Second patch ping

[DJ Delorie]

> Well, now wait a minute, let's think about this.  Suppose we set out
> to design an interface that didn't have that constraint,

Like http://gcc.gnu.org/ml/gcc-patches/2003-01/msg02407.html ?

Most of the status APIs you listed could easily be added to that
proposal, as the raw data is hidden anyway.

> All the operating systems of interest support 'launch-and-wait'
> execution of programs:

But not all of them allow you to launch first and then decide if
you're reading from or writing to the process.

> Most of the OSs of interest also support concurrent execution of
> more than one process, and all the ones that do have pipes.  (Unix,
> VMS, Windows)

We should assume neither of these, as they are not portable.