Boost logo

Boost :

From: David Abrahams (david.abrahams_at_[hidden])
Date: 2002-03-24 08:18:07


----- Original Message -----
From: "Jason Shirk" <jasonsh_at_[hidden]>

> First, 7.1 isn't even close to a rewrite. I have no idea where anyone
> might have gotten that impression. Please don't ever mention it
again,
> it will scare people unnecessarily.

Sorry, Jason!

It just seemed obvious to me given the number of things which don't work
in 7 (or work wrong) but which reportedly do work in 7.1; however, there
is often a gap between the obvious and the truth... I guess this is one
of those cases.

> We've rewritten access control and how we handle ambiguities (we've
> formalized disambiguation in our grammar, previously we had some awful
> hacks that failed often enough to throw the code out.)
>
> Other than that, we haven't really rewritten anything. We've just
been
> fixing bugs and implementing new features (and cleaning up some old
code
> in the process.)
>
> We have been more aggressive w/ "breaking" changes to become more
> conforming, but in general it's pretty easy to port your code to the
new
> compiler because the "breaks" we've made have been carefully chosen
and
> the diagnostics are usually helpful.

I was never worried that porting would be a big problem, but I can see
how my aside about a "rewrite" might have caused that concern. Again, my
apologies.

> Second, we are getting close to finishing up the work we have planned
> for 7.1. That doesn't mean we'll be shipping as soon as we'd like.
We
> are part of a much bigger product, so our schedule doesn't always
match
> the shipping schedule of the whole product.

That's about where I thought things stood.

It wouldn't hurt at all, I'm sure, to have a feature complete 7.1 in
testing for many months before the whole Visual Studio product finally
ships.

-Dave


Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk