Re: What databases have taught me

Keith H Duggar wrote:
> Dmitry A. Kazakov wrote:
>> Marshall wrote:
>> Side note: in a strongly typed language "extension" of an
>> operation can be accomplished only through an "extension"
>> of the type (actually a class of). This happens by adding
>> a new type to the class, so that the operation extension
>> be defined on that new type.
>>
>>> I'm not familiar with the term "predicate dispatch" but
>>> as I noted above, pattern matching is roughly the
>>> flipside of OO's subtyping polymorphism.
>> True. It does not add any safety, not even a feeling of.
>
> You know, there seems to be lots of jargon criss-cross here.
> For example, since type = class then "adding a new type to
> the class" is a senseless statement.

Ouch....

Keith, in some OO languages you are correct, a Type = Class.

However, from an OO PoV, Types dont have to be Classes.

This is especially true in dynamically Typed languages.

Or statically typed languages that support Interfaces.

>And as we can see
> "predicate dispatch" is at least unfamiliar to some and
> "pattern matching" can mean /many/ things.
>
> What would be great is if along with flying jargon around
> you all provided /examples/ (in some psuedocode) of what
> you are trying to describe.
>
> -- Keith -- Fraud 6
>
Received on Fri Jun 30 2006 - 23:56:25 CEST