Re: [Yaml-core] New Draft.

[Oren Ben-K. <or...@ri...>]

Stephane Payrard [mailto:s.p...@wa...] wrote:
> You did not answer about the appropriatness of the attached document.
> So here it is again, inlined this time:

Oh. Sorry, I missed it somehow.

> To make YAML more conformant to our first goal (maximum human
> readability) I would like to add two syntactical constructs:
> contextualizers and oneliners.

Hmmm. Well, I think that "contextualizers" won't fly. They
complicate things and it isn't clear to me how they help
readability - at least not from the examples you gave.

As for one-liners - it seems as though most of what you want
may be achieved by using an implicit type. Clark has suggested
a "tuple" type along these lines:

    3d vector: (1, 2, 3)

Other types (e.g., matrix) may be similarly handled:

    3d matrix:
        (1, 0, 0)
        (0, 1, 0)
        (0, 0, 1)

Both formats may be very useful for, say, a 3D graphics
package. But as there is a way of implementing such syntax
forms using implicit types I don't think there is a reason
to directly support them in the core spec. Instead, I think
it is rather neat that the above is possible while using
the core spec exactly the way it is now.

This also relates to the discussion Clark and I are having
about implicit types. I don't have any problem with some
group defining a set of implicit types which include the
above and use them privately to encode 3D data.

If, on the other hand, the same group wanted to promote this
encoding as a way to interchange 3D data, then they'd have to
make these types, and their schema in general, "well-known" -
which I hope would include registering it in yaml.org.

Have fun,

    Oren Ben-Kiki