Re: [colorforth] TCP State Engine
- Subject: Re: [colorforth] TCP State Engine
- From: Jonah Thomas <j2thomas@xxxxxxxxxx>
- Date: Sun, 18 Apr 2004 13:06:21 -0400
- Organization: Elysium
John Drake wrote:
--- Jonah Thomas <j2thomas@xxxxxxxxxx> wrote:
If it's your application, say something in process
control, and you
can keep it simple, maybe you won't need all of TCP.
All that short
messages need from TCP is to find out about dropped
messages. You
won't have to split messages and re-order packets
etc. So
implementing just that part is enough for that
application. Or maybe
implement over UDP, whatever works.
The last sentence is the only one that I can follow.
If your message fits in one UDP packet then just
use UDP. I don't see the point of stripping down
TCP to UDP.
TCP does more than splitting large packets, it also provides
confirmation whether messages get through.
And my point is to first do the part you need now.
1) Much of the complexity of IE4 has nothing to
do with TCP so it's a bad example to use.
The only reason I brought it up was that people talked about using TCP
for a web browser. The web browser looks like a complex project.
2) You're saying "it takes the full TCP". So, to
me, that's saying that you believe the TCP required
to handle a full web browser will take more then
3 blocks. Is that what you're saying or not?
I'm not ready to estimate how many blocks it will take. I don't
understand the specs yet. I'm encouraged that the uip guys could do a
subset of TCP in C in only about 4K of object code, for some
unspecified 8-bit processor.
Once more, I say that if you want TCP to use between Forth systems you
can leave out everything that you don't need and that the intermediate
nodes don't need. But if you want to communicate with foreign
applications (which includes among others writing a web browser to
interact with random websites or writing a web server to interact with
Microsoft products) then you have to be ready to handle whatever TCP
stuff they hand you, and also you have to be ready to deal gracefully
with their TCP mistakes. That's got to be more complicated than what
you need to communicate with other ColorForth systems.
---------------------------------------------------------------------
To unsubscribe, e-mail: colorforth-unsubscribe@xxxxxxxxxxxxxxxxxx
For additional commands, e-mail: colorforth-help@xxxxxxxxxxxxxxxxxx
Main web page - http://www.colorforth.com