superlifter design notes and a new proposal

[Martin Pool]

On  5 Aug 2002, Wayne Davison <wayned at users.sourceforge.net> wrote:

> I think we can easily choose a simple byte protocol now and finalize
> (and optimize) it later.

Clearly we agree more than we disagree, but there is a diversity of
opinion, which is a good situation at this stage of the design.

> That is, assuming that we've decided to go with the design
> that "everything that crosses the wire is a part of a message"?

Yes, I think that makes sense.  To me it is far preferable to
multiplexing streams on boundaries that are not meaningful to the
underlying content.

What are your thoughts about error reporting?  Are they going to be
messages back in the other direction?

-- 
Martin