OLPC: Sugar Not So Sweet?

26 Dec 2007 - 1:40am
6 years ago
28 replies
3166 reads
Dan Saffer
2003

I'd be interested in hearing from interaction designers who have
played extensively with the new One Laptop Per Child UI ("Sugar").

I saw the first public demo of Sugar back in August at AP's UX Week:

<http://www.adaptivepath.com/blog/2007/08/14/uxweek2007-lisa-strausfeld-on-one-laptop-per-child/
>

but never really got to play with it. I was willing to keep an open
mind about the UI, even though it uses very different paradigms than
the ones we're all used to.

But then Christopher Fahey's critique:

<http://www.graphpaper.com/2007/12-23_challenge-if-you-cant-say-something-nice-about-olpc
>

"From what I’ve seen, the UI bears all the hallmarks of a user
interface disaster, a case study in designer-driven design. I don’t
understand why the whole UX world isn’t awash in skepticism over an OS
that looks all the world like a Microsoft BOB for the Wallpaper* set."

and Adam Greenfield's

<http://speedbird.wordpress.com/2007/12/22/twenty-four-hours-with-my-olpc/
>

"Despite its inclusion of some innovative and useful features, I find
the OLPC device’s Sugar operating system poorly integrated with
applications (here nicely dubbed “activities”), to the degree that it
may well be impossible to evaluate whether the underlying idea ever
had any merit. My first impression - and I reiterate, it’s only that -
is that many of the applications bundled with the device epitomize
everything that’s wrong with FLOSS user interfaces, even when the OS
itself has been created by professional information designers."

And more...

<http://issues-in-publishing.blogspot.com/2007/12/first-reactions-to-olpc.html
>

"Sugar is the operating system on the XO, and it, too, is very cool,
but it is slow, and not intuitive for the hardcore windows and mac
users. It is just not as advanced an operating system, and it is clear
that it was built by developers for developers."

I'd be curious to see how the UI (and the machine in general) are
working in the field. I know for fact that no generative user research
was done for these, but I wonder if any testing has been done since
then.

And how about some heuristic evaluations from this community?

Dan

Comments

26 Dec 2007 - 7:39am
Fred van Amstel
2005

I didn´t observed directly the XO in use, but I´m very skeptical about
expert reviewers from outside of the target community. If the
interface is different maybe that´s because the situation is
different. The XO is a socially oriented machine. Any fair evaluation
must be in a social context, not by a context unaware individual.

The XO was tested in many schools in the world. Here are videos of XO
in use in a brazilian classroom (Luciana de Abreu School - Porto
Alegre):
http://br.youtube.com/watch?v=w3XlQgqAvWQ
http://br.youtube.com/watch?v=fRpCmV5zHYo
http://br.youtube.com/watch?v=FMcw0PtYTaA

--
.
.{ Frederick van Amstel }. Curitiba ´´ PR
¶ ...''''''''''|| www.usabilidoido.com.br
.
ICQ 60424910 / MSN e Gtalk usabilidoido at gmail.com
\\...................

--
.
.{ Frederick van Amstel }. Curitiba ´´ PR
¶ ...''''''''''|| www.usabilidoido.com.br
.
ICQ 60424910 / MSN e Gtalk usabilidoido at gmail.com
\\...................

26 Dec 2007 - 1:48pm
Robert Hoekman, Jr.
2005

> I didn´t observed directly the XO in use, but I´m very skeptical about
> expert reviewers from outside of the target community. If the
> interface is different maybe that´s because the situation is
> different.

I haven't used it *extensively*, but since getting my hands on an XO last
week, I've had a few chances to spend an hour or so with it and check things
out. I agree it would be foolish to judge the social aspects of the OS
without being surrounded by other people using them so the social components
could be examined, but I disagree that the difference in situation should be
given as much leeway as you're giving it. Good design principles can be
applied regardless, and the XO is lacking in this area.

For example, the #1 thing I've noticed with Sugar is the same #1 thing I've
noticed in most web apps: an incredible lack of instructive design elements.
I think the people working on this OS and on the multitude of applications
that can be used to extend it are doing fantastic things - don't get me
wrong - but they're in desperate need of a skilled interface designer, at
the very least.

>From the very first moment, it's difficult to tell even how to *open* the
laptop. Once you turn it on and start trying to make sense of the
environment, most clicks are guesses. Few are educated guesses. There's
simply nothing - *anywhere* - that communicates how to make things work,
what they do, what they mean, etc. There's nothing instructive about the UI,
nothing self-evident. And the Getting Started guide at
www.laptopgiving.org/start is more of a marketing piece than a how-to. It
would convince you to buy the product, but does a poor job of telling you
how to use it.

You don't have to be an ethnographic design researcher to know that getting
users up to speed is essential for a good experience.

One can only hope that those who deliver these machines are also sticking
around to train their recipients.

It's clearly stated on the OLPC site that they do not offer tech support
because they hope users will become savvy enough with the XO to fix issues
on their own. I think this is ... well, insane. Including an app for writing
basic Python does not, by any means, ensure that the kids on the other end
of the OLPC initiative will ever understand it.

So far, a huge number of design decisions have clearly been made by the
proverbial developer. And he's not just *any* developer, he's a *UNIX*
developer. He is - let's face it - in his own league of geekiness.

On another note, can I ask why you're exploring this, Dan? Are you
collecting info for a paper or some other purpose, or are you just curious?

-r-

26 Dec 2007 - 2:22pm
Fred van Amstel
2005

Sugar is presented as a challenge to the children: decipher-me or give
up. You can see in the videos I posted when they discover something
new the first thing they do is to show up to their peers. Soon there
are some kids that explore more the possibilites of the systems and
serve the others with support.

> It's clearly stated on the OLPC site that they do not offer tech support
> because they hope users will become savvy enough with the XO to fix issues
> on their own. I think this is ... well, insane. Including an app for writing
> basic Python does not, by any means, ensure that the kids on the other end
> of the OLPC initiative will ever understand it.

They are confident that their system will turn every child into
hackers. I cannot agree on that, but the system is clearly a hacker
artifact and as so, can stimulate hackering activities.

In this case, success is not a mean of technology adoption, but of
technology appropriation. If these people, in their culture and place
can appropriate this techonology to do something they find useful, so
it can be considered sucessful. Let users judge it.

--
.
.{ Frederick van Amstel }. Curitiba ´´ PR
¶ ...''''''''''|| www.usabilidoido.com.br
.
ICQ 60424910 / MSN e Gtalk usabilidoido at gmail.com
\\...................

26 Dec 2007 - 2:37pm
lachica
2006

These are interesting commentaries. I think there is a lot of value in
looking at this operating system being open to 'different paradigms' and
very little value looking at it from the perspective of a hardcore windows
or mac user.

These computers are going to be used by children with little or no exposure
to personal computers. From everything I've read about the development of
Sugar it seems that it was built to support the tasks of these children.
Whether it's successful or not needs to be tested in the field rather than
the laboratory.

Sugar is also based on the faith that a community of users will develop. If
these communities develop then the operating system has much more potential
than it would for individual users. I am hoping that once it is set free in
the world some unpredicted things will happen.

Cheers,
Julie

On Dec 26, 2007 12:40 AM, Dan Saffer <dan at odannyboy.com> wrote:

> I'd be interested in hearing from interaction designers who have
> played extensively with the new One Laptop Per Child UI ("Sugar").
>
> I saw the first public demo of Sugar back in August at AP's UX Week:
>
> <
> http://www.adaptivepath.com/blog/2007/08/14/uxweek2007-lisa-strausfeld-on-one-laptop-per-child/
> >
>
> but never really got to play with it. I was willing to keep an open
> mind about the UI, even though it uses very different paradigms than
> the ones we're all used to.
>
> But then Christopher Fahey's critique:
>
> <
> http://www.graphpaper.com/2007/12-23_challenge-if-you-cant-say-something-nice-about-olpc
> >
>
> "From what I've seen, the UI bears all the hallmarks of a user
> interface disaster, a case study in designer-driven design. I don't
> understand why the whole UX world isn't awash in skepticism over an OS
> that looks all the world like a Microsoft BOB for the Wallpaper* set."
>
> and Adam Greenfield's
>
> <http://speedbird.wordpress.com/2007/12/22/twenty-four-hours-with-my-olpc/
> >
>
> "Despite its inclusion of some innovative and useful features, I find
> the OLPC device's Sugar operating system poorly integrated with
> applications (here nicely dubbed "activities"), to the degree that it
> may well be impossible to evaluate whether the underlying idea ever
> had any merit. My first impression - and I reiterate, it's only that -
> is that many of the applications bundled with the device epitomize
> everything that's wrong with FLOSS user interfaces, even when the OS
> itself has been created by professional information designers."
>
> And more...
>
> <
> http://issues-in-publishing.blogspot.com/2007/12/first-reactions-to-olpc.html
> >
>
> "Sugar is the operating system on the XO, and it, too, is very cool,
> but it is slow, and not intuitive for the hardcore windows and mac
> users. It is just not as advanced an operating system, and it is clear
> that it was built by developers for developers."
>
>
> I'd be curious to see how the UI (and the machine in general) are
> working in the field. I know for fact that no generative user research
> was done for these, but I wonder if any testing has been done since
> then.
>
> And how about some heuristic evaluations from this community?
>
> Dan
>
>
>
>
> ________________________________________________________________
> *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
>

26 Dec 2007 - 2:42pm
.pauric
2006

http://www.cnn.com/2007/TECH/ptech/12/25/onelaptop.onevillage.ap/index.html

" Take Kevin, the aspiring trumpet player.

Sitting in his dirt-floor kitchen as his mother cooks lunch, he draws
a soccer field on his XO, then erases it. Kevin plays a song by
"Caliente," his favorite combo, that he recorded off Arahuay's
single TV channel. He shows a reporter photos he took of him with his
3-year-old brother.

A bare light bulb hangs by a wire from the ceiling. A hen bobs around
the floor. There are no books in this two-room house. Kevin's parents
didn't get past the sixth grade.

Indeed, the laptop project also has adults in its sights.

Parents in Arahuay are asking Mendoza, the visiting psychologist,
what the Internet can do for them.

Among them is Charito Arrendondo, 39, who sheds brief tears of joy
when a reporter asks what the laptop belonging to ruddy-cheeked
Miluska -- the youngest of her six children -- has meant to her.
Miluska's father, it turns out, abandoned the family when she was 1.

"We never imagined having a computer," said Arrendondo, a cook.

Is she afraid to use the laptop, as is typical of many Arahuay
parents, about half of whom are illiterate?

"No, I like it. Sometimes when I'm alone and the kids are not
around I turn it on and poke around."

Arrendondo likes to play checkers on the laptop.

"It's also got chess, which I sort of know," she said, pausing
briefly.

"I'm going to learn." "

Robert: "It's clearly stated on the OLPC site that they do not
offer tech support because they hope users will become savvy enough
with the XO to fix issues on their own. I think this is ... well,
insane."

I'd like to think that there's a future generation of nerds, who
like me, got their start supporting & fixing their friend's
computers.
http://www.youtube.com/watch?v=CmYDgncMhXw
Frankly, I think throwing kids in at the deep end of maker/hacker
culture is a stroke of genius.

Robert: "So far, a huge number of design decisions have clearly been
made by the proverbial developer. "
I'm sure you're aware Alan Kay is a principal designer on the
project although I dont know the extent of his involvement. From the
interviews I've read there's method to his madness, I might not
necessarily agree with the result, but it's certainly not your
typical OSS engineering centric featurefest.

I would have to disagree with the thinking that Sugar fails because
it does not "communicates how to make things work"

Children learn best by exploring their world.

regards -pauric

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Posted from the new ixda.org
http://gamma.ixda.org/discuss?post=23928

26 Dec 2007 - 2:54pm
Dan Saffer
2003

On Dec 26, 2007, at 10:48 AM, Robert Hoekman, Jr. wrote:

>>
> On another note, can I ask why you're exploring this, Dan? Are you
> collecting info for a paper or some other purpose, or are you just
> curious?

I'm just curious. Adaptive Path bought us all one (two) for a holiday
gift but they haven't arrived yet. And then there was those two
negative posts from people whose opinions I respect that caught my eye.

But a paper/post/detailed critique from this community ("How to
Improve the OLPC UI") would be valuable and a useful service.

Dan

26 Dec 2007 - 3:34pm
Matt Nish-Lapidus
2007

I couldn't agree more with pauric on this one.

Most people I know who are leaders in technology, programming, and
even a lot of designers, got their start with computers as children.
When we got our first computer I was 7 and it was a c64. No
instructions, no GUI.. just the BASIC language and a lot of curiosity.
I hacked away at it until I figured it out, and I attribute a lot of
my success to that early beginning.

If these kids are given access to a hacker friendly system they will
learn how to make it do what they want. Pauric is entirely right,
children learn best by exploring and being "thrown into the deep end."
I'm sure a lot of us here have stories from our childhood about our
first computers and how we learned how to use them without
documentation or helpful interfaces.

On Wed, 26 Dec 2007 11:42:28, pauric <radiorental at gmail.com> wrote:
[snip]
> I'd like to think that there's a future generation of nerds, who
> like me, got their start supporting & fixing their friend's
> computers.
> http://www.youtube.com/watch?v=CmYDgncMhXw
> Frankly, I think throwing kids in at the deep end of maker/hacker
> culture is a stroke of genius.
[snip]
> Children learn best by exploring their world.
>
> regards -pauric

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

26 Dec 2007 - 4:04pm
.pauric
2006

Dan: "And how about some heuristic evaluations from this community?"

The impression I get from the reviews you quoted is that on a scale
of DOS to OS X, Sugar is a missed opportunity.

I would say, wrong scale, it cant be measured by conventional
standards. Of course there is no such thing as a perfect design but
unless the IxDA sponsors some Ethnographic research in the depths of
the Amazon I Think its a fools errand to judge the endeavor by
western standards.

If anything we should define a new set of heuristics as I'm having a
lot of trouble applying any of 10 principles to the context of a dirt
floor classroom of 50 kids, 1 teacher, where the only other piece of
technology is a light switch (at best)
http://www.useit.com/papers/heuristic/heuristic_list.html

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Posted from the new ixda.org
http://gamma.ixda.org/discuss?post=23928

26 Dec 2007 - 4:24pm
desiree mccrorey
2007

pauric <radiorental at gmail.com> wrote:
> I would have to disagree with the thinking that Sugar fails because
> it does not "communicates how to make things work"
>
> Children learn best by exploring their world.

Please take what I'm about to say with a grain of salt, because I've not had
the chance to actually play with an OLPC.

I've two real life incidences to relay about my niece when she was a young'n;
about 17 years ago.
First was when I installed Kid Pix on my Mac, in anticipation of her visit. She
was 3 yrs at the time. Because of age, I assumed she had no experience with
Macs or any other computer. So I had every intention of getting the thing up
and running, launching Kid Pix and showing her how to use it.

She needed absolutely no help from me. She knew how to turn the computer on,
locate the app (it wasn't iconified the desktop), launch it (double click, mind
you) and get busy. Clearly, she'd been introduced to a Mac and that app before,
but given her age, I was floored that she had no trouble.

A few years later, I purchased and installed some kid's math application on a
PC for her. She was 7 or 8 yrs. Once installation was complete and the app
launched, we were presented with a display of cartoon characters positioned
around a colorful room. I saw no menubar, no apparent icons, dialogs, message
area nor any other items I would have taken queue from for next steps.

I was a bit embarrassed not knowing how to continue. As I stepped away from the
computer keyboard to search for the blasted users manual, my niece took
control, studied the screen for a second then immediately started interacting,
successfully, mind you, with the app; an app she'd never seen before.

My big point of all this is I've learned to keep an extra wide open mind about
what will work for kids.

Does Sugar fail to communicate how things work? I suspect that like most
computers, there's loads of room for improvement with Sugar, but it may not
fail to the degree we experts think it fails. The primary target audience is
almost a different species! :D (well they seem that way to me sometimes)

desiree

Desiree McCrorey
UI Architect/Web Producer
www.healthline.com
desiredcreations.com

____________________________________________________________________________________
Never miss a thing. Make Yahoo your home page.
http://www.yahoo.com/r/hs

26 Dec 2007 - 4:50pm
Dan Saffer
2003

On Dec 26, 2007, at 1:04 PM, pauric wrote:

> Dan: "And how about some heuristic evaluations from this community?"
>
> The impression I get from the reviews you quoted is that on a scale
> of DOS to OS X, Sugar is a missed opportunity.

BTW, I never suggested we should use DOS/Windows/Mac/Whatever as the
means of evaluation. Instead, we should do as Robert suggested and
look to the universal principles of design, perhaps combined with some
basic good interaction design characteristics, and evaluate using those.

> I would say, wrong scale, it cant be measured by conventional
> standards. Of course there is no such thing as a perfect design but
> unless the IxDA sponsors some Ethnographic research in the depths of
> the Amazon I Think its a fools errand to judge the endeavor by
> western standards.

The counterargument is that it was designed by people in US with no
input from any end-user or research whatsoever, to my knowledge. Why
shouldn't we judge it by Western standards?

Side discussion: I've heard argued is that there is no such thing
anymore as "Western" and "Eastern" any longer. Not touching that one,
although feel free, I'm interested! :)

> If anything we should define a new set of heuristics as I'm having a
> lot of trouble applying any of 10 principles to the context of a dirt
> floor classroom of 50 kids, 1 teacher, where the only other piece of
> technology is a light switch (at best)
> http://www.useit.com/papers/heuristic/heuristic_list.html

This is an excellent idea. I know Nokia has done a lot of work in
emerging markets. Perhaps that would be the right place to start.

Dan

26 Dec 2007 - 4:59pm
Dan Saffer
2003

On Dec 26, 2007, at 12:34 PM, Matthew Nish-Lapidus wrote:

> If these kids are given access to a hacker friendly system they will
> learn how to make it do what they want. Pauric is entirely right,
> children learn best by exploring and being "thrown into the deep end."
> I'm sure a lot of us here have stories from our childhood about our
> first computers and how we learned how to use them without
> documentation or helpful interfaces.

This argument doesn't work, logically. To summarize: let's give kids
sub-standard equipment. They're smart. They'll figure it out.

Now remove kids and put users in there. Or adults. Why would we
tolerate a lesser product for kids, simply because they can figure it
out? Adults can figure out Windows 3.1, but I wouldn't want to inflict
it on them.

The question for me becomes, what is the ultimate goal of giving the
kids laptops? My guess is that it gives them access to information
that they otherwise wouldn't have. Does the current device and UI
support that goal sufficiently, without forcing users to modify it to
achieve that goal? I don't know the answer.

Dan

26 Dec 2007 - 5:00pm
Robert Hoekman, Jr.
2005

> I'd like to think that there's a future generation of nerds, who
> like me, got their start supporting & fixing their friend's
> computers.

Oh, I'm sure there is, but is the goal here to create a generation of kids
in developing countries who can hack a computer, or is to provide tools for
educating themselves in a variety of other ways?

I'm sure you're aware Alan Kay is a principal designer on the
> project although I dont know the extent of his involvement.

>From what I can tell, Alan Kay's name isn't typically followed by
"designer". Am I missing something? How does Alan Kay's experience make up
for the apparent lack of quality interaction design?

I would have to disagree with the thinking that Sugar fails because
> it does not "communicates how to make things work"
>
> Children learn best by exploring their world.

I agree with this, but again, the goal shouldn't be to force these kids to
dedicate all sorts of attention to the tool, but on the content to which the
tool provides access.

-r-

26 Dec 2007 - 5:13pm
Robert Hoekman, Jr.
2005

> If anything we should define a new set of heuristics as I'm having a
> lot of trouble applying any of 10 principles to the context of a dirt
> floor classroom of 50 kids, 1 teacher, where the only other piece of
> technology is a light switch (at best)
> http://www.useit.com/papers/heuristic/heuristic_list.html

Seriously? These kids are in a different culture, not a different species.

They live in a different world, for sure, but their brains are still human,
and human behavior is what led Nielsen to this list of usability heuristics.
He's even stated elsewhere (
http://www.useit.com/alertbox/user-test-locations.html) that it doesn't
matter where you perform usability tests, they generally have the same
result regardless.

I'm not sure I've ever disagreed with you this sharply, Pauric, so I'm
hoping I just don't get what you're saying.

-r-

26 Dec 2007 - 6:16pm
.pauric
2006

Robert: "The goal shouldn't be to force these kids to dedicate all
sorts of attention to the tool, but on the content to which the tool
provides access"

I see this missing support mechanism as a challenge turned in to an
opportunity. I agree with your risk assessment and maybe have a more
optimistic view of the potential outcome. A purely academic approach
focusing on the content layer will not address the needs of the
do-ers
among us who understand through taking tools apart to see how they
work;

I hear, and I forget.
I see, and I remember.
I do, and I understand. - Confucius

Dan:"My guess is that it gives them access to information that they
otherwise wouldn't have."
Included is access to the basic concepts of information technology,
its more than just the content on the screen.

Robert: "From what I can tell, Alan Kay's name isn't typically
followed by "designer"."

Actually, he might be more in tune with the project at hand than you
might have realised:
http://www.vpri.org/
And I -strongly- urge you to watch this through..
http://video.google.com/videoplay?docid=-1109203988787201616

Robert:"How does Alan Kay's experience make up for the apparent
lack
of quality interaction design? "
I cant answer that question, but I ask you to think about the bigger
picture here, the context of the endeavor. I installed Sugar on my
mac and based on that I decided to put only $200 down for 1 laptop to
be sent off. I think its very unsuitable for western goals and would
bet yours will sit collecting dust in 6 months if not sooner.

"They live in a different world, for sure, but their brains are
still
human, and human behavior is what led Nielsen to this list of
usability heuristics."

I dont think Nielsen gets 'social', he's very goal-quantitate
result
driven and maybe not the best yardstick for the mushy-human-chaos
stuff.

As I interpret Nielsen's "Location is Irrelevant for Usability
Studies" I conclude that its true as long as there's a baseline in
collective understanding of technology. Remove any preceding
exposure
and the methodology falls apart. Take his Parking Meter example from
the linked page. In this case there is no concept of meter, parking
or even car - its as much use as providing feedback on the "Help and
documentation" that comes with negotiating Peruvian border control.

"so I'm hoping I just don't get what you're saying."
nope, I think you got me right and we have a different view on the
matter, but thats what makes for great debate and learning. I will
accept that I might have an optimistic view of the potential. But, I
also think that Sugar fails in preparing kids for practical
applications... however, spreadsheets dont engage kid's
imaginations.

regards - pauric

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Posted from the new ixda.org
http://gamma.ixda.org/discuss?post=23928

26 Dec 2007 - 6:32pm
.pauric
2006

Robert, I said "however, spreadsheets dont engage kid's
imaginations." and I'll take that back, I appreciate you're not
talking about a UI interaction that suits the needs and desires of a
developed country.

However, I have little trouble imagining being 5 again and being
given an XO. Could the UI design be better - undoubtedly. Would a
finely tuned Sugar with web 2.0 corners and feedback at every corner
made my wide eyed learning any more efficient in my Mongolian Yurt?
questionable.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Posted from the new ixda.org
http://gamma.ixda.org/discuss?post=23928

26 Dec 2007 - 7:25pm
Matt Nish-Lapidus
2007

I'm not saying we should give anybody substandard equipment, I don't
necessarily think Sugar is substandard (i haven't had enough time to
fully evaluate it yet). All I'm saying, like pauric, is that
sometimes things that seem unintuitive to adults (or at least adults
already familiar with a specific type of computing) are not so for
children.

I also believe that children shouldn't really be led or instructed as
much as some people do.. If, as pauric said, the OLPC is intended to
excourage exploration and thusly has a UI that is slightly obtuse,
that might just be the right approach.

If the UI hinders them from doing what they need to do it would be a
failure.... if it opens up other possibilities for deeper exploration
and hacking then it could be a huge success.

You're right, just because we -can- figure it out doesn't mean we
should have to. But that's not really what i intended to say.... I
just wanted to put forth the idea that exploratory interfaces could
work, depending on the goals. I see the OLPC program as more than
giving people access to information, it's also giving them access to
new technology and skills, building the system to encourage hacking
seems like a good idea to me.

On Dec 26, 2007 4:59 PM, Dan Saffer <dan at odannyboy.com> wrote:
>
> On Dec 26, 2007, at 12:34 PM, Matthew Nish-Lapidus wrote:
>
> > If these kids are given access to a hacker friendly system they will
> > learn how to make it do what they want. Pauric is entirely right,
> > children learn best by exploring and being "thrown into the deep end."
> > I'm sure a lot of us here have stories from our childhood about our
> > first computers and how we learned how to use them without
> > documentation or helpful interfaces.
>
> This argument doesn't work, logically. To summarize: let's give kids
> sub-standard equipment. They're smart. They'll figure it out.
>
> Now remove kids and put users in there. Or adults. Why would we
> tolerate a lesser product for kids, simply because they can figure it
> out? Adults can figure out Windows 3.1, but I wouldn't want to inflict
> it on them.
>
> The question for me becomes, what is the ultimate goal of giving the
> kids laptops? My guess is that it gives them access to information
> that they otherwise wouldn't have. Does the current device and UI
> support that goal sufficiently, without forcing users to modify it to
> achieve that goal? I don't know the answer.
>
> Dan
>
>
>
> ________________________________________________________________
> *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
>

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

27 Dec 2007 - 2:12am
stauciuc
2006

You are right. Of course, we should also keep in mind that enabling and
encouraging hacking is one thing, forcing it on all users of a system in
order to achieve basic tasks is another.
(I'm not saying this system is doing that - I just understood from the
discussion so far that there might be a risk of it doing that)

On Dec 27, 2007 2:25 AM, Matthew Nish-Lapidus <mattnl at gmail.com> wrote:

> [...] building the system to encourage hacking
> seems like a good idea to me.
>
>
Sebi
--
Sergiu Sebastian Tauciuc
http://www.sergiutauciuc.ro/en/

27 Dec 2007 - 8:18am
Dave Malouf
2005

I would like to make a suggestion for this thread, as I see it facing some
unnecessary poles, as I see the posters and all of them are equally
interested in good interaction design and more importantly successful
product design.

We don't have equal access to the tool, nor are we quite frankly qualified
to truly give a heuristic analysis because in my mind our most standard
heuristics of HCI don't necessarily apply the same way here. Not that we
throw them out, but there are other forces at play that I would imagine few
of us have been designing for.

I know there are a few of us on this list from LeapFrog for example. I would
love their insights as both interaction and instructional designers on this
topic.

Just as it is really hard for someone who has been doing consumer products
for years fall directly into place doing enterprise products, this is
exponentially true about moving from tools of task towards tools of
learning.

I agree that there are a ton of problems with the eco-system design. I would
say that the main cause is that there is a problem with the very problem
statement itself. It is very US-centric, not in that it is imperialistic in
nature as much as it assumes a US cultural outlook within societies where
the same soci-political-economics are not at play. That is, "give a man a
fishing rod and he can save himself from starvation." This individualism
based on the power of capitalism doesn't even work in the US due to outside
forces such as racism, classism and sexism (to name a few). These forces on
the context of design are even stronger in the developing (or actually NOT
developing) nations of the world.

I would love to see someone who has the device slap it down in front of
their kids (if they have any) don't tell them a thing and see what happens
over the course of days and weeks and video tape the whole thing. There is
the child observation project in Berkeley which has a ton of cameras for
just such testing. It would be amazing to do that type of observational
research on the tool.

Short of that, if all you are going to give me in the critiques are theories
based on existing heuristics, I think we are way off-base here.

Is there anyone on the list who has worked in instructional design for
little people? done so in a "non"-linguistic format, relying purely on
symbology to communicate?

The closest project I know of like this for adults is the MotoFone by
Motorola for the developing world. Gabriel White gave a great presentation
of this work at the SF IxDA event that coincided with ICSID/IDSA World
Congress. A lot went into thinking about "illiterate" users.

Anyway, I hope we will learn a lot from the case studies. I wish they came
less from obvious PR sources however. I wonder if some student here,
wouldn't want to change their thesis and work on a Fulbright grant to
travel to Peru to find out what is happening on the ground? Anyone?

-- dave

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

27 Dec 2007 - 9:39am
Dan Saffer
2003

On Dec 27, 2007, at 5:18 AM, David Malouf wrote:

> We don't have equal access to the tool, nor are we quite frankly
> qualified
> to truly give a heuristic analysis because in my mind our most
> standard
> heuristics of HCI don't necessarily apply the same way here.

This begs the question: are there universal principles of
(interaction) design that apply regardless of the context or user
base? Are there standards that apply equally to CEOs in London and
impoverished children in sub-Saharan Africa?

My guess is: yes, there are some. Fitts' Law, Hick's Law, Tesler's
Law, the Poka-Yoke Principle, and probably a few others are fixed.

What these are would be an interesting list.

Dan

27 Dec 2007 - 10:21am
Dave Malouf
2005

hmmm? I actually don't agree.
why? b/c few if any of these laws came out of research on young
people. They are all tests done on adults that have gone through the
same level of socialization. And if my reading is correct all of
these laws are based on research only done in specific western
industrialized settings.

By no means are these cross-culural/sub-cultural studies and I do
question them. They are also not age-based studies and further they
are contextualized around a specific use-context of computational
areas.

I also think that like in any product design, interaction design and
HCI theory are not the only parts of the puzzle. I think that there
may be pieces of instructional design that may counter our basic
interpretations of standard HCI theory. I don't know for sure if any
of this is true, but when I look at my kids toys, they really don't
follow what I would expect in standard HCI or even ergonomics, but
they are quite successful.

-- dave

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Posted from the new ixda.org
http://www.ixda.org/discuss?post=23928

27 Dec 2007 - 10:24am
Matt Nish-Lapidus
2007

Yes, exactly. Sorry if my original post was a little confusing, this
is the point I was trying to communicate. :)

On Dec 27, 2007 2:12 AM, Sebi Tauciuc <stauciuc at gmail.com> wrote:
> You are right. Of course, we should also keep in mind that enabling and
> encouraging hacking is one thing, forcing it on all users of a system in
> order to achieve basic tasks is another.
> (I'm not saying this system is doing that - I just understood from the
> discussion so far that there might be a risk of it doing that)
>
> On Dec 27, 2007 2:25 AM, Matthew Nish-Lapidus <mattnl at gmail.com> wrote:
> > [...] building the system to encourage hacking
> >
> > seems like a good idea to me.
> >

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

27 Dec 2007 - 10:51am
Murli Nagasundaram
2007

This sounds so much like the Nature vs. Nurture debate <which ended in a
tie, BTW, if it ended at all> that one might as well state it up front to
stave off a lot of discharged steam. Practically every discipline centered
around people (psychology, sociology, economics, anthropology, medicine,
etc.) has engaged (and will continue to engage) in this debate, explicitly
or implicitly. Consider the decline of Skinnerian behaviorism after
Chomsky's work on linguistics. Despite the bad rap that Behaviorism has
got, numerous organizations and institutions (the military, for instance)
apply it successfully.
So let's just say that there are many universal principles of design and a
significant fraction of these are a consequence of common biological and
physical constraints. And further, to the extent that there are common
determinants of culture, there are additional principles of design that
could be considered kinda-sorta universal. And then there are probably some
idiosyncratic cultural issues for which no 'universal principles' will work
(at all times). But then this has always been the case in every realm of
human endeavor since time immemorial. The variable plasticity of the human
mind (across age groups, cultures, contexts, etc.) in its ability to adapt
to circumstances will render this discussion quite fruitless.

On the other hand, there's nothing so fun as a good fight! Fruitlessness
apart, it's likely to bring out some interesting issues.

-murli

On 12/27/07, Dan Saffer <dan at odannyboy.com> wrote:
>
>
> This begs the question: are there universal principles of
> (interaction) design that apply regardless of the context or user
> base? Are there standards that apply equally to CEOs in London and
> impoverished children in sub-Saharan Africa?
>
> My guess is: yes, there are some. Fitts' Law, Hick's Law, Tesler's
> Law, the Poka-Yoke Principle, and probably a few others are fixed.
>
> What these are would be an interesting list.
>
> Dan
>
>

27 Dec 2007 - 11:32am
DrWex
2006

I'm wondering if I'm the only one on this list who got an OLPC for his
kid and is just letting the kid go with it?

Context: my kid is seven, and definitely above average in conventional
intelligence, reading/literacy metrics, and is reasonably well
experienced with conventional PCs, including both browser- and
desktop-based age-appropriate software.

He adores his OLPC. Even though he hasn't managed to connect it up
yet with anyone else. Even though it's slow (he noticed that) and
sometimes frustrating (particularly opaque in respect to connecting to
conventional wireless networks). He's using it to take pictures of
everything. And play music. The other morning he was all eager to
show us the music video he and his (four year old) brother created
with it. I didn't even know it could do that.

He's found the simplistic math games (he's well beyond that in skill
level) and he's found the python interface (I think I managed to keep
a mostly straight face when he showed me that - I'm not even saying
the word "programming" to him). I'm doing my best to let him be him
and do things in as natural a way as he can (though I still insist he
do his schoolwork first).

So what do I make of this? I think we're using the wrong yardsticks.
Jakob be damned, usability is NOT universal. The difference between
zero and one is much much bigger than the difference between one and
two. Sure, I can think of ways that the UI could be clearer. The
how-to-get-started two-pager that came in the box was just downright
awful. I had to look online to figure out how to get it hooked up to
my uncle's Apple Airport; once it did that I showed my boy the news
feeds. Pictures he liked, the rest he was bored with in about 2
minutes. Who cares about the proficiency of the usability techniques
of the UI if the users are bored?

Given the constraints and the goals of the OLPC project I think
they've done a pretty good job and I'm betting they have a lot of
success with their target audience. Which is, I assure you, is
absolutely no one reading this mailing list. It's not even my
seven-year-old, though he's much closer to the target population than
I am. He's exploring and learning and finding things I didn't find
because that's his way of interacting. He doesn't want manuals -
wouldn't read them if he had them - and instructions just bore him.
He wants to poke and prod and play and if the machine gets into some
state he doesn't understand or can't figure out how to get out of then
he might ask for some help. Mostly though he just plays with it some
more until he figures it out. If he had other kids around him they'd
all be poking and prodding and sharing knowledge in ways that make our
usual one-person-one-machine notion of computer interaction outmoded.
The last time I saw a usability evaluation that had any relationship
to what I'm seeing now it was done by Amy Bruckman and her team on the
kids MOO she created. (Google MediaMOO and MOOSE Crossing if you want
to read about that.)

So while it's nice to go on about "standards" I think it's important
to realize we are using the wrong standards of measurement here. It's
like trying to measure the thrill of a motorcycle's acceleration with
a slide rule and calipers. We have expectations and associations that
just aren't relevant. My boy knows what standard PCs are like and
what they do and he doesn't find anything to complain about here
(trust me, he's not shy about complaining when he finds something that
doesn't meet his standards).

I can't wait until he has other OLPCs in his local neighborhood.

27 Dec 2007 - 11:55am
DrWex
2006

On Dec 27, 2007 8:18 AM, David Malouf <dave.ixd at gmail.com> wrote:
> I would love to see someone who has the device slap it down in front of
> their kids (if they have any) don't tell them a thing and see what happens
> over the course of days and weeks and video tape the whole thing. There is
> the child observation project in Berkeley which has a ton of cameras for
> just such testing. It would be amazing to do that type of observational
> research on the tool.

I think this would lead to interesting, but distorted, results.
Here's the point: nobody uses these things in isolation. They're used
in the contexts of classrooms and homes where adults are present.
They're given by people with knowledge who share some of that
knowledge to get the children started. They're shared with other
children, possibly around the world, who share a pool of knowledge. If
you strip away that context you miss the crucial success criteria.
It's like taking a car from the showroom floor and complaining it
doesn't go anywhere because you didn't put fuel in its tank.

Putting one (mistake) in front of a single child (mistake) without
saying anything (mistake) would produce such a compounding of errors
that you'd still likely get irrelevant results. Or perhaps you'd
reduplicate some of the work that's been done on childrens'
explorative learning methods, but is that what you want to find out?
(I think Allison Druin did great work in this area, see
http://www.cs.umd.edu/hcil/kiddesign/ for what she's up to now.)

I see now that I've really soaked in the Papert sauce, despite never
taking his class. Blame six years at the Media Lab. Peer learning and
peer teaching is such an important and fundamental part of my thought
processes that I forget others don't see things that way. I cited Amy
Bruckman earlier because she was a friend and I've thought her work
was brilliant for over a decade. You can also look up some of Mitch
Resnick's publications (he was her advisor and Papert's student) or
any of the other work in that group if you want to learn about these
ideas.

I'm not prepared to say there aren't universal principles of design;
Fitt's Law still describes the time to move a pointer to a target and
so on, regardless of whether the person moving the pointer is me, my
kid, or a person who's never seen a computer screen and pointing
device before. What I'm saying is that it's not important to the last
situation as it is to the first.

The original point of this thread was about usability and its
relationship to the (possible) success of the XO. My argument is that
the universal principles of design we operate by aren't relevant to
this success/failure, not that they don't exist. The challenge "say
something nice about the design" written by someone who hasn't used it
and isn't in the target audience is an expectation that the UI would
somehow conform to adult, western, single-user notions of goodness.
This is not an iPod - never was, never will be. The nicest thing I can
say about the design is that I have to fight my kid to get him to
relinquish it. I hear similar stories coming back from real-world OLPC
deployments.

If there's a usability metric that trumps that, I have yet to see it.

--Alan

27 Dec 2007 - 12:16pm
Dave Malouf
2005

Sorry, great correction ... It does require more than one child with many
cameras. ;)

my main point was let's get a study going on the observational level, and
not rely on heuristics. I've never been a fan of them anyway.

-- dave

On Dec 27, 2007 11:55 AM, Alan Wexelblat <awexelblat at gmail.com> wrote:

> On Dec 27, 2007 8:18 AM, David Malouf <dave.ixd at gmail.com> wrote:
> > I would love to see someone who has the device slap it down in front of
> > their kids (if they have any) don't tell them a thing and see what
> happens
> > over the course of days and weeks and video tape the whole thing. There
> is
> > the child observation project in Berkeley which has a ton of cameras for
> > just such testing. It would be amazing to do that type of observational
> > research on the tool.
>
> I think this would lead to interesting, but distorted, results.
> Here's the point: nobody uses these things in isolation. They're used
> in the contexts of classrooms and homes where adults are present.
> They're given by people with knowledge who share some of that
> knowledge to get the children started. They're shared with other
> children, possibly around the world, who share a pool of knowledge. If
> you strip away that context you miss the crucial success criteria.
> It's like taking a car from the showroom floor and complaining it
> doesn't go anywhere because you didn't put fuel in its tank.
>
>
>

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

27 Dec 2007 - 10:42pm
Juhan Sonin
2003

For a taste of Sugar:
* install it on your own machine (win/linux/mac/etc)
* build it (preferable) or run in emulation
* find + report bugs
* see goodies that need design help, and
* participate in the Sugar community.

But the laptop is the wrong discussion.

The cell-top is THE platform. 2.2 billion are currently in use on
planet earth with another billion slated to be sold this year. The
open cell phone is it: iphone-ish services, Android, Linux....

Kids in Nairobi are hacking cell services now from monitoring
community blood banks to minute-swapping services. Local teenagers and
college kids have already been shaping the cell-top service in Africa
(and many other continents). The cell phone is your bank, shopping
center, medical record, concierge, and more.

The OLPC is cute, cuddly, and a fabulous concept. The cell-top is
dominating the global service scene.

-Juhan

PS: Android (or a particular Android service) would be a perfect
vehicle for the IXDA community to jump onto as a Design project.

28 Dec 2007 - 11:36am
Robert Hoekman, Jr.
2005

(Reposted from an offlist response to Pauric. I accidentally left off the
IxDA email.)

A purely academic approach
> focusing on the content layer will not address the needs of the do-ers
> among us who understand through taking tools apart to see how they
> work;

I'm just not so convinced that learning to hack around a unique UNIX
implementation will help those kids once they graduate from OLPC to
something else. Sure, they could learn concepts they can apply in a myriad
of other ways, but operating under the *assumption* that the kids will learn
to hack the system in the first place is perhaps misguided.

I installed Sugar on my mac and [...]
>

I think you're not understanding the full context of the design. To really
see how Sugar works, you should check it out on an actual XO laptop. There
are keys on the keyboard I still can't explain. Buttons on the monitor that
have no explanation whatsoever. They all apparently correspond to some
action in Sugar, but I have yet to discover their connection after several
hours of usage.

You also have to see for yourself how difficult it is to use the trackpad
because the mouse is so jumpy and it's so easy to accidentally move it to
the edge of the screen (which produces an overlay of the desktop
navigation). Scrolling windows with it, hitting the right area to click,
etc., are tricky actions at best.

Granted, these things are designed for people with smaller hands than my
own, but the lack of smooth interaction means you start getting very careful
about what you click. You start calculating your moves more. As I started
doing this, the first thing I wished for was better instructive design so I
knew what I was doing and could stop guessing and making mistakes.

[...] based on that I decided to put only $200 down for 1 laptop to be sent
> off. I think its very unsuitable for western goals and would bet yours will
> sit collecting dust in 6 months if not sooner.

It's totally unsuitable for Western goals, and I'm definitely not trying to
say it is suitable. What I'm saying is that it doesn't matter who you are -
things are easier to learn if they're designed to be easily learnable. And
Sugar, at the moment, has a low level of learnability.

I dont think Nielsen gets 'social', he's very goal-quantitate result
> driven and maybe not the best yardstick for the mushy-human-chaos
> stuff.

I'm not usually a big fan myself, but isn't his work focused entirely around
research on how humans interact with computers? I don't want to invite the
guy to a dinner party, but he sounds like a pretty good yardstick to me.

As I interpret Nielsen's "Location is Irrelevant for Usability
> Studies" I conclude that its true as long as there's a baseline in
> collective understanding of technology. Remove any preceding exposure
> and the methodology falls apart.

Two quotes from that article:
1) "[...] people usually interpret the screen elements the same, no matter
where they live. What's easy in one city is just as easy in another city."
2) "First, if a parking meter is intended for a region that had never before
charged for parking, it might be a good idea to test with users who were
completely new to the parking meter concept. Novices would doubtlessly
encounter more usability problems than more experienced users."

Your inference that a baseline understanding of the technology in question
is required for the testing to be meaningful is in no way stated in that
article, and I wholly disagree that it's a requirement.

OLPC will be introducing computers into cultures that may have never even
seen one up close. Just like a parking metter would be more difficult in a
country that previously had none, a computer that lacks any instructive
design whatsoever is going to be really difficult to digest when you've
never even used a computer.

This logic is exactly why so many people get their first PC home and barely
ever learn to do anything but check email or browse the web. Without
instructive and self-evident interfaces, Help, video training, or something
else in place, there's simply no way to become an intermediate or advanced
user without a ton of guessing, which has a decent chance at failure every
single time you do it.

Following this logic, it's like saying that furniture that requires assembly
should come without instructions.

But, I also think that Sugar fails in preparing kids for practical
> applications... however, spreadsheets dont engage kid's imaginations.

Agreed. But this is exactly why I think Sugar needs better instructive
design. The intent is not to teach kids about UNIX, it's to give them access
to a wealth of knowledge so they can do almost anything. It's to change
their whole way of life, and to use education as a pathway out of poverty.

But Sugar commits the cardinal sin of software design: it constantly gets in
the way of its own purpose. Sure, Sugar developers are giving them
computers, but interface-wise, they're giving them the UNIX equivalent of
"Fisher Price meets Windows 3.1". It's both way too basic and way too
complicated.

Try using it on an XO and tell me if your opinions change.

-r-

28 Dec 2007 - 3:21pm
Jim Leftwich
2004

I've been playing with the OLPC XO for a couple of weeks and
understand some of the criticism being at it in this thread.
However, I think it's definitely a system that invites exploration
and experimentation. The Tam Tam music apps are pretty fun, and if
you've got more than one, the mesh networking is cool. Clearly the
apps (the whole system actually) seem like they're at a beta level,
but there's still lots to play with.

My teenager nephew and niece took a break from Guitar Hero and other
XBox games to play with it (which says something), and thought it was
fun. But clearly, it's a system that's aimed at young kids with
unlimited time and lots of friends with them. Just watching two
teenagers showing each other things they were discovering convinced
me that this will definitely work in its intended environment.

I did find the touchpad to be pretty tweaky, but quickly got the hang
of it. It also didn't take me very long to map out the various
functions on the keyboard.

I guess I'm a bit confused by some of the criticism here though,
comparing it to desktop Operating Systems and the kind of software
used by Western adults. Using the OLPC XO reminded me a lot more of
the great raw computing experiences that I had on Apple IIs in the
late 1970s, and I know that I was happy to spend kajillions of hours
hacking around on those.

If our sophisticated desktop software (not to mention games) are so
great, you'd think we'd be churning out a lot more programmers and
developers of unique and new things than we are. I think that the
raw peer-sharing-and-teaching model of the XO is pretty brilliant.

I'll be anxious to see how little computer performs once they begin
distributing them in larger numbers. My observations are that kids
are far more tenacious and far less judgemental than adults and
experts. I also don't think that every kid's experience will be
equal, and that's actually one of its benefits. Some will naturally
dig deeper and appreciate its open-ended nature. It wouldn't
surprise me if twenty to thirty years from now there are some
successful people who will say that their first formative experiences
on a computer were on the OLPC XO.

That's just the sense I get. I know that the race is on now to
compete with the OLPC XO with small, inexpensive computers running
Windows. I'd say it's all good, but I think that there's some
real advantages to having this little Linux system out there.

Jim

James Leftwich, IDSA
CXO - Chief Experience Officer
SeeqPod, Inc.
Emeryville, California
http://www.seeqpod.com

Orbit Interaction
Palo Alto, California
http://www.orbitnet.com

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Posted from the new ixda.org
http://www.ixda.org/discuss?post=23928

Syndicate content Get the feed