[thelist] IA - Visio Tricks - Menu Templates

Sam-I-Am sam at sam-i-am.com
Fri Nov 2 12:22:19 CST 2001

I think Hassan's point is that if you know what you what you are doing,
and to a degree, sure you can build your prototype in the final
And sometimes that is true - as a developer it certainly helps you
understand any problems that might arise from a particular architecture,
spot opportunities.. and sometimes things *dont* change so much and you
can just polish and publish

I have been bitten many times though by prototypes looking too "real".
It's difficult to justigy the further 50+% fee and n months for the
build phase if the client is looking at something that to him looks
built already. Or, as happened recently to me, while the client contact
fully understood what we'd done, he knew that it would get people *too*
excited if he presented it to his superiors and he needed to build a
careful case for the hires he'd have to make, the budget he'd need, the
time the project would take etc.. 

It also narrows the solution space prematurely - assumptions made on
interaction and workflow in your protoype might never get evaluated. And
even if you as the prototype developer are quite happy to scrap it all
and start again... repeatedly, your team members and the client might
feel obliged to build on what you've done to avoid "wasting" good work. 

Also, while you (as a programmer) might be quite comfortable thinking in
the abstract, many people are not  - whiteboard work, paper prototypes,
etc. are all good tools to counter that brain-ache, and help those
people actively participate in the design.
(and those people, if we are honest, probably includes most of us) When
well done, you should be able to step back and grasp the problem and
understand where the solutions are going to come from. There's nothing
like big sheets of paper to do that! 

Me, I like to be able to walk around a problem.. the more tangible it
gets the greater my understanding and the better my ideas and solutions.
I can't get that from a web/html prototype. I want to keep the concept
seperate from the medium so I don't have to deal with that baggage of
conventions and issues just yet. You want to be able to think freely and
focus people on the *design* problem when designing. 
Just as when developing a function spec for a product, you want to avoid
thinking in a particular language or platform - you try to understand
what needs to happen, how it will work without necessarily saying we'll
use this syntax or this application

_all_ that said. I am also one of those people who thinks best working.
So I'm very interested in seeing a flexible and generic web-technology
based prototyping toolset. 
Mainly because output from the tools like Visio is so entirly un-useful
when it comes to doing functional prototypes. I should be able to import
a structure, a decision tree, section names and page titles to get me
started. As it is I typically end up rekeying everything from a print


> I would NOT prototype in any form of final "code," whatever that may be.
> It's way too detailed and specific for the task at hand.  And you WANT to be
> able to tear it down and start over quickly.

More information about the thelist mailing list