JLS3 - can we get a peek?

Question

Hello!
I read somewhere on this forum that the JLS3 is very close to completion and it's just being "proofed" before going out.
Can we get a look at this latest pre-press version? It can't be an issue of secrecy since the draft version is already available at
http://java.sun.com/docs/books/jls/jls-proposed-changes.html
Many eyes make for shallow bugs, as the saying goes. It goes both ways - we won't have to wait for an updated version and Sun gets extra-plus-plus-good distributed proofing for free.
One question I have, for example:
"The direct supertypes of a type variable (�4.4) are the types listed in its bound."
What if there are no listed bounds? Does java.lang.Object become the direct supertype? I think it's fairly obvious that this is the case, but it's the kind of thing that should be clearly stated.
Another one:
"a type variable is a direct supertype of its lower bound."
How does a type variable have a lower bound? It doesn't. The synthetic type variables produced by capture conversion might have a lower bound, I guess. But those are not known as "type variables" - they seem to be defined as "intersection types". Are intersection types a form of type variable? Shouldn't the statement then be written "an intersection type is a direct supertype of its lower bound" ? Furthermore, the syntax for intersection types has no way to express the lower bound. I realize that we can't write intersection types programmatically, but then why give a syntax for them at all? Presumably it's so that they can be reported in error messages? Therefore I ask again, shouldn't there be a provision for expressing the lower bound?
Those two are minor examples - they're just what I'm looking at at the moment. There are several other places where the draft is unclear or ambiguous or seemingly wrong.

Answer 1

Thank you for your attention. I really appreciate it.
Can we get a look at this latest pre-press version?You will have ask the author. He doesn't read this forum.I was sort of hoping you (or someone else who reads the forum) would ask for us, if you consider it a reasonable idea. You have a little more purchase seeing as you know him (gafter?) personally. ie "Dear colleague, I thought I should let you know that the throngs are clamoring for a look at the latest version of the jls3. Perhaps we can release it informally, somehow?"
I don't know - maybe it's going to come out next week. It's tough, though, not knowing precisely when! It has been Real Soon Now for a while.
Thanks for trying to answer my question. Those two RFEs are indeed interesting, but don't quite apply to my question. I see now where I got confused, however:
Basically my problem was just that "intersection types" sound alot like those synthetic type variables "Capture of ? whatever". They come up during capture and inference, you can't write them textually, and they have the same members as a type variable with the same bounds.
Upon further review, I see now that they are not the same thing. Please excuse me.
I guess my confusion just comes from the fact that type variables with non-nulltype lower bounds are not normatively described in the section for type variables. Lower bounds for type variables are not mentioned at all except by reference, in these two places:
"If Ti is a wildcard type argument of the form ? super Bi..."
"a type variable is a direct supertype of its lower bound."
So... it seems like: either there needs to be a mention of lower bounds in the definition of type variables, or there needs to be a 'subtype' of type variables called "captured variables" or something similar.
interface TypeVariable {
     List<Type> getUpperBounds();
interface CapturedVariable extends TypeVariable {
     Type getLowerBound();
}no?