Re: Serializations and data structures (was Re: Topic Maps

Lars Marius Garshol (larsga@ifi.uio.no)
Wed, 25 Nov 1998 19:17:16 +0100


* Lars Marius Garshol
>
> Well, you can't very well transmit the data model itself between computers
> or down through time. In other words: you must have the serialization
> syntax. The question is: is an explicit specification of the data model
> of any use?

* W. Eliot Kimber
>
> Do you mean the definition of the data model or instances of data that
> conform to it?

Instances of the data model, hamlet.xml, for example. To be really precise
(as I should have been all along): in-memory instances encoded as electrical
impulses in the memory chips of a particular computer (or a set of computers),
possibly swapped out to disk, but part of runtime memory of a specific
program.

> If you mean the latter, again, we have defined serialization mechanisms
> that allow the data instances to be reliably transmitted: [actual examples
> snipped]

Let me quote myself: "In other words: you must have [a] serialization
syntax." Whether it is XML or serialized CORBA objects is immaterial, but
you do need a syntax to ensure that the information persists and that you
can transmit it to another computer or program.

So the question remains: do you need an explicit specification of the data
model?

--Lars M.