Hi Ann,
> In Computer Science, the term 'representation' or 'physical
> representation' means a way of making some structured item,
> or whatever, into a series of character string bytes (or I
> suppose strictly into a seqeunce of 1s and 0s) so that it can
> be transmitted over a network, or stored in a file. The
> trouble is that the 'man in the street' understands something
> different by the term - imagining a picture or some such.
>
> Another word for the same thing that is becoming accepted
> because XML talks about it (and much more trendy!) is
> 'serialisation'. So maybe we could have:
>
> Each rich value... is some text... that serialises the
> resource...
>
> Then we argue about whether it is spelt with a 'z' or an 's'!
I think "serialisation" (c'mon, resist "z"-eification!) would be the
wrong term here.
In the XML context, I think it usually refers to converting some
in-memory (not sure if that's necessarily so?) data structure to a
byte-stream, usually an XML document (typically for transmission to an
end point where it is "de-serialised"), and I don't think that's what we
are referring to here.
The "value" in the DCAM (a resource) may be anything, a person, an
animal, a place, a concept.
I think it would be _inappropriate_ to say that the value string
"My pet unicorn Bob"
or the rich value
[some binary stuff here]
is a "serialisation" of my pet unicorn Bob.
Yes, I can serialise a table or a tree or a graph or an object (in the
data structure sense, Tom...!), but I can't convert my unicorn into that
byte-stream (or de-serialise that byte-stream to obtain my unicorn).
Pete
|