RE: [Yaml-core] Sparse Syntax

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

Brian Ingerson [mailto:in...@tt...] wrote:
> So it's # for throw away, and // for round-trip?

Yes.

> I say we add # to the spec, and reserve // for this use. I 
> think it needs
> some more time and feedback from real users.

That's exactly what I have in mind. By defining '//' as being a member of an
implicit type we effectively reserve it for that use. Then we see whether
people actually like it or not. If they don't, we can "release" it back for
other uses.

> You know we could make the lack of the space mandatory.
> 
> -[55] fifty-five
> 
> I think I like this rule even if we use #.
> 
> -#55 fifty-five

That's inconsistent with our general descriptor syntax...

> So are we in enough agreement to bang out another spec? I 
> think we should do
> the spec ASAP. I can get people to start implementing that way :)

Well, the document-separation issue still open. Both you and Clark still
haven't decided if you like my "document stream as MIME subset" concept.
Either way I'll be happy to hammer out a new draft this weekend.

> BTW, I am good friends with a guy who did a Perl/Java binding 
> tool. He said
> he'd be interested in trying to do a YAML implementation for 
> Java. Should I
> have him proceed?

Originally I was going to do that, but (wisely or otherwise) I focused my
attention on the spec itself instead. I collected plenty of interesting
ideas on how to do a Java implementation along the way, and I'm itching to
try them out. That said, I don't see when I'll have the time to actually
start working on it, if I'm doing more spec drafts (I've no illusions about
the next one being final :-).

How about you put us in touch? I could tell him my ideas and perhaps we
could cooperate on implementing this.

Have fun,

	Oren Ben-Kiki