[Sigia-l] The A>B, B>A problem

[Boniface Lau]

> From: Listera
>  
> I think the underlying assumption here is that
> 
> a) this is a novel/new problem
> b) it deserves its own name
> c) and that that name will uniquely describe the issue/process.

There is nothing unusual in looking for a name for something observed.
Properly naming things helps to characterize them. And that is how I
see the original request, i.e. how best to characterize the observed?


> 
> A lot of things in IA, including the notion of facets, is nothing
> new to folks who have been handling data and RDBMs for ages. I and
> others have already said that the reason why you have relational DBs
> is so that you can present same data in different hierarchies
> depending on the context.

Properly naming the observed helps to better understand a situation.
Such understanding may lead to choosing one solution over another.

RDBMS is a solution for producing different hierarchies. But that does
not eliminate the need to characterize a situation with seemingly
different hierarchies. After all, what seems like different
hierarchies may in fact be a manifestation of some underlying factors
that can be controlled and therefore resulted in only one hierarchy.
If that is indeed the case, rushing to apply an RDBMS will likely lead
to less than optimal outcome.


> 
> Why is this important? Because it's not helpful to invent
> highfalutin names specific to IA for things that are common in other
> domains.

I agreed with you that highfalutin names are not helpful. The term
"Information Architecture" is a classic example. While the term hypes
book sales, it does great damage by introducing confusion and
therefore slows the progress in getting a better understanding of the
domain.

Highfalutin names are bad apples. But that doesn't mean apples (naming
things) are bad. Properly naming things highlights their essence. It
is a consolidation of understanding. In the process of coming up with
a name, one often develops a better understanding.


Boniface