[Sigia-l] IA system components - add to the list!

[Listera]

"Arno Reichenauer" wrote:

It would help if you didn't invent words for me. I never asked "Why a
theory?" at all. I have nothing against "a theory". Not much use, but
hopefully harmless as well. What I did object to was:

>> it is really necessary for the IA community to work towards something like an
>> underlying IA theory...
> 
> Why? 

In other words, I questioned the notion that a (perfect) theory/definition
will (somehow) advance IA, in practice.

The cartographic example I gave amply illustrates the fact that "perfect
theories" handed down from above (the Church) often have little relevance in
the field (portolanos for seamen). In fact, the more "perfect" the Church's
theory/definition of the world got, the less relevant and useful it became
for actual navigation and discovery.

> First: Solving problems: A theory allows you to draw on generalized knowledge
> to find solutions for problems.
 
Well, worry no more. Jakob has already theorized and defined all you ever
need to know about IA, navigation and usability. :-)

> Isn’t the goal of it to share ideas, ask and answer questions and thereby help
> each other to mature as an IA and in the end to advance IA itself?

Again unlike others, I have no problems with people discussing *anything* on
the list. I hope you realize, though, the irony here: the more "perfect" a
definition of IA you divine, the less discussion there's left for
discussion.
 
> Third: Communication to clients: After all, we have to sell IA to our clients.
> What if after your perfect presentation, your prospective client says, “but
> your competitor down the street told me he would also be doing
> HTML-programming for me!” and you don’t?

This is a contrived example in the sense that a client sophisticated enough
to solicit an IA wouldn't usually be clueless enough to not know the
difference between architecture and coding. If it's you who approached the
client then I'd suggest you better pre-qualify your prospective clients. In
any case, if the client wants HTML coding, just contract a third party to do
it, if you can't provide it yourself. It's little use throwing a piece of
"perfect theory/definition" into the face of that client 'proving' for
example why "IAs don't do HTML". This is not how you succeed in business.
The practice of IA, like the practice of nautical navigation, turns out to
be murkier than "perfect definitions/theories".

Ziya
Nullius in Verba