[Sigia-l] Pillars of Ergonomics

[Listera]

"Livia Labate" wrote:

> You talk about the importance of implementing things regardless of the why
> ("An unimplemented idea is...nothing") and say this is a matter of
> efficiency? What's the use of efficiency without efficacy?
 
Obviously, somehow we're not communicating properly on this.

Simple example: suppose you're tasked with designing an app that displays
several dozen high-res images from a DB. Let's stipulate for the sake of
this argument, that it's a Flash app. You can have a menu (on the left, of
course :-) that lists the pictures and a pane on the right that displays the
actual picture. Simple. If you sit in front of your Flash MX 2004 IDE, you
could complete this app in minutes. This is a straightforward and utterly
simple-to-use and simple-to-implement arrangement.

Or is it? If you don't know that Flash players earlier than v5 don't load
external JPEGs, you will have wasted a lot of time and money until the
fundamental flaw is discovered by some user. In this case, since preloading
all images up-front would probably be inadvisable, you may have to scrap the
notion of a Flash app altogether. If you knew the tool, Flash, it would have
taken you a mere 20 seconds to reject the notion and move on to other
options, rather than waste days on this blind alley. There's neither
efficiency nor efficacy in that.

Ideas are cheap. Even good ideas are cheap. But users don't get to use
unimplemented ideas. What's difficult is to implement an idea well. Can you
have a bad idea implemented well? Sure, look at Hollywood. Can you have a
good idea implemented badly? Of course, look at the web.

As I have written about this here a while ago, designers/IAs/analysts/etc
don't deal with raw data. Whether it's an API, SQL table, chunk of content,
a chart, videostream, etc., the stuff we deal with are at various levels of
abstraction well above raw data. We give shape, context and direction to
that data, before it gets to the user. In other words, we are the people who
provide the interface to data, which makes it useful and usable for the
customer. 

But information without interface is just data, and that's not our domain.
So I can't think of what we do in the abstract, in a vacuum, without the
interface. And, interface necessarily involves more tangible stuff, like
objects, widgets, languages, rendering rules, etc. And these have certain
limitations and opportunities. You'd be at a distinct disadvantage not
knowing the boundaries of the tools that you have to use to implement the
necessary interface.

> I'm surprised that you of all people would say that. What's the point of
> having information architects, business analysts, designers, etc, then, if
> we can totally rely on programmers to deliver great efficient solutions?

I'm not sure how this equates to relying on programmers to dictate
solutions. Actually, it's the opposite. It makes it efficient because you
don't have to rely on programmers at all to come up with implementable ideas
(remember unimplemented ideas are...nothing. :-)

> I'm just perplexed by your sudden enthusiasm with technology despite of
> results.

Know your enemies. :-) The principal reason why I can propose, defend and
get implemented good interfaces among a group of geeks is because I
thoroughly speak their language. I'm not about to let a programmer derail a
good idea from getting implemented because he can speak technology and I
can't. By the time he gets my prototype, most if not all the fundamental
technical issues will have already been vetted. Knowing the pros and cons of
implementation is the best insurance policy that guarantees good ideas
actually get implemented, i.e. users get to benefit from them.

----
Ziya

Design is the art of gradually applying constraints
until only one solution remains.