[AccessD] Number vs text data type

Hi Lambert

I think you are mixing up two things here.

Private formats controlled by some management may, of course, be
changed with no warning.

Public formats, however, are so widely used that you'll always be
given a lead time counted in years. This holds true for postal code
formats, telephone number formats etc.

Have for example a look at how slowly but steadily the IBAN
international numbering format for bank accounts is promoted while
you still, but at increasing costs, can use the old system with bank
names, BIC (SWIFT) codes and account numbers.

But I will vote with you.

/gustav


> Drew,

> True the (US) post office may or may not be planning a zip code reorg, but
> the point is that the Zip code is just an example of a number that's not
> really an number at all.

> Other things often get "numerical" values, like Customer Numbers, Account
> Numbers, Part Numbers etc. but they are not numbers in reality. They are
> simply identifiers. These are exactly the kind of things that management can
> and does change its mind about at the drop of a hat. There's no guarantee
> that you'll get a 'decent amount of lead time' when the decision come down
> the line.

> To your second point, if today's processors are such that speed is not
> critical anymore, then there's surely not much chance that tomorrows CPU's
> will be significantly slower. So we have processing power to spare, and
> searching string values is no big deal.

> The really significant point about searches is to make wise choices as to
> which fields will be indexed. Good indexing equals fast searches. No
> indexing means massive network traffic and lots of waiting around.

> My vote is for the philosophy that says "use a data type that's appropriate
> to the type of data".

> JM2CW

> Lambert