Does anyone use a fixed width format for your addressing?

[SarawutTangadunrat]

I curious to know what is the benefit of that format compare to the parsed format.

Fixed format :

Address
HAVENBROOK DR 909
J R HAWKINS RD 5301
MARQUIS CIR W 2705

Parsed format:

Number	Dir	Street	Type
909		HAVENBROOK	DR
5301		J R HAWKINS	RD
2705	W	MARQUIS	CIR

[ChrisSmith7]

As far as best practices, addresses are messy. It really depends on how you're using them... The fixed format provides easier editing - if you're just storing addresses to print on letters, or something, this would probably be ok. It keeps the table thinner/smaller, too, depending on the datatype of the field.

The parsed format would make it easier to analyze and search; additionally, you can apply fk constraints and indexing.

If you are dealing with international addresses, that's another ball of wax - I would check-out EndsWithSaurus...: A lesson in address storage  if it applies to you. Here's some info on US addresses - http://pe.usps.gov/text/pub28/welcome.htm

Additional info on best practices:

http://stackoverflow.com/questions/310540/best-practices-for-storing-postal-addresses-in-a-database-...

One thing is for sure, make sure you carefully choose how you architect your db now, otherwise you may be hurting later.

[JoeBorgione]

Chris offers sage advice.

I have to ask: are storing address points and are you geocoding against them?  I do nothing but geocoding for 9-1-1 and for that application parsing is the only way to go.  Also, it is my understanding that the latest iteration of ESRI locators do not perform addressing standardization as they used (ie 9.3 locators) so that my influence your decision as well; it makes a big difference if you will be passing on geocoded information to outside applications as is the case with some 9-1-1 Computer Aided Dispatch (the other CAD) software packages.

So back to what Chris said; it depends on how you are using the data.