Conference DVDs: Design principles

[Wilhelm Sanke]

Just received the Conference DVDs along with some accompanying materials 
and I might give some feedback on the basis of my first and still 
incomplete impressions.

First of all, it is surely an interesting experience to see and watch 
some of the chosen Revolution protagonists in action. It adds to and 
possibly corrects the pictures that have slowly developed in your mind 
over time while reading their contributions on the Rev lists.  In one 
part of the disks Björnke von Gierke presents his "Animation Engine", 
but unfortunately never appears in person, he is only allowed to talk 
from the "off".    

As for the contents of the discs, I do not yet have a good overview, but 
it appeared to me that they also contain, at least at some places, very 
basic stuff, as for instance explaining the fundamental differences 
between PNG, JPEG, and GIF image formats (on a world-wide conference!). 
Maybe it would be a good idea for future conferences to sort the 
contents at least in two categories - low end stuff in the first four 
disks, high end for the other four - and then offer them separately to 
the corresponding target audiences.

The short lists on the back of the disk boxes surely help as a first 
orientation. I would appreciate if some more detailed information (sort 
of short summaries) would be provided on two extra sheet of papers, 
which could also contain more information about the conference, its 
proceedings outside of the presentations, about participants, results, 
perspectives etc..

The technical quality of the recordings is average at best, probably 
reflecting the small budget allocated here. Light conditions vary, audio 
quality is sometimes very poor and some parts are hardly understandable, 
also the displayed script examples and other written material are very 
difficult to read.-

As I could not participate personally in the conference, I was surprised 
and pleased to find out that at least one of my stacks had made it to 
Las Vegas as a sample stack. Scott Rossi used it in his "Multimedia II" 
presentation - the "II" in the section title probably indicating that 
this belongs in the advanced high-end stuff category - to demonstrate 
design principles (DVD 7, minutes 47 to 48).

On the one hand he praised my stack of  April 2007 ("Imagedata Toolkit 3 
Preview" <http://www.sanke.org/Software/ImagedataToolkitPreview3.zip>) 
for containing "a lot of powerful stuff" and "lots of features and 
functions" (soundtrack Rossi), on the other hand he used it to 
demonstrate some of the "Dont's" of interface design. As I did not have 
the opportunity to comment on design principles earlier, I think I am 
entitled to a few remarks here.

I agree with Scott that the stack - unfinished as it is - lends itself 
as an example to discuss questions of interface design. So would have 
other stacks.

Scott complains that he did not know where to look first on the card. 
This is somehow understandable as the stack has a rather complex 
structure, as one fact containing 542 controls on one card, having 452 
buttons (and an extra script library in a substack), of which 116 are 
visible and cluttering the card.
Taking the time to read the three introductory texts available on the 
card (two buttons on top center), surely would have helped.

The stack is not an exercise in interface design nor intended as a model 
for design, it is a "preview" stack (not a finished commercial product) 
and I wanted to get out this "powerful stuff" to the Rev community as we 
had already discussed details of its contents on the lists. In the 
meantime more than 60 other functions for imagedata processing have been 
added, and probably I will divide up the whole stuff in several separate 
stacks. One will be the already announced "Photo Patchworks" stack, 
another one has already been released ("Seamless Tiles 2" 
<http://www.sanke.org/Software/SeamlessTiles2.zip>-) which probably can 
be assessed more favorably for its surface design.

I myself have thought about various alternatives of structuring the 
interface, but this was not yet one of the essential points on my 
agenda. Among other solutions, I had thought about a strict menu 
structure (e.g. like in "PaintShop Pro"), but at least for the time 
being (and for time reasons) I decided against it, because I preferred 
the different functions - which can interact in various sequential ways 
to produce special effects - to be available right at your fingertips 
side by side, instead of digging deep into menu structures and start at 
the first level each time when you want to access a related function. 
Such considerations are also design principles, although not of the 
visual surface , but of internal structure and usability.

Because of the complexity of the stack you might think of it as an 
adventurous, creative enterprise, which requires some time to detect all 
of its features and possibilities, given the about 300 (?) functions 
that can additionally interact in various ways.

Another point Scott raised was why the 116 visible buttons on the card 
were not or insufficiently grouped according to their specific functions 
and that he could not understand the logic of the color codes. But a 
closer look would have revealed that indeed the majority of the buttons 
*are* ordered in functional groups, most indicated as such - surrounded 
by a field, having a title - others being in obvious neighborhood to 
each other. There are however some "stray" buttons, which had been added 
at a later time and placed at a free place (like button "Glassy 
Reflections" which should have been added to the "mirrors" group). There 
is also and certainly some logic in the coloring of the buttons and 
groups, but this may indeed be improved.

Scott has even invested the effort to produce and present an alternative 
surface for my stack in his presentation which is absolutely convincing 
in showing his design principles, but he put only about 20 percent of my 
functions on his card, probably  by that conveying a hint that such a 
complex stack should be really divided into several parts - like I 
mentioned above - but then conflicting with the other mentioned 
principle to have most of the essential functions available at your 
fingertips.

There are a number of aspects present in the theory and practice of 
design that have to be brought into a balance and often conflict with 
each other, among them the difficult task to streamline and organize the 
interaction of scripts, aspects of usability, of an esthetic surface and 
more. There can be, however, products with a shiny and polished 
interface and almost no contents, empty packages as it were.

Having said this, I remember a stack I used in a programming seminar 
some years ago, which certainly possesses a sound didactic step-wise 
structure, but is equipped with a shiny surface (with windowshapes, 
semi-transparent dialogs and buttons, and gradient colors) because I 
wanted to demonstrate at the same time some of the graphical 
possibilities of Revolution. It is a stack about developing 
"multiple-choice exercises". You can find it here: "Multiple-Choice 
Tutorial English" <http://www.sanke.org/Software/MultipleChoice.zip>.

Best regards,

Wilhelm Sanke