I like "type transfer" nicely. No point in calling it a URI
since it really isn't... it is a syntax which we've defined
a mapping into URIs with. That is, we don't put !taguri:...
nor to we require a scheme.
The important thing about a "type transfer" is that it has
an equality operator; for example, in the !int type transfer
both 10 and 0xA are "equal". This equality operator must
be a restriction of string equality, thus, the !int type
transfer cannot provide that two scalars with string value
of "10" are different due to location in the document.
By equality operator, we mean one that is an equivalence
relation (symmetric, transitive, impotent). That is,
if A = B, then B = A (symmetric)
if A = B and B = C then A = C (transitive)
A = A (impotent)
On Sun, Sep 22, 2002 at 11:55:36AM -0700, Brian Ingerson wrote:
| Could somebody (Oren) please explain to me the current wisdom on the exact
| meaning and relationships of the following terms:
| - Type Family
| - Transfer Method
| Do we need the word "family" at all. It doesn't make sense to me. I've been
| musing about the term "type transfer". As in "type transfer URI".
| As far as I can see, family doesn't hold any more impact than banana. As in
| this translation:
| While type bananas are preserved in all the data monkeys, the native
| monkey may do so by implicitly associating a specific type banana
| with certain native types.
| ...has a nice jungle feel.
| Um, anyway, could someone lend me a hand?
| Cheers, Brian
| This sf.net email is sponsored by:ThinkGeek
| Welcome to geek heaven.
| Yaml-core mailing list