RE: [ciq] Identifiers in CIQ schemes

[Max Voskob <max.voskob@paradise.net.nz>]

Fulton,

I'll quote you here

"My follow through point on "use cases" was to suggest demonstrating that a CIQ set of data could
serve as the data source for populating and updating key master tables in SAP or Oracle, etc. (The
payload data would of course have to undergo some mechanized transformation to be shoved into a
relational database, etc). "


That is exactly what I'm after! 

We've got too much stuff in xNAL that is rarely found in real life databases in the form we require
it.
There is a hot example of Street Name, Type and Direction.

How many non-postal DBs do you know that keep the data in that highly normalised form?

I can think of very few. Most of DBs have Street Name Type Direction in one line. Those that do
split them are sophisticated enough to parse an address into those elements, otherwise how would
they get them in the database in that form from users in the first place?

The other thing I still don't understand is why would anyone want to parse a street name in Name,
Type and Direction?
What useful information do those elements have on their own?
I can think of only one use case from my personal experience:

Rostov - on - Don is a city in the south or Russia. All mid-size streets that go along the river
(east-west) have type STREET, all those going north-south are called crescents. All large ones are
called avenues regardless of the direction. 
This was true back in the days of communism and total planning. Now they call them whatever. 
Here, in NZ we don't even have a definitive list of street types. 

The same applies to many other xAL elements. Providing use cases we can figure out what the
practical level of complexity is.

The problem with xNAL always was that it's just too complex and too flexible to be practical on a
wide scale. Even the latest xNL/xAL-Basic versions are not simple enough in my opinion. 

Makes sense?

Cheers,
Max