[Etoys] Re: A few notes on eToys interface

[Richard Karpinski]

Wild idea for geographic navigation.

On 2007, Apr 2, , at 9:00, Milan Zimmermann wrote:

>    - I think the largest potential issue is ease of eToys  
> navigation for
> someone who is running the system for the first time. I am thinking  
> how to
> modify the UI, without significant changes, to help a new user to  
> not "get
> lost" (for example by following a few projects from the cloud  
> menu). I think
> people's sense "not getting lost" is greatly satisfied when there  
> is a way to
> know "how to go back" - either one step back, or all the way to the
> beginning. I realize this is where the "Navigator->Prev" is used,  
> but it is
> not very obvious for a first time user, mostly because it's  
> contents (the
> Prev button) is "hidden". I am thinking if the following would help:

Background - Why geographic navigation works

When Jef Raskin, the father of Macintosh, proposed a zooming user  
interface for a hospital information system, it was built that way.  
It worked well, but the part that interests me now is that utter  
novices learned the system quickly. They became comfortable and  
competent in literally less than a single minute. Of course, computer  
experts took longer but they succeeded in less than two minutes. This  
was using a mouse with two buttons designated Zoom In and Zoom Out.

My analysis is that for millions of years, our ancestors succeeded in  
getting back to the nest. This imbues their living descendants with a  
natural talent for learning to navigate in a geographic world, even  
one where there is a lot of content in zoomable regions. I want to  
take advantage of that native ability in an even simpler system, one  
which uses the graphical input device alone with no keys or buttons  
at all.

Utility - Why auto-zooming is good for you

The first  thing to notice is that there is room for unbounded  
content if you can write it very small and zoom in to read what  
interests you. If you showed a shelf full of products, there would  
always be room for more detail, even a complete user manual, in a  
region no larger than ten or sixteen pixels on a side. The rollover  
event could be interpreted to auto-zoom into the region. The amount  
of zooming could be arranged in the construction of the zoom world to  
be just right for reading text or viewing images in that region.

Next, we can observe that rollover works the other way too. If you do  
the natural thing to return to where you were, you will cross the  
same region boundary in the other direction and be back there again.  
This is so simple that nearly everyone, I claim, will discover it on  
their own. They shouldn't have to, when less than a minute of  
training is required, but they could, especially if they are kids and  
not yet overly concerned with making no mistakes. Indeed, it works  
for the preliterate as well as those who can read and compose text.

Beyond zooming - Making the graphical input device alone do even more

I would go further and devise a mouse only (or touch pad only)  
gesture for select, and at that exact moment, make a pie menu appear  
to permit the second half of a noun-verb command to be chosen and  
activated. But I shall leave further discussion of such extensions  
for future occasions.

Initial introduction - How to make first use painless

In gaming arcades, there is often an attract mode which gives hints  
about what this game offers. For us, this includes several  
introductory eToy projects. With the auto-zooming interface, it would  
make some sense for the initial screen to have each of them in a  
region with a well defined border and let folks auto-zoom in to  
whichever strikes their fancy. Re-crossing that border, anywhere,  
naturally returns you to the original screen contents. Rapid learning  
can be expected.

Since there is always room for another help sheet or demonstration  
video on any screen, gentle training can always be at hand. Even a  
direct passage to the initial screen, despite violating the strict  
geographic navigation paradigm can be arranged to be awfully clear  
and quite effective. Any navigation problem which is discovered to be  
common can be addressed in such a way without much time or effort  
required to set it up.

Barriers - Why this scheme cannot be adopted at this time

We need to press forward and we already have much too much to do.  
This would be a radical change which we have no time to consider at  
present. It would take weeks to set this up just to see if it works.  
Nobody needs this. It won't work. People will fall into the auto- 
zooming regions and be lost and stuck. It is not now such things are  
done.

Richard Karpinski, World Class Nitpicker
148 Sequoia Circle, Santa Rosa, CA 95401
dick at cfcl.com  Home +1 707-546-6760   Cell +1 707-228-9716
http://cfcl.com/twiki/bin/view/Karpinski

ps Put (or leave) "nitpicker" in the subject line to get past my spam  
filters.