[Sigia-l] Intelligent signs at Microsoft

[Alexander Johannesen]

On 8/16/05, Dwayne King <pinpointlogic at gmail.com> wrote:
> 1) Everything we produce is dead at the moment of client sign-off.

I'd like to say that that depends totally on what sort of IA you're
doing. If you're doing static design, then sure it will be static by
its very nature. When I do IA, I design dynamic structures and
behaviour analysis, or, as some has dubbed, macro-IA, and this
documentation evolves even after I'm long gone.

> While requirements, use cases, technical specifications have found a
> way to live throughout the life of a project, on a universal scale we
> have not. On most projects, I hand off my documentation, it is rebuilt
> and my documents at that point are dead documents.

I think that IA's need to come up with better means of flowing with
the time, so I agree that traditional IA deliveries are rooted in
dead-documentation. A site will change over time, and so should the
IA, we all know this. The question comes down to how IA is intergrated
into the solutions, and if indeed there is only a printed-out Excel
spreadsheet it *will* be dead. This is why we need IA tools that can
integrate into the solutions more directly.

> 2)  Our process depends on the IA. While there are certainly nuances
> in the RUP, for the most part a requirements analysis is a
> requirements analysis, Use cases are Use cases, object diagrams are
> object diagrams, etc. etc. In IA a wireframe is whatever the IA
> perceives a wire frame to be (author of this note included). \

Maybe its time to revise deliveries again? For instance, in all my
projects I've integrated IA documentation from the GUI all the way
down to technical protocols, so that a GUI fiddler and a programmer
debugging code will read the purpose of the source files he's working
with right there and then. Some might say this is redundant at this
stage, but somehow these little nuggets of IA wisdom throughout seems
to be the very thing that saves the dead-line. Documentation in a
traditional layered fashion seems to be a dead end in my experience.

> So, what I'm getting at is that I agree with Listera what this guy is
> doing is a step. Does it look perfect,I don't think so. But, it's
> forward momentum. We need to look towards creating living documents
> that are standardized.

Amen.

> Lastly, to you point of "Programmers can't design, and designers can't
> program." I sure hope you're wrong [...]

Of course he's wrong, given the arbitrary definitions of both words. :)


Alex
-- 
"Ultimately, all things are known because you want to believe you know."
                                                         - Frank Herbert
__ http://shelter.nu/ __________________________________________________