>The word "streams" means something much different when you have to
>shout it.

My own memory is fuzzy, but I believe this is a paraphrase of
words from Dennis Ritchie himself (I think the original statement
was `Streams means something different when shouted').  (The
V8 stuff was called Streams, with one big S, at least in some
of the papers I remember.  Streams explicitly did *not* attempt
to solve the mux/demux problem: dataflow was strictly two-way.
This alone makes an enormous difference in complexity.)

A better name for the System V stuff is `sewers'.  (I saw some
of it once....)

>(it's been so long since I've seen mention of STREAMS/[b]streams that I
>don't know the status of that portion of 4.4(.[12]) anymore.  Was that
>Karels' project or McKusick's?)

The former, and as far as I know it is still in the same state it
was a year ago:  Mike had his `final' (?) version of the data
structures designed.

Chris