> > > to solve the Designer > Developer workflow problem instead of trying
> > > to solve the designer problem and THEN the designer > developer
> > > workflow problem. Which in my mind, if you solve the designer
> > > problem, there will be less issues between the two.
> > >
> > > If Blend is not a good designer tool, then you haven't solved the
> > > workflow issue b/c designers won't use it.
> > >
> > > I think adding the additional role of "integrator" as I think they
> > > are now calling it, is a HUGE mistake.
> > >
> > > >From an MS perspective though, there are 20million MS developers, and
> >
> > > MAYBE 100,000 interactive Designers (not developers). You do the math.
> > > ;)
> > >
> > > And how many of those designers are going to choose MS over Adobe?
> > >
> > > -- dave
> > >
> > >
> > > . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
> > > Posted from the new ixda.org
> > > http://www.ixda.org/discuss?post=25002
> > >
> > >
> > > ________________________________________________________________
> > > *Come to IxDA Interaction08 | Savannah*
> > > February 8-10, 2008 in Savannah, GA, USA
> > > Register today: http://interaction08.ixda.org/
> > >
> > > ________________________________________________________________
> > > Welcome to the Interaction Design Association (IxDA)!
> > > To post to this list ....... discuss at ixda.org
> > > Unsubscribe ................ http://www.ixda.org/unsubscribe
> > > List Guidelines ............ http://www.ixda.org/guidelines
> > > List Help .................. http://www.ixda.org/help
> > >
> >
> >
> >
> > --
> > ~ will
> >
> > "Where you innovate, how you innovate,
> > and what you innovate are design problems"
> > -------------------------------------------------------
> > will evans
> > user experience architect
> > wkevans4 at gmail.com
> > -------------------------------------------------------
> > ________________________________________________________________
> > *Come to IxDA Interaction08 | Savannah*
> > February 8-10, 2008 in Savannah, GA, USA
> > Register today: http://interaction08.ixda.org/
> >
> > ________________________________________________________________
> > Welcome to the Interaction Design Association (IxDA)!
> > To post to this list ....... discuss at ixda.org
> > Unsubscribe ................ http://www.ixda.org/unsubscribe
> > List Guidelines ............ http://www.ixda.org/guidelines
> > List Help .................. http://www.ixda.org/help
> >
>
>
>
> --
> Russell Wilson
> Vice President of Product Design, NetQoS
> Blog: http://www.dexodesign.com

--
David Malouf
http://synapticburn.com/
http://ixda.org/
http://motorola.com/

23 Jan 2008 - 7:02pm
Mark Schraad
2006

nice...

On Jan 23, 2008, at 7:56 PM, David Malouf wrote:

> but they are a developer (wolf) in sheep's
> clothing

23 Jan 2008 - 7:59pm
Andrei Herasimchuk
2004

On Jan 23, 2008, at 4:56 PM, David Malouf wrote:

> What is the core element that most (sorry Andrei) IxD's work with?
> Wireframes!!!!
> What we want more than anything is a way to sketch, not to draw,
> interactions as easily as we can wireframe.

I'm not sure why you are apologizing to me. 8^) I tend to work that
way myself. I just tend to sketch all on paper or a whiteboard and
not dive into using a vector program if I can avoid, as I find paper
and whiteboards to be faster. But as a process and workflow, I agree
with you.

--
Andrei Herasimchuk

Principal, Involution Studios
innovating the digital world

e. andrei at involutionstudios.com
c. +1 408 306 6422

23 Jan 2008 - 8:19pm
Matt Nish-Lapidus
2007

On Jan 23, 2008 7:56 PM, David Malouf <dave at ixda.org> wrote:
> The Thermo stuff is definitely interesting, but thinks like a graphic
> designer, not like an interactive designer. Expression in their tutorials
> anyway is thinking similarly.
i've always done my wireframes and prototypes with html/javascript ...
html and css are amazing for putting together a quick wireframe.. all
form elements are done, the box model makes it really easy to lay out
distinct areas... and now, with all the nice javascript frameworks,
all the heavy lifting is done for you. throw in something like ruby
on rails and you can have a functional prototype with pretty basic
technical skills.

the best part is that, if you're like me, the final product will end
up being html/css/javascript anyway, so if somebody else needs to do
production work all your prototypes can be a great starting point.. or
if you're like me you probably do most of the production work too :)

then, when it comes time to make things pretty, you can fire up
photoshop/fireworks/illustrator and start skinning elements and
building a style guide.

i will often do html wireframes and use screenshots as a background in
photoshop to do the graphic design...

> Basic assumption that is false that both are making is that I'm going to
> make a finalized high fidelity graphic UI as a flat screen to start and THEN
> add interactive elements. This goes against the very way IxD's have been
> working.
>
> What is the core element that most (sorry Andrei) IxD's work with?
> Wireframes!!!!
> What we want more than anything is a way to sketch, not to draw,
> interactions as easily as we can wireframe.
>
> THEN! build the styles on top of that!

--
Matt Nish-Lapidus
work: matt at bibliocommons.com / www.bibliocommons.com
--
personal: mattnl at gmail.com / www.nishlapidus.com

23 Jan 2008 - 9:28pm
Dave Malouf
2005

HI Matt,

I would NEVER want to use Expression (except Web Designer) if I was doing a
web site.
This is for Silverlight and WPF solutions only. If you ain't doing one of
those, don't bother.

But that being said, one of my issues with the method below is that it
assume that standard HTML controls are right for everything. This has been
my problem with HTML from the beginning. There needs to be a much easier
mechanism (I know there are custom controls for sale) to manage custom
controls and custom skins on controls within HTML.

I do have to say that the one brilliant part of Blend are the controls and
the control part templates. the layout controls in particular are amazing. I
also love that every control doesn't only have a width and height, but they
can all be inherited from a parent with a click of a button AND you can set
minimum and maximums on sizing ... AND WAIT! ... so far we have not seen any
degradation on nesting of layout controls at all. Nest away!!! It makes no
difference what so ever. I really love my layout controls.

Yes, I'm deep in the weeds in an Expression Blend desktop application right
now.

-- dave

On Jan 23, 2008 9:19 PM, Matthew Nish-Lapidus <mattnl at gmail.com> wrote:

> On Jan 23, 2008 7:56 PM, David Malouf <dave at ixda.org> wrote:
> > The Thermo stuff is definitely interesting, but thinks like a graphic
> > designer, not like an interactive designer. Expression in their
> tutorials
> > anyway is thinking similarly.
> i've always done my wireframes and prototypes with html/javascript ...
> html and css are amazing for putting together a quick wireframe.. all
> form elements are done, the box model makes it really easy to lay out
> distinct areas... and now, with all the nice javascript frameworks,
> all the heavy lifting is done for you. throw in something like ruby
> on rails and you can have a functional prototype with pretty basic
> technical skills.
>
> the best part is that, if you're like me, the final product will end
> up being html/css/javascript anyway, so if somebody else needs to do
> production work all your prototypes can be a great starting point.. or
> if you're like me you probably do most of the production work too :)
>
> then, when it comes time to make things pretty, you can fire up
> photoshop/fireworks/illustrator and start skinning elements and
> building a style guide.
>
> i will often do html wireframes and use screenshots as a background in
> photoshop to do the graphic design...
>
>
>
> > Basic assumption that is false that both are making is that I'm going to
> > make a finalized high fidelity graphic UI as a flat screen to start and
> THEN
> > add interactive elements. This goes against the very way IxD's have been
> > working.
> >
> > What is the core element that most (sorry Andrei) IxD's work with?
> > Wireframes!!!!
> > What we want more than anything is a way to sketch, not to draw,
> > interactions as easily as we can wireframe.
> >
> > THEN! build the styles on top of that!
>
> --
> Matt Nish-Lapidus
> work: matt at bibliocommons.com / www.bibliocommons.com
> --
> personal: mattnl at gmail.com / www.nishlapidus.com
>

--
David Malouf
http://synapticburn.com/
http://ixda.org/
http://motorola.com/

24 Jan 2008 - 5:32am
Stew Dean
2007

On 24/01/2008, David Malouf <dave at ixda.org> wrote:
> The Thermo stuff is definitely interesting, but thinks like a graphic
> designer, not like an interactive designer. Expression in their tutorials
> anyway is thinking similarly.
>
> Basic assumption that is false that both are making is that I'm going to
> make a finalized high fidelity graphic UI as a flat screen to start and THEN
> add interactive elements. This goes against the very way IxD's have been
> working.

Well said David,

My view is that it goes even beyond wireframes and UI. I need a tool
that understands all the elements I put into get to the final
interfaces elements. I don't want to add the logic at the interface
level but a level above.

This means a tool that understands site maps and process flows. At the
moment I see lots of folks doing diagram tools, I see a few prototype
tools like Axure that are good for wireframes and out put
specification documents (do people still use them?) and I see Designer
to Dev tools like Expression and the forthcoming Thermo. Both of
these, as a user experience specialist, are of no good to me but may
be of use to the front end devs - they're implimentation tools not
user experience tools.

The tools I need is part CMS (this is more important that you'd
think), part diagram tool (visual representation is very important - I
want control over how things look), part development tool (it needs to
tie into a developement process not create a throw away prototype) and
part project management tool (version control, issue logging, change
tracking).

So get a good CMS, add on a new kind of sitemap / process flow tool,
allow sketching, allow me to create libraries of items with
inheritance (so I can set up a relationship between items, change all
my drop downs from my black and white version to, say, Vista style
boxes), allow content to be entered and managed tagged and organised.

Yes it does sound like I want a swiss army knife but I feel in my work
all these elements are so closely related and are part of the same
story. I've come to accept that Adobe or Microsoft would never design
such a tool as for both of these companies user experience is about
Visual Design and Developers, use user exeprience folks don't appear
to exist in their world.

I feel there is a huge gap in the market but I have yet to see
anything get even remotely close to what I'm imagining. I live in hope
that someone has been secretely working on something like this - if
you are let me know :)

Stewart Dean

24 Jan 2008 - 7:31am
Adam Connor
2007

I love this thread.

And I can't help but think that there is an opportunity here (for
those of us who have the time and desire) to collaborate and begin
down the path of designing such a tool. A tool for User Experience
designers by User Experience designers.

That'd be pretty sweet.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Posted from the new ixda.org
http://www.ixda.org/discuss?post=25002

24 Jan 2008 - 11:32am
Matt Nish-Lapidus
2007

Hey Dave,

> But that being said, one of my issues with the method below is that it
> assume that standard HTML controls are right for everything. This has been
> my problem with HTML from the beginning. There needs to be a much easier
> mechanism (I know there are custom controls for sale) to manage custom
> controls and custom skins on controls within HTML.

There are a couple things to consider about those points:

1) all the modern/cool javascript libraries have easy to use
interface widgets for many different UI elements. (http://extjs.com/,
http://developer.yahoo.com/yui/)

2) the ability to skin form elements is slightly controversial ...
i'm of two minds about it. On one hand, having standard form widgets
improves usability because people always know how to use the same
elements... on the other hand, windows form elements are ugly :)

anyway, lots of cool stuff going on in the world of prototyping, and
always an interesting discussion. I don't have much experience with
Air or Expression Blend, but I might take a quick look just to see
what you're all talking about :)

--
Matt Nish-Lapidus
work: matt at bibliocommons.com / www.bibliocommons.com
--
personal: mattnl at gmail.com / www.nishlapidus.com

24 Jan 2008 - 11:57am
Andrei Herasimchuk
2004

On Jan 24, 2008, at 3:32 AM, Stew Dean wrote:

> So get a good CMS, add on a new kind of sitemap / process flow tool,
> allow sketching, allow me to create libraries of items with
> inheritance (so I can set up a relationship between items, change all
> my drop downs from my black and white version to, say, Vista style
> boxes), allow content to be entered and managed tagged and organised.

This sounds suspiciously like what we do with our custom rolled
prototyping tools.

Also, you left out the aesthetic aspect that such a product would
need. I don't think enough people appreciate just how hard it is to
create a tool hat does all that you want it to for the interactive
side and can still draw sophisticated graphics like Photoshop,
Illustrator or Fireworks. Once you toss in the need for the tool to
handle HTML layout rules or custom layout rules...

Well. It's quite hard product to build if the audience isn't large
enough.

If the product lacked the drawing tools, it would be mostly useless I
think. And the rant from the kind of designer that started this
thread, Jason Santa Maria, would find it useless as well.

It's a touch nut to crack.
>

--
Andrei Herasimchuk

Principal, Involution Studios
innovating the digital world

e. andrei at involutionstudios.com
c. +1 408 306 6422

24 Jan 2008 - 11:57am
Andrei Herasimchuk
2004

On Jan 24, 2008, at 3:32 AM, Stew Dean wrote:

> So get a good CMS, add on a new kind of sitemap / process flow tool,
> allow sketching, allow me to create libraries of items with
> inheritance (so I can set up a relationship between items, change all
> my drop downs from my black and white version to, say, Vista style
> boxes), allow content to be entered and managed tagged and organised.

This sounds suspiciously like what we do with our custom rolled
prototyping tools.

Also, you left out the aesthetic aspect that such a product would
need. I don't think enough people appreciate just how hard it is to
create a tool hat does all that you want it to for the interactive
side and can still draw sophisticated graphics like Photoshop,
Illustrator or Fireworks. Once you toss in the need for the tool to
handle HTML layout rules or custom layout rules...

Well. It's quite hard product to build if the audience isn't large
enough.

If the product lacked the drawing tools, it would be mostly useless I
think. And the rant from the kind of designer that started this
thread, Jason Santa Maria, would find it useless as well.

It's a touch nut to crack.
>

--
Andrei Herasimchuk

Principal, Involution Studios
innovating the digital world

e. andrei at involutionstudios.com
c. +1 408 306 6422

25 Jan 2008 - 9:13am
Stew Dean
2007

On 24/01/2008, Andrei Herasimchuk <andrei at involutionstudios.com> wrote:
>
> On Jan 24, 2008, at 3:32 AM, Stew Dean wrote:
>
> > So get a good CMS, add on a new kind of sitemap / process flow tool,
> > allow sketching, allow me to create libraries of items with
> > inheritance (so I can set up a relationship between items, change all
> > my drop downs from my black and white version to, say, Vista style
> > boxes), allow content to be entered and managed tagged and organised.
>
> This sounds suspiciously like what we do with our custom rolled
> prototyping tools.
>
> Also, you left out the aesthetic aspect that such a product would
> need. I don't think enough people appreciate just how hard it is to
> create a tool hat does all that you want it to for the interactive
> side and can still draw sophisticated graphics like Photoshop,
> Illustrator or Fireworks. Once you toss in the need for the tool to
> handle HTML layout rules or custom layout rules...

I googled out the post that launched this post (at least it appears to be it).

http://www.jasonsantamaria.com/archive/2008/01/23/mucking_up_the_fireworks.php

In the work I do the aesthetic aspect comes a few steps down in the
interation design process (as is increasing the norm for interactive
projects), even in the most visual rich applications I've worked on
wireframes and even interactive prototypes are created either before
or along side visual design.

I agree the tool needs to be able to throw around bit maps and vectors
(good vector drawing is vital I feel for any diagraming tool) but I
would say the tool that allows easy production of graphics almost
feels like a seperate but linked tool to the one I had in mind.

It's almost like the same problem but seen from a visual design flow
rather then a user experience flow (visual design being a
complimentary role to mine) and the post does hit several nails on the
head from that point of view - you want the visual design to plug into
the logic. So if I define a widget using a few lines and even add in
the logic then this can be replaced by a hi-fi version. These elements
could have styles applied to them and again have inheritance (possibly
via CSS).

It's always worth noting that new two folks work the same way so any
tool would be able to deal with different types of skills.

--
Stewart Dean

30 Jan 2008 - 11:00am
Narciso Jaramillo
2006

Coming in on this thread a bit late...lots of good stuff here!

David Malouf wrote:

> The Thermo stuff is definitely interesting, but thinks like a graphic
> designer, not like an interactive designer. Expression in their
> tutorials anyway is thinking similarly.

> Basic assumption that is false that both are making is that I'm going
> to make a finalized high fidelity graphic UI as a flat screen to start
and
> THEN add interactive elements. This goes against the very way IxD's
have
> been working.

Just a quick note on this... Our current public Thermo demo does
emphasize the graphic-design-to-production workflow, but we're also
interested in the early-stage interaction design workflow. We do plan to
have basic drawing tools and built-in components to let you do
wireframing, and you can build custom components and try out
interactions and transitions in wireframe as well.

It sounds like the overall message of this thread is that the
interaction design workflow is much more about early-stage
experimentation, screen architecture, and prototyping, and that while
visuals are somewhat important (for sketching and presentation
purposes), exact visual bits aren't. Is that a fair characterization?

I'd definitely be curious to hear what other kinds of interaction design
needs aren't being addressed by visual-production-oriented design tools
today.

By the way, I'm going to be at Interaction '08 along with other Thermo
folks from Adobe--if you see one of us, please stop us and have a chat!

Thanks,

Narciso (nj) Jaramillo
Adobe Thermo team

30 Jan 2008 - 11:47am
Grady Kelly
2007

But doesn't Thermo limit you to just Flex? I think that is a bad thing. I
have had horrible experiences with Flex development. I would consider
myself a fairly savvy at what I do. Like some, I can design a composite in
Fireworks and then build the xhtml/css/javascript prototype. Thermo outputs
mxml for flex. If I want to make an element do something I am at a loss as
I do not know Flex or ActionScript well enough to do anything. I can find
numerous amounts of ajax/js/css resources to build the interactions that I
know are possible. 99% of the time I can create a UI composite in Fireworks
and build the High Fidelity Prototype in no time.

To follow the article form the first post, Fireworks is an awesome tool for
creating UI. I have used it for the last 8 years and have never been able
to not do something that I needed.

Grady

On Jan 30, 2008 10:00 AM, Narciso Jaramillo <njaramil at adobe.com> wrote:

> Coming in on this thread a bit late...lots of good stuff here!
>
> David Malouf wrote:
>
> > The Thermo stuff is definitely interesting, but thinks like a graphic
> > designer, not like an interactive designer. Expression in their
> > tutorials anyway is thinking similarly.
>
> > Basic assumption that is false that both are making is that I'm going
> > to make a finalized high fidelity graphic UI as a flat screen to start
> and
> > THEN add interactive elements. This goes against the very way IxD's
> have
> > been working.
>
> Just a quick note on this... Our current public Thermo demo does
> emphasize the graphic-design-to-production workflow, but we're also
> interested in the early-stage interaction design workflow. We do plan to
> have basic drawing tools and built-in components to let you do
> wireframing, and you can build custom components and try out
> interactions and transitions in wireframe as well.
>
> It sounds like the overall message of this thread is that the
> interaction design workflow is much more about early-stage
> experimentation, screen architecture, and prototyping, and that while
> visuals are somewhat important (for sketching and presentation
> purposes), exact visual bits aren't. Is that a fair characterization?
>
> I'd definitely be curious to hear what other kinds of interaction design
> needs aren't being addressed by visual-production-oriented design tools
> today.
>
> By the way, I'm going to be at Interaction '08 along with other Thermo
> folks from Adobe--if you see one of us, please stop us and have a chat!
>
> Thanks,
>
> Narciso (nj) Jaramillo
> Adobe Thermo team
>
> ________________________________________________________________
> *Come to IxDA Interaction08 | Savannah*
> February 8-10, 2008 in Savannah, GA, USA
> Register today: http://interaction08.ixda.org/
>
> ________________________________________________________________
> Welcome to the Interaction Design Association (IxDA)!
> To post to this list ....... discuss at ixda.org
> Unsubscribe ................ http://www.ixda.org/unsubscribe
> List Guidelines ............ http://www.ixda.org/guidelines
> List Help .................. http://www.ixda.org/help
>

30 Jan 2008 - 3:28pm
Fred Beecher
2006

On 1/30/08, Narciso Jaramillo <njaramil at adobe.com> wrote:
>
>
> Just a quick note on this... Our current public Thermo demo does
> emphasize the graphic-design-to-production workflow, but we're also
> interested in the early-stage interaction design workflow. We do plan to
> have basic drawing tools and built-in components to let you do
> wireframing, and you can build custom components and try out
> interactions and transitions in wireframe as well.

!!!!!

It sounds like the overall message of this thread is that the
> interaction design workflow is much more about early-stage
> experimentation, screen architecture, and prototyping, and that while
> visuals are somewhat important (for sketching and presentation
> purposes), exact visual bits aren't. Is that a fair characterization?

That's exactly it.

The tools we work with for wireframing and lo-fi interactive prototyping are
very simple. Text boxes, basic shapes, GUI controls (list boxes, radio
buttons, etc.). If you give us a snapable grid (or guides) to lay these
objects out on, it would keep many of us happy for a good long while. : )
Based on what you describe above, that is exactly what you *have* given us.
(Thank you!) I have some more comments about this below...

When it comes to interactions, our needs are less simple. Existing
prototyping tools (outside of Flash, but it's primary purpose is *not*
prototyping) limit the interactions we can represent. While we can fudge a
lot of interactions, giving us a broader palette here would be good. The
interactions shown in the Thermo videos and the way they are applied to
objects is what makes me, personally, the most excited about it.

I'd definitely be curious to hear what other kinds of interaction design
> needs aren't being addressed by visual-production-oriented design tools
> today.

The need to produce or at least facilitate the production of printed
documentation based on text annotations to the wireframes. You guys make
Acrobat, so this shouldn't be a big challenge. : )

Back to the wireframing issue... You say that the current demos focus on the
visual design to development workflow... I think it's crucial that the UXD
to visual design workflow is well thought out. It should be easy for the VDs
(sorry, really I am) to swap out wireframe elements with design elements,
but they shouldn't be locked into the wireframe layout. While these layouts
are always well and rationally thought out, it's not very collaborative to
basically say to the VDs, "Here, color this in for me." The
not-at-all-thought-out idea that pops into my head is something like layer
comps in Photoshop... there might be separate Wireframe and VD comps where
the VD comp uses the WF comp as its starting point.

Another issue is the ability to create re-usable components. It sounds like
you're already planning to build this in, but I want to elucidate the need a
little bit. Currently, I use and love Axure, and have gotten spoiled by some
of its great functionality. Axure allows me to create two (well three, but I
only use two) types of custom components (called masters). You create
masters and then drag them to wireframe pages. With normal masters, editing
the master changes *every instance* of that master. Any
wireframing/prototyping tool I use MUST have this functionality. You can
also set masters to be "custom widgets," meaning that when you change the
master, its instances are NOT updated. This is useful for things like page
types and design patterns. Again, this is *must have* functionality.

Another issue... I don't want to have to buy five pieces of software to do
this. I want to do it *one.* I'm not going to code any Flex, so that one
won't count. But I don't want to have to import stuff into Thermo from all
the other Adobe products. I'd love to be able to just build it in Thermo. I
*could* see the argument for doing a lot of the work in Fireworks and then
importing to Thermo, but you'd need a major update to FW to make that a
viable option.

By the way, I'm going to be at Interaction '08 along with other Thermo
> folks from Adobe--if you see one of us, please stop us and have a chat!

Oh man, you just lost every last second of free time you thought you had. :
)

- Fred

31 Jan 2008 - 12:34pm
Narciso Jaramillo
2006

Hi Grady,

That's actually the point of Thermo J Currently, you have to drop to
code to build behavior in Flex. The idea of Thermo is to make it so you
can prototype and build interactions without having to write code (but
the code is there behind it, so a developer can take it and turn it into
a real app).

Glad to hear you like Fireworks! I'm curious what aspects of it you use
most when designing UI.

Thanks,

nj

Adobe

3 Feb 2008 - 2:18pm
Stew Dean
2007

On 30/01/2008, Narciso Jaramillo <njaramil at adobe.com> wrote:
> Coming in on this thread a bit late...lots of good stuff here!
>
> David Malouf wrote:
>
> > The Thermo stuff is definitely interesting, but thinks like a graphic
> > designer, not like an interactive designer. Expression in their
> > tutorials anyway is thinking similarly.
>
> > Basic assumption that is false that both are making is that I'm going
> > to make a finalized high fidelity graphic UI as a flat screen to start
> and
> > THEN add interactive elements. This goes against the very way IxD's
> have
> > been working.
>
> Just a quick note on this... Our current public Thermo demo does
> emphasize the graphic-design-to-production workflow, but we're also
> interested in the early-stage interaction design workflow. We do plan to
> have basic drawing tools and built-in components to let you do
> wireframing, and you can build custom components and try out
> interactions and transitions in wireframe as well.

That's good to hear Narciso.

The ease of creating custom components is going to be vital here as
the potential problem is that the tool you use only really thinks of
interaction in certain ways - like a windows UI or a web form. Whilst
most of what I work on is web I often find myself creating custom
forms of interaction so the customization is vital.

What is also required is to substittued lo-fi with hi-fi later on in
the process. If the prebuild widgets are too hi-fi it makes the lo-fi
element of the interface sketches look strange.

> It sounds like the overall message of this thread is that the
> interaction design workflow is much more about early-stage
> experimentation, screen architecture, and prototyping, and that while
> visuals are somewhat important (for sketching and presentation
> purposes), exact visual bits aren't. Is that a fair characterization?

Spot on. It's also about diagrams and flow that relate to these
sketches. Notations, clickable site maps and user flow are all used by
the agencies i have worked in the UK and I all major web agencies now
use these in the creation of their websites. The same appears to be
true in the US, so I wouldn't underestimate the audience of a more
diagram based tool as I would say very few large websites are build
these days without going through a fairly standard flow, site map,
wireframe process.

> I'd definitely be curious to hear what other kinds of interaction design
> needs aren't being addressed by visual-production-oriented design tools
> today.

See above for a start. To simplify things I would love to be able to
draw a flow or site map and then be able to link these to my
wireframes. I'd like to be able to set cases (a limited set of
variables like Axure) and I'd like to have an annotation layer that
allows paper/pdf presentation and screen based presentation.

The wireframes created tend to fall into two main categories -
functional pages (forms, interactive tools and heavily interactive
elements) and content (text, pictures and pages where interaction is
about changing the view of the content). The content pages often take
the form of 'templates', one wireframe covering at times the majority
of the site, but both the functional and content templates share
elements so the ability for parts to inherited (and understand their
context) is another thing I'd like to see in any interaction design
tool.

At the moment I'm looking at Flash and Axure to try and create a new
kind of wireframes that include interaction and allows prototyping for
client/user validation so if Thermos can deliver as much drawing
ability (multiple pages/screens is a must have) as Visio but allows
more on top then it starts to become a viable contender for a
Information Architect / User Experience / Interaction Design tool.
It's just in the teams I work in the graphic designers are part of a
team and folks like myself are the ones that create architecture and
main interaction for a project.

If you'd like more detail or examples feel free to contact me.

--
Stewart Dean

6 Mar 2008 - 4:19am
Bryce
2008

We have been frustrated with the same issues, so while still improving we have developed a plug-in, Screen Architect (www.screenarchitect.com) to Enterprise Archtiect (a UML Case tool) www.sparxsystems.com that allows designers to develop UI designers and generate HTML prototypes for users to walk through. We are still adding functionality like .css support, reusability in the form of templates and custom controls but because Screen Architect integrates directly into UML case tool all controls and screens developed are automatically added as UML elements into the underlying analysis model.

This allows designers to tag controls with requirements that the analyst can then pick directly up from the UML model and reuse. It also allows tracability between analysis models and GUI design elements, something that current tools don't allow you to do, so you have to work independently within two tools and keep them in sync throughout the project life.

The future as we see it is allowing users to develop interaction models in UML and tagging screens to these models creating scenarios and flows through your prototype. This will also allow reusability across role and through the complete SDLC.

Syndicate content Get the feed