Re: Interesting Flash interface

28 Oct 2004 - 2:47pm
9 years ago
44 replies
685 reads
sgp
2004

Apologies for coming late on the scene, but I worked back in the day for a design firm known as Plumb Design who developed a tool called Thinkmap. You may recall the Visual Thesaurus project. Plumb is now known as Thinkmap and the software is available for sale on their website http://www.thinkmap.com

I'd never heard of the Tom Sawyer product but its range of examples is intriguing. Inxight, Thinkmap and it seems Tom Sawyer all use some similar core java libraries to create lines and nodes. As far as I'm aware, Flash will freeze up your processor trying to render a really dense data visualization, especially if there's animation of any kind. Joshua Davis of Praystation demonstrated this limit awhile ago with his motion widget studies. A counter example might be Josh On's They Rule project http://www.theyrule.net which performs well. I'm not advocating any particular solution, but you best do your homework on the various issues people have already on this thread - viz metaphor, parameters, biz process, etc.

Best,
[sgp]

Comments

29 Oct 2004 - 8:20am
Craig Marion
2004

Visual Thesaurus and the engine that powered it are examples of hyperbolic browsers.

This is a brief intro to them: http://www.sapdesignguild.org/community/book_people/visualization/controls/hypBrowser.htm. The links to StarTree have changed, though. This is the new link: http://www.inxight.com/products/sdks/st/index.php. There's a link to a product walk-through in the lower right.

This thread seems to be have focused on the value of this interface to huge databases, and that's certainly true. But it's also a valuable alternative to tree hierarchies in just-plain-large databases. We used it in a project a couple of years ago and our users commented that they found it intriguing and fun to use.

Craig Marion
Performance-Centered Designer, SI International, Inc.
http://www.si-intl.com/
Editor, Software Design Smorgasbord
http://www.chesco.com/~cmarion/

sgp <aisgp at earthlink.net> wrote:
Date: Thu, 28 Oct 2004 15:47:19 -0400 (GMT-04:00)
From: sgp
To: discuss-interactiondesigners.com at lists.interactiondesigners.com
Subject: [ID Discuss] Re: Interesting Flash interface

Apologies for coming late on the scene, but I worked back in the day for a design firm known as Plumb Design who developed a tool called Thinkmap. You may recall the Visual Thesaurus project. Plumb is now known as Thinkmap and the software is available for sale on their website http://www.thinkmap.com

---------------------------------
Do you Yahoo!?
Yahoo! Mail Address AutoComplete - You start. We finish.

29 Oct 2004 - 1:45pm
dmitryn
2004

Hyperbolic browsers are an interesting visualization paradigm, but
studies of their effectiveness in information visualization literature
have come up with mixed results at best. The most recent one is a
comparison of four tree visualization systems and Windows Explorer
presented at Infovis this year:

http://www.ics.uci.edu/~kobsa/papers/2004-InfoVis-kobsa.pdf

This study found StarTree to be an "average" system which performed
especially poorly on certain tasks which required users to rotate the
tree structure and had "a number of usability problems". Food for
thought.

Dmitry

(Full disclosure: One of my supervisors developed H3, one of the first
hyperbolic tree browsers, but has not pursued that research direction
for a number of years. :))

--
Dmitry Nekrasovski
http://www.cs.ubc.ca/~dmitry

29 Oct 2004 - 2:34pm
Andrei Herasimchuk
2004

On Oct 29, 2004, at 11:45 AM, Dmitry Nekrasovski wrote:

> Hyperbolic browsers are an interesting visualization paradigm, but
> studies of their effectiveness in information visualization literature
> have come up with mixed results at best. The most recent one is a
> comparison of four tree visualization systems and Windows Explorer
> presented at Infovis this year:
>
> http://www.ics.uci.edu/~kobsa/papers/2004-InfoVis-kobsa.pdf

No offense, but when I read these types of studies and examine the
interfaces they use for them, I'm left with the uneasy feeling that
even in 2004 there seems to be little about how interface design should
be be practiced versus what should be left for another field which is
more pure science or research. These sorts of studies are the sort of
thing that turned me off from CHI over the past ten years.

I mean really... Those have to be some of the worst visuals I have seen
in a long time. It's 2004! Can we at least make some marginal attempt
to make the interfaces of these sorts of things pleasing and useful?
Isn't it obvious that the visuals have a significant impact on the
interaction?

I can't tell what the hell is going on in those interfaces and I'm an
expert computer user. Would the iPod be successful if it looked like a
concrete block of crud? No matter how efficient the interaction of the
spin wheel? If the display of the type and the product itself looked
like these interfaces, would the iPod be worth anything?

How can anyone come up with relevant evidence or research data on
interfaces that fail the basic test of visual design, at least a full
1/3 of what an interface constitutes?

We really have to stop testing interfaces that obviously fail the basic
test of being well designed across the board. Interfaces are about
visual design, about information design and about interaction design.
You can't test the product or its interface unless all three are
well-designed. Well... you can, but all you'll get, IMHO, is useless
data about useless interfaces.

</rant>

Andrei

29 Oct 2004 - 2:43pm
Listera
2004

Andrei Herasimchuk:

> Would the iPod be successful if it looked like a concrete block of crud?

Would DOS/Windows? :-)

Ziya
Nullius in Verba

29 Oct 2004 - 2:53pm
Dave Malouf
2005

> Andrei Herasimchuk:
>
> > Would the iPod be successful if it looked like a concrete
> block of crud?
>
> Would DOS/Windows? :-)
>
> Ziya

Ziya, I think your point is well taken, but I think tehre is a different
type of success that the iPod has. It to me is successful on 2 levels. 1
economically (like DOS/Windows) but it is also successful in that it has
garnered the attention of at least the US community so much so that it is a
cultural icon that no other item its class has achieved, including those
that preceeded it like the Rio. Windows on the other hand is a cultural
joke, which might make iconographic in its own way, but not a successful
one. A better example would be "Google", which I would call the next
platform to contend with.

-- dave

29 Oct 2004 - 3:18pm
Listera
2004

David Heller:

> but I think tehre is a different type of success that the iPod has.

Yes, the non-facetious part of my remark was a plea for a definition of
"success"?

Ziya
Nullius in Verba

29 Oct 2004 - 4:00pm
dmitryn
2004

Andrei,

Your point is valid, but I am not sure you have picked the right target:

1) Two of the four systems examined in that study (three of five if
you count Explorer) are commercial products or parts thereof.

2) The study was done by a researcher unaffiliated with the developers
of any of the tools. Whatever the faults of his work (and there are a
few), I don't believe they have to do with the selection of the
systems involved. Like it or not, these are most of the well known
tree visualization systems out there.

I am not arguing that the "researchers are in an ivory tower far away
from real world design considerations" viewpoint has no basis in
reality. It does. But that doesn't mean that every single research
paper with aesthetically questionable visuals can be used as fodder
for it. Research should and does have merits apart from pretty
pictures.

Dmitry

--
Dmitry Nekrasovski
http://www.cs.ubc.ca/~dmitry

29 Oct 2004 - 4:43pm
Andrei Herasimchuk
2004

On Oct 29, 2004, at 2:00 PM, Dmitry Nekrasovski wrote:

> 1) Two of the four systems examined in that study (three of five if
> you count Explorer) are commercial products or parts thereof.

That means nothing to me. As Ziya pointed out, Windows is a commercial
product as well. Its got a lot of crap too. That has nothing to do with
my larger point.

> 2) The study was done by a researcher unaffiliated with the developers
> of any of the tools. Whatever the faults of his work (and there are a
> few), I don't believe they have to do with the selection of the
> systems involved. Like it or not, these are most of the well known
> tree visualization systems out there.

There are more than a few, there are a ton of flaws, especially with a
vital aspect of the success of the UI which is the visual. That's a big
problem. A huge "there's a white elephant in the living room" type of
problem. I'm not complaining about a lack of pretty pictures. I'm
complaining about a complete lack of any moderate attempt to make the
visuals of these interfaces work at a BASIC level.

Look at these two examples:

http://acg.media.mit.edu/people/fry/zipdecode/
http://www.marumushi.com/apps/socialcircles/

A large part of the success of these interfaces has precisely to do
with the fact they are also visually appealing. The interaction is well
executed, which is strengthened by the visuals. Would you buy a BMW if
it looked like a Yugo? Would you test the success of a BMW as a car if
it had the body of a Yugo?

I wouldn't. It's a waste of time and money.

> I am not arguing that the "researchers are in an ivory tower far away
> from real world design considerations" viewpoint has no basis in
> reality. It does. But that doesn't mean that every single research
> paper with aesthetically questionable visuals can be used as fodder
> for it. Research should and does have merits apart from pretty
> pictures.

Then I can only conclude you missed my point. In fact, that you used
this study as an "example" for the rest of us to ponder shows precisely
the problem I was making with this kind of research. You gave it
credibility by citing it and passing it around to this list of
professionals, even though what it researched and studied has little
credence when it comes to successful interfaces due to the poor and
flawed execution of a major aspect of the UIs in question, their
visuals.

Again, would you spend time and money researching how successful a BMW
is when crammed into a Yugo body? More to the point, would you use any
data or results from a research about successful car design if you saw
they used a Yugo body on a BMW engine?

Andrei

29 Oct 2004 - 4:50pm
Daniel Harvey
2004

You forgot your open rant tag. :)

Daniel

-----Original Message-----
From:
discuss-interactiondesigners.com-bounces at lists.interactiondesigners.com
[mailto:discuss-interactiondesigners.com-bounces at lists.interactiondesign
ers.com]On Behalf Of Andrei Herasimchuk
Sent: Friday, October 29, 2004 3:34 PM
To: IxD Discussion'
Subject: Re: [ID Discuss] Re: Interesting Flash interface

No offense, but when I read these types of studies and examine the
interfaces they use for them, I'm left with the uneasy feeling that
even in 2004 there seems to be little about how interface design should
be be practiced versus what should be left for another field which is
more pure science or research. These sorts of studies are the sort of
thing that turned me off from CHI over the past ten years.

[snip'

We really have to stop testing interfaces that obviously fail the basic
test of being well designed across the board. Interfaces are about
visual design, about information design and about interaction design.
You can't test the product or its interface unless all three are
well-designed. Well... you can, but all you'll get, IMHO, is useless
data about useless interfaces.

</rant>

Andrei

_______________________________________________
Interaction Design Discussion List
discuss at ixdg.org
--
to change your options (unsubscribe or set digest): http://discuss.ixdg.org/
--
Questions: lists at ixdg.org
--
Announcement Online List (discussion list members get announcements already)
http://subscribe-announce.ixdg.org/
--
http://ixdg.org/

29 Oct 2004 - 4:54pm
Alain D. M. G. ...
2003

--- Andrei Herasimchuk <andrei at designbyfire.com> a écrit :

> > Hyperbolic browsers are an interesting visualization paradigm, but
> > studies of their effectiveness in information visualization
> literature
> > have come up with mixed results at best. The most recent one is a
> > comparison of four tree visualization systems and Windows Explorer
> > presented at Infovis this year:
> >
> > http://www.ics.uci.edu/~kobsa/papers/2004-InfoVis-kobsa.pdf
>
> No offense, but when I read these types of studies and examine the
> interfaces they use for them, I'm left with the uneasy feeling that
> even in 2004 there seems to be little about how interface design
> should
> be be practiced versus what should be left for another field which is
>
> more pure science or research. These sorts of studies are the sort of
>
> thing that turned me off from CHI over the past ten years.
>
> I mean really... Those have to be some of the worst visuals I have
> seen
> in a long time. It's 2004! Can we at least make some marginal attempt
>
> to make the interfaces of these sorts of things pleasing and useful?

You have to realize that these things are made in university labs where
there is usually no money to hire a professional illustrator with
training in engineering drawings and industrial design.

Just a few days ago I was at UIST 2004 and a group of computer
scientists from France (Intuilab and CENA) made an interesting
presentation on a toolkit that adressed these kinds of problems.
(Revisiting visual interface programming: Creating GUI tools for
designers and programmers.) They stressed however that they always
worked with illlustrators and graphics specialist who had a training
and experience in industrial design, working closely with engineers in
auto and aerospace industries. Some persons in the audience questioned
the utility of such tools given what they thought to be the lack of
desire of graphic persons to work under constraints. The authors noted
that in their experience, the professional industrial designers they
worked with requested in fact more constraints, correcting the toolkit
as it was developed.

Alain Vaillancourt

__________________________________________________________
Lèche-vitrine ou lèche-écran ?
magasinage.yahoo.ca

29 Oct 2004 - 4:58pm
dmitryn
2004

> Look at these two examples:
>
> http://acg.media.mit.edu/people/fry/zipdecode/
> http://www.marumushi.com/apps/socialcircles/

Very nice systems, but do you have an example of a visually appealing
tool that could have been used in that study (i.e. a tree
visualization that scales to thousands of nodes)?

Dmitry

29 Oct 2004 - 5:20pm
Andrei Herasimchuk
2004

On Oct 29, 2004, at 2:58 PM, Dmitry Nekrasovski wrote:

> Very nice systems, but do you have an example of a visually appealing
> tool that could have been used in that study (i.e. a tree
> visualization that scales to thousands of nodes)?

Why is this relevant? What is the point of your question? I haven't
looked in depth for one that could be used in place of any one in that
study, but in looking at the examples from that study, it's obvious
where just the basics of visual design would dramatically help the
interface in conveying dense and complex information.

Tell whomever designed these interfaces to buy Tufte's "Envisioning
Information" and read it. All the basics are covered perfectly by Mr.
Tufte.

Andrei

29 Oct 2004 - 5:47pm
Kevin Cheng
2004

Just so I truly understand your point Andrei (because I don't really
find your BMW analogy translates very well).

You're saying they're wasting their time and money because they're
researching something that LOOKS crap and is obviously therefore,
crap.

Yes?

Kevin Cheng (KC)
OK/Cancel: Interface Your Fears
kc at ok-cancel.com
www.ok-cancel.com

29 Oct 2004 - 7:11pm
Andrei Herasimchuk
2004

On Oct 29, 2004, at 3:47 PM, Kevin Cheng wrote:

> Just so I truly understand your point Andrei (because I don't really
> find your BMW analogy translates very well).

It is a poor analogy, I acknowledge that.

> You're saying they're wasting their time and money because they're
> researching something that LOOKS crap and is obviously therefore,
> crap.

It's more than just looks.

Interface design is about three key areas: visual, information and
interaction. When *any* one of those areas is obviously compromised or
fundamentally flawed, then I claim the entire interface is compromised.
(In the example provided, all of the interfaces are obviously
compromised at the visual level.) You may test it and get marginal data
on it, but to test the "success" or "user satisfaction" of the
interface is folly.

The question is, what constitutes "obvious."

That would be a good discussion. There are well known fundamental
principles in each area that should be fairly well known by now, or
should become agreed on by those who practice the craft. As I mentioned
to Dmitry, the examples used in the study could use from following the
principles of visual and analytical design found in Tufte's Envisioning
Information. Bringhurst's Elements of Typographic Style is the best
choice with regard type, imho. There are other books and principles we
could cite with regard to interaction and information architecture as
well, Dreyfuss's "Designing for People" being one that comes to mind
for interaction principles or process of design.

Andrei

29 Oct 2004 - 7:57pm
dmitryn
2004

On Fri, 29 Oct 2004 15:20:21 -0700, Andrei Herasimchuk
<andrei at involutionstudios.com> wrote:
> [Please voluntarily trim replies to include only relevant quoted material.]
>
> Why is this relevant? What is the point of your question? I haven't
> looked in depth for one that could be used in place of any one in that
> study, but in looking at the examples from that study, it's obvious
> where just the basics of visual design would dramatically help the
> interface in conveying dense and complex information.

The point of my question was that, while all your points are valid in
theory, it is much easier to design a visually appealing interface for
a toy system like your examples than for one that actually has to
handle non-trivial amounts of information. And yes, I am familiar with
Tufte's work, and realize that the systems in the study likely violate
some of his principles. But incorporating them requires skills that,
as Alain already pointed out, are not likely to be available in a
research environment.

Imagine if research had to meet a set of guidelines of whatever
constitutes "best practices" in a field before it could be published.
If that were the case, Brin and Page would still be trying to submit
PageRank to some conference. :)

Dmitry

--
Dmitry Nekrasovski
http://www.cs.ubc.ca/~dmitry

29 Oct 2004 - 8:40pm
Listera
2004

Dmitry Nekrasovski:

> Imagine if research had to meet a set of guidelines of whatever
> constitutes "best practices" in a field before it could be published.
> If that were the case, Brin and Page would still be trying to submit
> PageRank to some conference. :)

Wait, did you just dis the notion of "best practices" as bogus? :-)

Ziya
Nullius in Verba

29 Oct 2004 - 8:55pm
dmitryn
2004

On Fri, 29 Oct 2004 21:40:14 -0400, Listera <listera at rcn.com> wrote:
> [Please voluntarily trim replies to include only relevant quoted material.]
>
> Dmitry Nekrasovski:
>
> > Imagine if research had to meet a set of guidelines of whatever
> > constitutes "best practices" in a field before it could be published.
> > If that were the case, Brin and Page would still be trying to submit
> > PageRank to some conference. :)
>
> Wait, did you just dis the notion of "best practices" as bogus? :-)
>
> Ziya
> Nullius in Verba
>
>
>
>
> _______________________________________________
> Interaction Design Discussion List
> discuss at ixdg.org
> --
> to change your options (unsubscribe or set digest): http://discuss.ixdg.org/
> --
> Questions: lists at ixdg.org
> --
> Announcement Online List (discussion list members get announcements already)
> http://subscribe-announce.ixdg.org/
> --
> http://ixdg.org/
>

--
Dmitry Nekrasovski
http://www.cs.ubc.ca/~dmitry

29 Oct 2004 - 8:56pm
Andrei Herasimchuk
2004

On Oct 29, 2004, at 5:57 PM, Dmitry Nekrasovski wrote:

> The point of my question was that, while all your points are valid in
> theory, it is much easier to design a visually appealing interface for
> a toy system like your examples than for one that actually has to
> handle non-trivial amounts of information.

Not to offend once again, but that is bullshit. That's what I thought
you might be implying. It's not "valid in theory." It's either valid or
its not. I claim it is valid. Entirely valid.

Further, good design takes time. If one is going to take the time to
engineer a something as complex as those examples, they have the time
to go the distance. Stop making excuses for people ignoring the most
basic of visual design principles

Finally, be careful what you call "toy systems." Both Ben Fry and
Marcos Weskamp have and are working on some of the most complex
information and interaction design problems around these days. You
should do a little research on what they are working on before taking
that kind of swipe. It might come back to haunt you that you put that
out on record.

> And yes, I am familiar with
> Tufte's work, and realize that the systems in the study likely violate
> some of his principles. But incorporating them requires skills that,
> as Alain already pointed out, are not likely to be available in a
> research environment.

Then they better get some of those skills or stop providing studies
like this that do very little to forward the field of design by
researching fundamentally flawed interfaces. IMHO.

Further, it's not that they violate "some" of his principles, the
visuals in those examples aren't even on the same planet in terms of
passing a basic test. As others have heard me say, it takes work to
make stuff that ugly, so why not put that effort into using the most
basic of visual design principles to at least not make it ugly? Would
you show up to a job interview in ripped jeans and sweat-stained
t-shirt? No? then why on earth would you allow your interface to do the
equivalent when being introduced to users for the first time?

> Imagine if research had to meet a set of guidelines of whatever
> constitutes "best practices" in a field before it could be published.
> If that were the case, Brin and Page would still be trying to submit
> PageRank to some conference. :)

Oh come on. It's not that hard.

It's just not. If it were hard, I'd have some sympathy, but it's just
not that hard!

And don't get me started about Google's poorly executed visual design.
I already did a small example of moderate clean-up as an exercise on my
blog. http://www.designbyfire.com/000039.html And that example took me
less than 2 hours to do with minimal thinking or work. My solution has
many problems, but the general visual clean-up helps quite a bit I
think while also providing more data to view. That Google is successful
in spite of bad design just says more about our tolerance for bad
design.

In the words of Mr. Rand: "The public is more familiar with bad design
than good design. It is, in effect, conditioned to prefer bad design,
because that is what it lives with. The new becomes threatening, the
old reassuring."

But just because that's so doesn't make it right, nor something we --
as designers -- should tolerate.

I am continually amazed and somewhat appalled 15 years into this career
the degree some people in the field of "interaction" design continually
make excuses for lacking the most basic of visual design skills for
work in interface design. I'm not asking for the next great design of
the century. I'm just asking for stuff that should be fundamental and
basic to everything we do in this field.

Andrei

29 Oct 2004 - 9:56pm
Dave Malouf
2005

I think I really believe in what Andrei is staying.
Research is the one place where TIME is not an issue and thus people should
take that time and use it to their fullest. When one is in grad school the
oness is on THEM to the do the work, it should be a requirement to have the
craft and understanding of theory not only to do testing, but to do the
environment worth testing. If nothing else, it is so easy in a grad school
environment to create collaborative projects that even if the
students/researchers themselves were not all information designers, they
should at least collaborate with them. THAT is the real world, and the lack
of this type of work is why people like Andrei (and myself) are just so
frustrated with the academy.

They don't seem to take the wholistic veiw very seriously.

<admin hat>
Now, I know sometimes being direct cuts through all the "shite" of the
conversation ... But I do think that it would be better to be cordial than
to ere on the side of rudeness. Not pointing fingers but the tone is getting
a little out there. Thank you.
</admin hat>

-- dave

29 Oct 2004 - 10:04pm
Listera
2004

David Heller:

> They don't seem to take the wholistic veiw very seriously.

Why should they?

Ziya
Nullius in Verba

30 Oct 2004 - 12:23am
Pete Gordon - U...
2004

As a Java Developer/Architect/Programmer/whatever, I empathize with
those that developed the interfaces in the research report. It seems
often times the technology and tool itself dictates or at least
influences the visual design aspect of a project, I believe.

As one who appreciates quality visual design, I do appreciate more the
appeal of the Ben Fry example. And, have learned that it is anything
but a "toy system", in fact it has been built on Java technology; just
as was apparently used in the research report--although it has been
abstracted out using Processing (http://processing.org/), which looks
amazing. Oh, yea and Ben Fry is apply his programming and visual
design skills to the human genome, an enormous amount of data.

I think Ben Fry addressed somewhat the issue that is being debated, in
this interview
(http://since1968.com/article/11/benjamin-fry-interview).

The Interview Question:
...To what extent is the type of data we visualize limited by the
simplicity of the tools, and how soon should web developers begin
looking for more robust tools to display more complex data?..

Here is what Ben says....
"Part of the problem with the simplicity of the tools is that it sets
the expectation for what types of things can be “seen.” My feeling is
that as a result, the current expectation is set lower than it should
be.
...
As for when to look for the tools, it’s difficult to say. There are a
small handful of companies who are making interesting tools/projects
currently, but for myself I’d be more interested in having more people
(designers and programmers) thinking about these issues and trying to
build interesting things themselves. Maybe Proce55ing and things like
it can start enabling that, or maybe it’ll take a few books or some
people to set a few more good examples to set the standard higher."

Be sure to check out the whole interview
(http://since1968.com/article/11/benjamin-fry-interview).

Thanks for the diversion from my norm, I learned a lot.

Pete Gordon
usersfirst.com
software tools to observe, capture, and communicate user experience

30 Oct 2004 - 4:26am
pabini
2004

Andrei

I'm entirely with you on your comments about hyperbolic browsers, though I
wouldn't use such a nice word to describe them. ;-) They are horrendous from
a visual design standpoint, and their interaction design is just as bad.

I looked at your examples of "visually appealing" design though and was not
favorably impressed. The zipdecode page suffers from a lack of contrast
between the text and its background, and it's not usable. In fact, it's
downright cryptic. Visual design isn't successful unless it effectively
communicates how to use the functionality on the page. Social Circles is
more visually appealing, but not wholly satisfying. It's not very useful or
usable.

Pabini

----- Original Message -----
From: "Andrei Herasimchuk" <andrei at involutionstudios.com>
To: "IxD Discussion'"
<discuss-interactiondesigners.com at lists.interactiondesigners.com>
Sent: Friday, October 29, 2004 2:43 PM
Subject: Re: [ID Discuss] Re: Interesting Flash interface

> [Please voluntarily trim replies to include only relevant quoted
material.]
>
> On Oct 29, 2004, at 2:00 PM, Dmitry Nekrasovski wrote:
>
> > 1) Two of the four systems examined in that study (three of five if
> > you count Explorer) are commercial products or parts thereof.
>
> That means nothing to me. As Ziya pointed out, Windows is a commercial
> product as well. Its got a lot of crap too. That has nothing to do with
> my larger point.
>
> > 2) The study was done by a researcher unaffiliated with the developers
> > of any of the tools. Whatever the faults of his work (and there are a
> > few), I don't believe they have to do with the selection of the
> > systems involved. Like it or not, these are most of the well known
> > tree visualization systems out there.
>
> There are more than a few, there are a ton of flaws, especially with a
> vital aspect of the success of the UI which is the visual. That's a big
> problem. A huge "there's a white elephant in the living room" type of
> problem. I'm not complaining about a lack of pretty pictures. I'm
> complaining about a complete lack of any moderate attempt to make the
> visuals of these interfaces work at a BASIC level.
>
> Look at these two examples:
>
> http://acg.media.mit.edu/people/fry/zipdecode/
> http://www.marumushi.com/apps/socialcircles/
>
> A large part of the success of these interfaces has precisely to do
> with the fact they are also visually appealing. The interaction is well
> executed, which is strengthened by the visuals. Would you buy a BMW if
> it looked like a Yugo? Would you test the success of a BMW as a car if
> it had the body of a Yugo?
>
> I wouldn't. It's a waste of time and money.
>
> > I am not arguing that the "researchers are in an ivory tower far away
> > from real world design considerations" viewpoint has no basis in
> > reality. It does. But that doesn't mean that every single research
> > paper with aesthetically questionable visuals can be used as fodder
> > for it. Research should and does have merits apart from pretty
> > pictures.
>
> Then I can only conclude you missed my point. In fact, that you used
> this study as an "example" for the rest of us to ponder shows precisely
> the problem I was making with this kind of research. You gave it
> credibility by citing it and passing it around to this list of
> professionals, even though what it researched and studied has little
> credence when it comes to successful interfaces due to the poor and
> flawed execution of a major aspect of the UIs in question, their
> visuals.
>
> Again, would you spend time and money researching how successful a BMW
> is when crammed into a Yugo body? More to the point, would you use any
> data or results from a research about successful car design if you saw
> they used a Yugo body on a BMW engine?
>
> Andrei
>
> _______________________________________________
> Interaction Design Discussion List
> discuss at ixdg.org
> --
> to change your options (unsubscribe or set digest):
http://discuss.ixdg.org/
> --
> Questions: lists at ixdg.org
> --
> Announcement Online List (discussion list members get announcements
already)
> http://subscribe-announce.ixdg.org/
> --
> http://ixdg.org/

30 Oct 2004 - 3:00pm
Andrew Otwell
2004

> The point of my question was that, while all your points are valid in
> theory, it is much easier to design a visually appealing interface for
> a toy system like your examples than for one that actually has to
> handle non-trivial amounts of information.

Right, Fry's attempt to visualize the entire human genome, that's a
*toy system*.
http://acg.media.mit.edu/people/fry/

30 Oct 2004 - 3:08pm
Dave Malouf
2005

> Right, Fry's attempt to visualize the entire human genome, that's a
> *toy system*.
> http://acg.media.mit.edu/people/fry/

What do people think of David Small's work?
His visualization of the Genome and his work with the Torah/Talmud both
really interest me. It is not hyperbolic, it is zoom and fade (I guess you
call it; I don't know the "official" name for it). I think it accomplishes
structure, behavior and presentation (that's the way I break down what
Andrei was calling information, interaction, and visual: see my blog @
http://synapticburn.com/ for more on that).

David's work can be seen at ... http://www.davidsmall.com/

-- david

30 Oct 2004 - 4:06pm
Listera
2004

David Heller:

> What do people think of David Small's work?

"Moving away from the domain of sheer visuality into the realm of the
tactile and interactive, the corporeal capacity of language is dissected
into workable spatial and temporal applications for the digital age."

"application"? Is it art or does it pass the usable/useful test? :-)

Ziya
Nullius in Verba

30 Oct 2004 - 4:14pm
Dave Malouf
2005

> > What do people think of David Small's work?
>
> "Moving away from the domain of sheer visuality into the realm of the
> tactile and interactive, the corporeal capacity of language
> is dissected
> into workable spatial and temporal applications for the digital age."
>
> "application"? Is it art or does it pass the usable/useful test? :-)

I saw a demo of his work awhile back and I have to say that it is both,
which I think is Andrei's point. Beautiful and useful is the ideal we are
striving for. The very behaviors we make in the direction of an application
will be mitigated by our emotional responses to it.

I admi that a lot of David's work might be more artistic than useful, but I
was purposely pointing people to the Torah/Talmud and Genome work which is
further in the past.

-- dave

30 Oct 2004 - 4:15pm
whitneyq
2010

At 05:06 PM 10/30/2004 -0400, Listera wrote:
>Is it art or does it pass the usable/useful test? :-)

That's a very interesting question for me. In the world of voting systems
(one in which I'm currently immersed):

COMPARE

The Voting Booth Project at Parsons School of Design - a group of artists
and designers were given actual Vote-o-matics to work with
http://a.parsons.edu/~voting_booth/

TO

Design for Democracy's redesign of the Vote-o-matic booth for the Chicago
board of elections
http://americanhistory.si.edu/vote/design.html
(scroll down, or go directly to the enlarged image:
http://americanhistory.si.edu/vote/large/8_04_lrg.jpg

Perhaps I am painting a fine verbal line here (though the distinction is
important), but the first seems to me to be firmly "art" while the second
is "design."

Back in a former life of theatre design, we used to talk about the
difference between "craft" and "art" with the same implication of
useful/usable vs. interesting/decorative.

Whitney Quesenbery
Whitney Interactive Design, LLC
w. www.WQusability.com
e. whitneyq at wqusability.com
p. 908-638-5467

UPA - www.usabilityprofessionals.org
STC Usability SIG: www.stcsig.org/usability

30 Oct 2004 - 4:35pm
Listera
2004

Whitney Quesenbery:

> Perhaps I am painting a fine verbal line here (though the distinction is
> important), but the first seems to me to be firmly "art" while the second
> is "design."

Incidentally, while at the Parson show, someone was showing around an
unbelievably confusing voting card from, I think, somewhere in the mid-west.
It listed Kerry on top and Bush the bottom, except that their respective
punch slots were literally crossed, Bush on top Kerry at the bottom. Anyway,
there was a reporter from a European paper at the show and ended up asking
me what I thought of the booths on exhibit and the actual ones. I struggled
a bit but was able to say that the former was meant for us to think, the
latter to catch us not thinking.

Ziya
Nullius in Verba

30 Oct 2004 - 5:44pm
Andrei Herasimchuk
2004

On Oct 30, 2004, at 2:26 AM, Pabini Gabriel-Petit wrote:

> I looked at your examples of "visually appealing" design though and
> was not
> favorably impressed. The zipdecode page suffers from a lack of contrast
> between the text and its background, and it's not usable.

I would disagree with you on this point, as I find the low-contrast
aspect somewhat refreshing in this case. However, outside of that
issue, it should be noted that fixing the contrast in an example such
as zipcode is one line of code and extraordinarily easy given the
overall design approach, which is a two-toned information design
approach. Given that, I don't think its that big of a deal.

> Visual design isn't successful unless it effectively
> communicates how to use the functionality on the page.

I would disagree with you here as well. That is not the ultimate goal
or measure of visual design. Visual design has to compliment and
support both the information and interaction pieces creating harmony,
communicating what it needs to depending on the goal of the overall
design. But it's job is not necessarily to communicate how to use
something specifically.

Example: Once again the iPod. There's nothing in its visual appearance
or aesthetics that effectively communicates how to use the spin wheel.
In fact, if you give an iPod to people, invariably they ask you how to
use it. However, once you use it and understand how to use it (because
it takes all of five seconds to grasp its function when you play with
it), it's nearly perfect. It's entirely satisfying as well. The visual
design does nothing here, except to reinforce the simplicity of the
device.

Does the iPod fail at a visual and aesthetic level? I'd say no way.
It's one of the most successful in the past decade.

> Social Circles is more visually appealing, but not wholly satisfying.
> It's not very useful or
> usable.

You'll have to be more explicit in detail than that. I agree it has
problems, but overall, for a design exercise, which is what it was for
Marcos, I think it succeeds far more than most commercial examples we
have these days.

Andrei

30 Oct 2004 - 6:03pm
Andrei Herasimchuk
2004

On Oct 30, 2004, at 1:08 PM, David Heller wrote:

> David's work can be seen at ... http://www.davidsmall.com/

Interesting stuff. I'd love to see these in person before I made my own
persona judgment about them though. Maybe next time I'm in Boston.

Andrei

30 Oct 2004 - 7:02pm
dmitryn
2004

On Sat, 30 Oct 2004 15:00:55 -0500, Andrew Otwell <andrew at heyotwell.com> wrote:

> Right, Fry's attempt to visualize the entire human genome, that's a
> *toy system*.

Rest assured that I meant no disrespect to Fry. As someone currently
involved in a project on visualizing large amounts of biological
information (phylogenetic trees), I can recognize the tremendous
challenges involved in an endeavour like genomic visualization.
However, I was making a very specific point about the scale on which a
specific system, unrelated to this work, was designed. I am sure that
it can be refuted with better arguments than those based on the
designer's reputation. :)

Dmitry

--
Dmitry Nekrasovski
http://www.cs.ubc.ca/~dmitry

30 Oct 2004 - 11:27pm
pabini
2004

Hi Andrei

Pabini Gabriel-Petit wrote:
>
> > I looked at your examples of "visually appealing" design though and
> > was not favorably impressed. The zipdecode page suffers from a lack of
contrast
> > between the text and its background, and it's not usable.

Andrei wrote:
> I would disagree with you on this point, as I find the low-contrast
> aspect somewhat refreshing in this case. However, outside of that
> issue, it should be noted that fixing the contrast in an example such
> as zipcode is one line of code and extraordinarily easy given the
> overall design approach, which is a two-toned information design
> approach. Given that, I don't think its that big of a deal.

***[Pabini] Of course, it would have been easy to fix. All the more reason
why the problem should have been rectified. Putting out low-contrast pages
does a disservice to all sight-impaired users. I like the two-toned info
design approach, but it would have been more successful with better
contrast.

Pabini wrote:
> > Visual design isn't successful unless it effectively
> > communicates how to use the functionality on the page.

Andrei wrote:
> I would disagree with you here as well. That is not the ultimate goal
> or measure of visual design. Visual design has to compliment and
> support both the information and interaction pieces creating harmony,
> communicating what it needs to depending on the goal of the overall
> design. But it's job is not necessarily to communicate how to use
> something specifically.

***[Pabini] I didn't say that was the *only* criterion for successful
design, but it's certainly an important one. If there is functionality on a
page (There always is if navigation is present.), the page isn't successful
unless it effectively communicates that functionality. *Of course*, it's
important for visual design to complement and support information, too.

Andrei wrote:
> Example: Once again the iPod. There's nothing in its visual appearance
> or aesthetics that effectively communicates how to use the spin wheel.
> In fact, if you give an iPod to people, invariably they ask you how to
> use it. However, once you use it and understand how to use it (because
> it takes all of five seconds to grasp its function when you play with
> it), it's nearly perfect. It's entirely satisfying as well. The visual
> design does nothing here, except to reinforce the simplicity of the
> device.

***[Pabini] We've jumped from visual design to industrial design here. A
hardware affordance does to some extent communicate its functionality. The
precise function of that spin wheel may not be intuitable, but it sounds
like it's easy to learn and use. I've heard nothing but good things about
the iPod.

Andrei wrote:
> Does the iPod fail at a visual and aesthetic level? I'd say no way.
> It's one of the most successful in the past decade.

***[Pabini] No argument here. :-)

Andrei wrote:
> > Social Circles is more visually appealing, but not wholly satisfying.
> > It's not very useful or
> > usable.

Andrei wrote:
> You'll have to be more explicit in detail than that. I agree it has
> problems, but overall, for a design exercise, which is what it was for
> Marcos, I think it succeeds far more than most commercial examples we
> have these days.

***[Pabini] After playing with it a bit, I ended up with a morass of text
upon text upon text. Nothing was readable. The initial state was visually
appealing, but everything went downhill from there. I agree with you that
there's a lot of awful visual design on the Web--some of it on the sites of
renowned professionals who should know better. I also agree with you that
designers must attend to visual design if they want to create truly
successful designs.

31 Oct 2004 - 6:07am
Andrei Herasimchuk
2004

On Oct 30, 2004, at 9:27 PM, Pabini Gabriel-Petit wrote:

> ***[Pabini] Of course, it would have been easy to fix. All the more
> reason
> why the problem should have been rectified. Putting out low-contrast
> pages
> does a disservice to all sight-impaired users. I like the two-toned
> info
> design approach, but it would have been more successful with better
> contrast.

Fry did the project as a student project. It was nothing more than an
exercise from what I know of it. In that regard, I assume he was
experimenting with more subtle tones to make the bright dots appear
without having to resort to using color. In that regard, it seems like
a reasonable UI decision. And in fact, a really good UI decision.

Further, the UI is not really low-contrast. It uses a dark gray with a
light yellow, which is in fact high contrast. Low contrast would be two
tones that have nearly the same gray or gamma values.

The instructions are a bit on the lower-contrast side, but not the UI
itself. That the instructions are somewhat lower contrast (so they
don't conflict with the important data, the UI itself) does not bother
me. Reflecting on this further, I'd have to say your opinion that the
approach fails due to low-contrast seems wrong to me. The UI itself not
really low-contrast at all, just the instructions.

> ***[Pabini] We've jumped from visual design to industrial design here.

And? People respond to how the iPod *looks* before they respond to how
it works. The fact the object is white, shades of white, very little
gaudy iconography, clean type, etc. That visual aesthetic is what I was
referring to.

> ***[Pabini] After playing with it a bit, I ended up with a morass of
> text
> upon text upon text. Nothing was readable. The initial state was
> visually
> appealing, but everything went downhill from there.

It didn't for me. I moved objects so that the text wasn't overlapping.
In fact, I explicitly did that as the first thing to do after the
initial start point. Why did you move text so it covered itself? Seems
like an odd thing to do to me. I think you might be playing devil's
advocate for the sake of playing devil's advocate here. The UI itself
doesn't necessarily create a "morass of text" and you admitted that you
were the one who made the mess from a reasonable starting point. Does
that mean the UI fails? Because it gave you enough rope to hang
yourself? I just don't buy that.

I'll grant the UI to Social Circles does has it own set of issues, but
allowing users to create "morasses of text" is not one of them.

Andrei

31 Oct 2004 - 6:29pm
Kevin Cheng
2004

------------------------------------------
Andrei said:
It's more than just looks.

Interface design is about three key areas: visual, information and
interaction. When *any* one of those areas is obviously compromised or

fundamentally flawed, then I claim the entire interface is
compromised.
(In the example provided, all of the interfaces are obviously
compromised at the visual level.) You may test it and get marginal
data
on it, but to test the "success" or "user satisfaction" of the
interface is folly.
------------------------------------------

So the researchers here said, presumably, something like, "let's find
out whether this type of interface, in terms of interaction design, is
usable/useful" and took a look at what was on the market. You're
saying they shouldn't have bothered because a key ingredient - visuals
- was sorely lacking and thus undermined any kind of results they
might get.

What do you propose then, given the three areas (visual, info,
interaction) are tightly intertwined and can't really be separated?

My thought is this - there are indeed academic studies that look at
things which seem blatantly obvious ("well of COURSE they're going to
say that, those products were obviously fugly. I could have told you
that!"). Yet there are plenty of studies which show that what is
thought to be common knowledge is in fact, a misconception.

So these people came back with the expected results. Was that a waste
of time? I don't think necessarily. Certainly, I don't think it's
realistic to say these people, who are unlikely to be visual
designers, to go and BUILD one that's perfect first. Instead, maybe
the researchers can simply ascertain there are problems and then
identify where they come from.

If they said something like "these things are completely unusable but
that doesn't mean the interaction is useless, many of the problems
were found to be stemmed from confusion with the distracting visuals"
then that would be useful.

So summarizing:
1) nobody can separate info/visual/interaction from a product. They're
intertwined
2) given this, researchers must test knowing this and actively test
all aspects instead of trying to isolate
3) the "obvious" is not always so and research exists partially for
this reason

And to Dave's point: time is VERY MUCH a factor in research. Any
thoughts that they have all the time in the world to just build the
perfect test tools is unrealistic.

Kevin Cheng (KC)
OK/Cancel: Interface Your Fears
kc at ok-cancel.com
www.ok-cancel.com

31 Oct 2004 - 9:41pm
Andrei Herasimchuk
2004

On Oct 31, 2004, at 5:43 PM, Pabini Gabriel-Petit wrote:

> ***[Pabini] Since the functionality isn't very usable, because the page
> doesn't do a good job of visually communicating its functionality, the
> instructions are key, but are hard to read, which was the main point I
> was
> making. You've often made the point that people neglect visual design
> and
> that the quality of user interfaces is diminished thereby. I'd agree,
> but
> Web designers often overemphasize visual design at the expense of
> usability
> and readability, and that's at least as big a problem.

I disagree with you that a majority of "web designers often
overemphasize visual design at the expense of usability." I think
that's a strawman argument that has done nothing but allow too many in
the field to avoid the issue on the importance of visual design. I've
seen no one prove this is the case and further, I doubt it is the case
that it happens all that often in commercial website design. I don't
see it it as the primary trend when I browse the commercial web.

Does zipcode fail due to it's instructions being lower contrast than
some might prefer? I just don't buy that it does. You'd have to
quantify how many people it is a problem with, then come to an
agreement on what percentage of the user base it is a problem for
before it can be deemed a significant problem or failure. (In other
words, does it have to be a problem for 10% of the user base or 30%
before its a problem?) You have to find out if indeed that agreed upon
base can't read the instructions outright, or simply don't like the
low-contrast aspect of it? Etc etc etc.

> ***[Pabini] Nevertheless, industrial design is not visual design.
> However, I
> appreciate aesthetics in both.

You still haven't responded to the basis for my analogy: that the
visual aesthetics of the iPod does nothing to communicate the function
of the iPod, yet the iPod succeeds at both a visual level and an
interaction level. How can that possibly be if the primary goal of
visual design is to, as you say, communicate the function of the
product?

>>> ***[Pabini] After playing with it a bit, I ended up with a morass of
>>> text upon text upon text. Nothing was readable. The initial state was
>>> visually appealing, but everything went downhill from there.

>> ...The UI itself
>> doesn't necessarily create a "morass of text" and you admitted that
>> you
>> were the one who made the mess from a reasonable starting point. Does
>> that mean the UI fails? Because it gave you enough rope to hang
>> yourself? I just don't buy that.
>
> ***[Pabini] Give me a break, Andrei. I didn't do or say any such thing.

You said you ended up with a "morass of text upon text" out of the UI
to render it unreadable even though you started with something that was
acceptable. Ok... So what exactly did you do to get the morass of
unreadable text then? I assumed "playing with it a bit" meant
interacting with the UI. Moving objects with mouse, pressing a few
buttons, etc. When I do those tasks, I don't get the result you
described, hence my criticism.

> What you're saying doesn't reflect reality at all--any more than most
> of what
> Bush says. Perhaps you didn't fiddle with the UI as much as I did.

Pabini , don't get into the sort of argument with me that implies
questioning my competence. You and I both know that's useless. So let's
not don't go down that path. You said something happened in the UI and
disagreed with you. IMHO, you are not being clear about what it is that
you did to make the UI unreadable, so I made an assumption. My
apologies for a bad assumption then, but I have no idea what it is you
are talking abut unless you state it more clearly.

Further, I think it's easily possible to prove that not all users would
end of with a "morass of unreadable text." I'd be more than happy to
send you screenshots of what happens when I use the system for example.

Again, I'll grant the system has issues. But you are making sweeping
proclamations about it that I simply don't buy. Does that mean I'm the
one living in a false reality here? I don't think so.

> Did you click that button? That's what caused the problem.

What button? Specifically? What are you referring to or talking about?

> It quickly became
> impossible to manipulate things so they didn't overlap--layer upon
> layer.
> Whether this was because of a bug in the software or by design I
> cannot say,
> but it was a mess of its own making.

I have no ida what you are referring to. When I move the objects
around, I move them so they don't overlap each other. When I try the
various animation options, the system does a reasonable job of moving
things out of the way. There are times when the system has so much
data, text overlaps a little too much. Sure. But when I *move* text --
which is what I thought you meant when you said "playing with the
system" as I don't consider pressing whatever button you are referring
to as playing -- I don't overlap the text. I move it so things are out
of the way. So I have no idea how you arrive at unreadable text when
the first thing I do with the system is make it more readable than the
way it started.

> ***[Pabini] Sorry I can't send you a screen shot. You'd have to eat
> those
> words.

Why can't you send me a screenshot? Why is it important for me to eat
my words? Is this going to get personal? We can certainly discuss each
other's competence openly if you like, although I'm sure it would bore
others.

Andrei

31 Oct 2004 - 10:47pm
pabini
2004

> On Oct 31, 2004, at 5:43 PM, Pabini Gabriel-Petit wrote:
>
> > ***[Pabini] Since the functionality isn't very usable, because the page
> > doesn't do a good job of visually communicating its functionality, the
> > instructions are key, but are hard to read, which was the main point I
> > was
> > making. You've often made the point that people neglect visual design
> > and
> > that the quality of user interfaces is diminished thereby. I'd agree,
> > but
> > Web designers often overemphasize visual design at the expense of
> > usability
> > and readability, and that's at least as big a problem.

Andrei Herasimchuk wrote:
> I disagree with you that a majority of "web designers often
> overemphasize visual design at the expense of usability." I think
> that's a strawman argument that has done nothing but allow too many in
> the field to avoid the issue on the importance of visual design. I've
> seen no one prove this is the case and further, I doubt it is the case
> that it happens all that often in commercial website design. I don't
> see it it as the primary trend when I browse the commercial web.

***[Pabini] You do seem to have a knack for putting words into people's
mouths that they never would have used themselves. I never said "a majority"
of Web designers are guilty of doing this, but it does occur much too
frequently. Either extreme--overemphasis or underemphasis on visual
design--is bad. I'm not one who devalues or ignores the importance of visual
design, so maybe you should go have that argument with someone else.

> Does zipcode fail due to it's instructions being lower contrast than
> some might prefer? I just don't buy that it does. You'd have to
> quantify how many people it is a problem with, then come to an
> agreement on what percentage of the user base it is a problem for
> before it can be deemed a significant problem or failure. (In other
> words, does it have to be a problem for 10% of the user base or 30%
> before its a problem?) You have to find out if indeed that agreed upon
> base can't read the instructions outright, or simply don't like the
> low-contrast aspect of it? Etc etc etc.

***[Pabini] That isn't its only point of failure. I never said it was.
However, anything that *unnecessarily* causes problems for *any* portion of
the user base should be intolerable. There's no trade-off here. It was
merely the designer's choice.

> > ***[Pabini] Nevertheless, industrial design is not visual design.
> > However, I
> > appreciate aesthetics in both.
>
> You still haven't responded to the basis for my analogy: that the
> visual aesthetics of the iPod does nothing to communicate the function
> of the iPod, yet the iPod succeeds at both a visual level and an
> interaction level. How can that possibly be if the primary goal of
> visual design is to, as you say, communicate the function of the
> product?

***[Pabini] Visual aesthetics that don't distract are helpful in letting
users focus on function. Does the visual appeal of an object itself matter?
Of course. I never said that communicating function was the "primary goal of
visual design," just that, if a Web page fails to communicate its
functionality visually, it's a failure. Visual design can contribute many
things.

> >>> ***[Pabini] After playing with it a bit, I ended up with a morass of
> >>> text upon text upon text. Nothing was readable. The initial state was
> >>> visually appealing, but everything went downhill from there.
>
> >> ...The UI itself
> >> doesn't necessarily create a "morass of text" and you admitted that
> >> you
> >> were the one who made the mess from a reasonable starting point. Does
> >> that mean the UI fails? Because it gave you enough rope to hang
> >> yourself? I just don't buy that.
> >
> > ***[Pabini] Give me a break, Andrei. I didn't do or say any such thing.
>
> You said you ended up with a "morass of text upon text" out of the UI
> to render it unreadable even though you started with something that was
> acceptable. Ok... So what exactly did you do to get the morass of
> unreadable text then? I assumed "playing with it a bit" meant
> interacting with the UI. Moving objects with mouse, pressing a few
> buttons, etc. When I do those tasks, I don't get the result you
> described, hence my criticism.

***[Pabini] Clicked the button. Perhaps it is a bug then, but it's a very
strange bug.

> > What you're saying doesn't reflect reality at all--any more than most
> > of what
> > Bush says. Perhaps you didn't fiddle with the UI as much as I did.
>
> Pabini , don't get into the sort of argument with me that implies
> questioning my competence. You and I both know that's useless. So let's
> not don't go down that path. You said something happened in the UI and
> disagreed with you. IMHO, you are not being clear about what it is that
> you did to make the UI unreadable, so I made an assumption. My
> apologies for a bad assumption then, but I have no idea what it is you
> are talking abut unless you state it more clearly.

***[Pabini] How you got there, I have no idea. I never said anything meant
to imply that you're incompetent. Being awfully sensitive, aren't you?
Having a bad day? What I meant to communicate was that I agreed with your
arguments about the importance of visual design, but didn't think you chose
examples that supported your arguments very well. Why disagree with someone
so violently when you can't see what I'm seeing on my screen? If I was
seeing something different from what you're seeing, there was no way for me
to know that. Short of sending you a screen shot, which I can't do on this
list, I think I described what I saw well enough. Computers being what they
are, it's not a fair assumption on your part that I did something to create
the mess on my screen and certainly not to deny that it occurred. I'll
accept your apology though.

> Further, I think it's easily possible to prove that not all users would
> end of with a "morass of unreadable text." I'd be more than happy to
> send you screenshots of what happens when I use the system for example.

***[Pabini] I don't care enough about this to warrant your making the effort
to do that.
>
> Again, I'll grant the system has issues. But you are making sweeping
> proclamations about it that I simply don't buy. Does that mean I'm the
> one living in a false reality here? I don't think so.

***[Pabini] Your false reality was your sense of what I was saying. That's
all. Much too much projection going on.

> > Did you click that button? That's what caused the problem.
>
> What button? Specifically? What are you referring to or talking about?

***[Pabini] I don't recall what the button's label was, nor am I now going
to dig through all those email messages to find the link to the page again
and try to recreate what happened.
>
> > It quickly became
> > impossible to manipulate things so they didn't overlap--layer upon
> > layer.
> > Whether this was because of a bug in the software or by design I
> > cannot say,
> > but it was a mess of its own making.
>
> I have no ida what you are referring to. When I move the objects
> around, I move them so they don't overlap each other. When I try the
> various animation options, the system does a reasonable job of moving
> things out of the way. There are times when the system has so much
> data, text overlaps a little too much. Sure. But when I *move* text --
> which is what I thought you meant when you said "playing with the
> system" as I don't consider pressing whatever button you are referring
> to as playing -- I don't overlap the text. I move it so things are out
> of the way. So I have no idea how you arrive at unreadable text when
> the first thing I do with the system is make it more readable than the
> way it started.

***[Pabini] Glad it works properly for you then. It didn't for me.

> > ***[Pabini] Sorry I can't send you a screen shot. You'd have to eat
> > those
> > words.
>
> Why can't you send me a screenshot? Why is it important for me to eat
> my words? Is this going to get personal? We can certainly discuss each
> other's competence openly if you like, although I'm sure it would bore
> others.

***[Pabini] Because I didn't take one. This already got personal when you
first replied to my comments. You chose not to give any credence to what I
said. Your tone was very dismissive. You were basically denying the reality
of what I was seeing on my screen and chose to blame the user (me) for what
was appearing on the screen. I didn't have nearly that much control over it.
Since you and I have never met, and you're completely unfamiliar with my
work, that would be a bit stupid, wouldn't it? How can you possibly
interpret someone's disagreeing with you on one point among many as
questioning your competence?

31 Oct 2004 - 11:47pm
Andrei Herasimchuk
2004

On Oct 31, 2004, at 7:47 PM, Pabini Gabriel-Petit wrote:

> ***[Pabini] You do seem to have a knack for putting words into people's
> mouths that they never would have used themselves. I never said "a
> majority"
> of Web designers are guilty of doing this, but it does occur much too
> frequently.

You said, "You've often made the point that people neglect visual
design and that the quality of user interfaces is diminished thereby.
You've often made the point that people neglect visual design and that
the quality of user interfaces is diminished thereby. I'd agree, but
Web designers often overemphasize visual design at the expense of
usability and readability, and that's at least as big a problem."

You used no qualifier when referring to web designers., an obvious
generalization. Further, you used the word "often" with regard to
theior actions. What are we supposed to think you meant if that does
not constitute a majority?

> However, anything that *unnecessarily* causes problems for *any*
> portion of
> the user base should be intolerable. There's no trade-off here. It was
> merely the designer's choice.

Again, that's your opinion. I just don't share that opinion. I have no
issue with making certain things "intolerable" for certain people if I
feel the design succeeds on certain levels. People like to use the word
"intolerable" when tit's usually nothing more than a nuisance. As
for"unnecessarily that's just a subjective point of view. So who is
right when it comes to determining what's "unnecessary?"

In the Fry example, I'm nearly positive the instructions are lower
contrast so as not to battle visually with the data and the UII itself.
Is that "unnecessary?"

> I never said that communicating function was the "primary goal of
> visual design," just that, if a Web page fails to communicate its
> functionality visually, it's a failure. Visual design can contribute
> many
> things.

You said, "If there is functionality on a page (There always is if
navigation is present.), the page isn't successful unless it
effectively communicates that functionality."

So if someone were interested in "success," (which most good designers
I know are very much interested in almost all of the time) wouldn't
they make it their primary goal to make the visual design communicate
functionality? How is that not making that a primary focus? Let me put
it this way... What can constitute success if the visual design does
not communicate functionality?

>> Ok... So what exactly did you do to get the morass of
>> unreadable text then? I assumed "playing with it a bit" meant
>> interacting with the UI. Moving objects with mouse, pressing a few
>> buttons, etc. When I do those tasks, I don't get the result you
>> described, hence my criticism.
>
> ***[Pabini] Clicked the button. Perhaps it is a bug then, but it's a
> very
> strange bug.

What button? There are like three buttons and a menu on the UI. Which
one re you referring to? (You stated the initial state of the UI was
readable, so there's some button you clicked that made it not readable.
Which one then?)

> ***[Pabini] How you got there, I have no idea. I never said anything
> meant
> to imply that you're incompetent. Being awfully sensitive, aren't you?

Pabini, you followed a comment that what I said doesn't reflect reality
(a la my remark about comparing Nielsen to Bush) with a comment that
"perhaps" I didn't fiddle with the UI as much as you did. Try reading
your statement again. I'm not being sensitive. I'm telling you that
sort of slant on the discussion seems pointless an unnecessary to me.

> Having a bad day?

I could say the same of your comments as well. Let's just move on...

> What I meant to communicate was that I agreed with your
> arguments about the importance of visual design, but didn't think you
> chose
> examples that supported your arguments very well.

I know. I obviously disagree with you.

> Why disagree with someone so violently when you can't see what I'm
> seeing on my screen?

Because of the manner in which you presented you arguments. If I had
agreed with you, I wouldn't have said anyhting. And I'm not disagreeing
violently. I'm just disagreeing. I will be more than happy to
acknowledge your points, but I'm of the opinion you are really not
making them effectively. Maybe you are... maybe it is just me.

> If I was seeing something different from what you're seeing, there was
> no way for me
> to know that. Short of sending you a screen shot, which I can't do on
> this
> list, I think I described what I saw well enough.

You can always email it me.

> ***[Pabini] Your false reality was your sense of what I was saying.
> That's
> all. Much too much projection going on.

I guess you take no responsibility for using phrases like, "I'd agree,
but Web designers often overemphasize visual design at the expense of
usability and readability, and that's at least as big a problem."

Or absolute, sweeping phrases like, "If there is functionality on a
page (There always is if navigation is present.), the page isn't
successful unless it effectively communicates that functionality."

Or vague sentences like, "Did you click that button? That's what caused
the problem."

Laying the blame on me having a "false reality" seems like sidestepping
the issue. You said those things (and more) and I'm calling them into
question. That's what's going on here.

> ***[Pabini] I don't recall what the button's label was, nor am I now
> going
> to dig through all those email messages to find the link to the page
> again
> and try to recreate what happened.

http://www.marumushi.com/apps/socialcircles/

And until you do, I'll just disagree with you on this point. Fair
enough?

> You were basically denying the reality
> of what I was seeing on my screen and chose to blame the user (me) for
> what
> was appearing on the screen.

I was questioning your assertion that since *you* had a bad reaction to
the UI examples I provided, they was therefore flawed in some
fundamental way. And i was questioning your statements about the role
of visual design. (You said, "Visual design isn't successful unless it
effectively communicates how to use the functionality on the page.
Social Circles is more visually appealing, but not wholly satisfying.
It's not very useful or usable.")

You made sweeping statements about the UI and the role of visual design
as some sort of fact. When I questioned that, you seemed to have taken
it personally. My apologies if you did, but I also don't feel I was
out of line in questioning your broad, sweeping statements about the
purpose of visual design or how it was used in the examples I provided.

Andrei

1 Nov 2004 - 12:00am
Andrei Herasimchuk
2004

On Oct 31, 2004, at 3:29 PM, Kevin Cheng wrote:

> If they said something like "these things are completely unusable but
> that doesn't mean the interaction is useless, many of the problems
> were found to be stemmed from confusion with the distracting visuals"
> then that would be useful.

Certainly. Further, if they tested the UI for specific things that had
nothing to do with "user satisfaction" I would feel far more
comfortable. Again, the study that started this thread was trying to
measure things like "user satisfaction." Why? Given the product and its
obvious flaws that seems like an entirely pointless thing to measure or
research.

> So summarizing:
> 1) nobody can separate info/visual/interaction from a product. They're
> intertwined

IMO, yes.

> 2) given this, researchers must test knowing this and actively test
> all aspects instead of trying to isolate

If they want to measure things like overall success or user
satisfaction, yes.

> 3) the "obvious" is not always so and research exists partially for
> this reason

My choice of words with "obvious" is not accurate. The word I'm looking
for is more akin to "fundamentals." The sorts of things you would learn
in a 101 type course on [insert your topic here.] Products that used
for research in this capacity should at least adhere to fundamentals.

> And to Dave's point: time is VERY MUCH a factor in research. Any
> thoughts that they have all the time in the world to just build the
> perfect test tools is unrealistic.

My reaction to this is that often you only have a set amount of time to
write a thesis. Well, if you turn in that thesis in handwritten form,
with smudge marks and ink stains on the page, you'll fail. How is that
any different from taking the time to do the basic work to make
something look presentable for research purposes?

Andrei

1 Nov 2004 - 1:08am
Alain D. M. G. ...
2003

--- Andrei Herasimchuk <andrei at designbyfire.com> a écrit :

> My reaction to this is that often you only have a set amount of time
> to
> write a thesis. Well, if you turn in that thesis in handwritten form,
>
> with smudge marks and ink stains on the page, you'll fail. How is
> that
> any different from taking the time to do the basic work to make
> something look presentable for research purposes?
>

Why?

Because it is extremely easy to type a thesis in a computer and have it
print out without smudge marks or stains, compared to the time and
practice it takes to get the graphic elements right in an interface.

Even if you are doing absolutely nothing original in the graphic
segments of your interface research (innovating in other matters such
as input and output devices, or new visualizations with standard
commands) and even if you are doing your prototype in JAVA and using
the exquisitely well done JAVA look and feel guidelines (again I take
this opportunity to praise Don Gentner and his team) you have some
paingstakingly long work just getting all these elements together. It
is so much easier to put words and letters together in a word
processing program!

My good friend Michel has a daughter who was born with a natural sense
for drawing and illustration. Even before she reached the age of ten
and before she had any training she could take a fast look at an
airport scene, and upon returning home sketch out a full perspective
view of a tarmac with airplanes and support staff! At the other end of
the spectrum you have quite a few visually challenged individuals like
another good friend of mine who is a programming whiz, but has nearly
no spatial or visual feeling for things.

Yes, training can compensate for lack of natural ability but it can
never completely replace it. And training is nothing without the
experience to hold it up. And training and experience take time and
time and more time. I went back to try to do a PhD after working in
many domains and as others have noted, I noted that grad school means
an endless list of things to do (doing research, taking classes, giving
classes, preparing conferences as a speaker or as support staff, and
reading, reading, reading, reading) and no time to do them. So unless
you have graphic training and experience beforehand, you just cannot
squeeze it in at that point.

And no, you can't get help from other students in other faculties or
departments more concerned with visual things because a university is
not a trade school. They are also there to do research, not to get
technical results.

I have the impression from reading all your messages in the last days
that you, Andrei Herasimchuk, have been blessed with a greater than
average talent in graphic creation (and perhaps a bit of training, and
quite a lot of experience) and the visual domain in general, but thay
you are totally blind to the fact that most of the human race is not
your equal in these matters.

Alain Vaillancourt

__________________________________________________________
Lèche-vitrine ou lèche-écran ?
magasinage.yahoo.ca

1 Nov 2004 - 1:48am
Andrei Herasimchuk
2004

On Oct 31, 2004, at 10:08 PM, Alain D. M. G. Vaillancourt wrote:

> Because it is extremely easy to type a thesis in a computer and have it
> print out without smudge marks or stains, compared to the time and
> practice it takes to get the graphic elements right in an interface.

I'm sorry, I just buy that. It's easy and straight-forward to use
fundamental color theory, layout structure and typographic rules in UI
design. Heck, even just using the principles as laid out by Tufte in
Envisioning Information is probably enough, forget the details of type
and color for now.

> It is so much easier to put words and letters together in a word
> processing program!

Has it really come to this? That to use simple complimentary colors is
so difficult to follow when designing a UI? Or knowing what constitutes
good use of whitespace? That adjusting line spacing and line-length
creates hardship?

> Yes, training can compensate for lack of natural ability but it can
> never completely replace it.

I would argue that in graphic design, training or study can compensate
immensely. In art, I would agree with your concern more. But graphic
design is just as much about problem solving as pure artistic talent.
What we do with interaction design in our profession is far more akin
to graphic design than it is to art.

> And training is nothing without the
> experience to hold it up. And training and experience take time and
> time and more time.

Indeed it does. In fact, I'd say you'll be training and learning until
the day you die. No time to waste then, right?

> I have the impression from reading all your messages in the last days
> that you, Andrei Herasimchuk, have been blessed with a greater than
> average talent in graphic creation (and perhaps a bit of training, and
> quite a lot of experience) and the visual domain in general, but thay
> you are totally blind to the fact that most of the human race is not
> your equal in these matters.

My skills in visual design pale in comparison to the entirety of the
design blogosphere. I have no notion that I excel in this area at all
quite frankly. I have to work at it and I'm rarely satisfied with my
results. But before the work can occur, one has to acknowledge the
importance in seeking it. Anyone interested in interface design,
interaction design, or design related to the high-tech field needs to
seek it, IMHO. And those who do research on it, need to understand what
it is they are researching, and make sure they qualify what it is they
are studying.

Andrei

1 Nov 2004 - 6:51am
Dave Malouf
2005

There is one point in all this that I want to take on ...
Visual Design is a skill. It is a skill that uses a larger portion of one
side of the brain over the other (just a metaphor), so it means that to do
it well requires people to retrain their minds. Just like with programming,
playing music, playing tennis etc. All of these are learnable. Yes, there
will be some who can "master" it beyond what the rest of us can achieve. But
just to become mediocre at it if it is directly involved in your field, and
you are doing that field at the doctoral level, then you should do it.

To me it is more important than learning "German". Here in the US (I don't
know about others), you HAVE to learn a foreign language to do your PhD. So
most take German for some reason. It is more important b/c it is directly
related.

I think what makes it "hard" is that few people have figured out how to
teach this skill in such a way where people feel they learned the 3 elements
of Visual Design that are required to actually translate it into practice:
Theory, Craft (hand coordination), knowing what is right (eye coordination).
All of the things I mentioned above have a mixture of these. One of the main
reasons I have decided to go for a masters in Industrial Design is to
actually take on these issues. I know it would be hard for me to do this
without the discipline and focus of school, but that is because I don't
learn well otherwise. Other people have had huge success with just learning
from the blog space and doing the practice on their own. All depends on who
you are.

So to me the question isn't whehter or not you should do this, but when.

The last part is ... Where are the institutions that can help?

Ok, now the last part ... I believe a lot of this has to do with where you
are in your career and where you want to go. If you want to be an in-house
specialist and you don't mind working in teams and you can learn enough of
the theory and what looks right on your own so you can lead that team, all
is good. If you want to go out and become a guru level consultant where
people are expecting you to do everything, then of course you need to push
yourself. I think the difference here is a 5 year experienced lead IxD and a
15 year experienced IxD. I'll be damned if I hit that 15 year mark (only 5
years away) and still can't draw for example ... Now back to my homework. ;)
We can talk about why drawing is important in another thread at another
time, but lets just say it is about design process and communication.

-- dave

1 Nov 2004 - 10:07am
Coryndon Luxmoore
2004

Andrei said:
"Interface design is about three key areas: visual, information and
interaction. When *any* one of those areas is obviously compromised or
fundamentally flawed, then I claim the entire interface is compromised.
(In the example provided, all of the interfaces are obviously compromised at
the visual level.) You may test it and get marginal data on it, but to test
the "success" or "user satisfaction" of the interface is folly."

I don't disagree that interface design is comprised of these three elements
and for a successful design you must incorporate all of them. I agree that
for a professional to succeed they need to be able to blend all of these
elements.

As a interaction designer and a former industrial designer I have always
viewed this kind of test/research as breadboard prototypes. Since designers
rarely have the tools to effectively include all aspects of a proposed
design into a single prototype (short of building the solution) we
frequently need to resort to breaking the prototypes down into visual
appearance models and functional/breadboard models

Testing each type will absolutely benefit the design process. Each will
answer questions that the designer has about a given design approach and
saves us significant time. We can measure the success of the prototypes with
users but we need to keep an eye out for elements that might skew the
results. It takes an experienced designer to be able to filter the noise and
be able to know when to try and fix the design or to start over.

Given the choice of having to spend a lot of time building a perfect
prototype that incorporates all three interaction design elements or
building and testing a NUMBER of crude prototypes that fail in one of the
three areas for the same effort. I will gladly take the second option.

Now it is another question why most of this hci "research" tends to focus on
the function vs. the appearance....

--Coryndon

1 Nov 2004 - 1:45pm
Andrei Herasimchuk
2004

On Nov 1, 2004, at 7:07 AM, Coryndon Luxmoore wrote:

> Given the choice of having to spend a lot of time building a perfect
> prototype that incorporates all three interaction design elements or
> building and testing a NUMBER of crude prototypes that fail in one of
> the
> three areas for the same effort. I will gladly take the second option.

Fair enough, and I agree. If you qualify what you build and what you
study, then definitely the results and research data are very useful.
In my experience, I see too many research papers (and corporate testing
research too btw) study with examples that are not complete. But they
test the entirety of the project. If they did what you are suggesting,
I would not lodge the complain I diid.

Andrei

1 Nov 2004 - 2:43pm
Listera
2004

David Heller:

> I'll be damned if I hit that 15 year mark (only 5
> years away) and still can't draw for example ...

Drawing can be taught. And it's fun learning, as you usually get to
speed-draw naked people.:-)

But you're right, it's very difficult to do IA/UI/UX without constantly
drawing in front of clients/stakeholders. I can't remember the last meeting
where I didn't draw at least a dozen sketches on the board.

Ziya
Nullius in Verba

Syndicate content Get the feed