Xref: utzoo comp.protocols.tcp-ip:7940 comp.unix.wizards:17515
Path: utzoo!utgpu!jarvis.csri.toronto.edu!rutgers!iuvax!purdue!haven!uvaarpa!randall
From: randall@uvaarpa.virginia.edu (Randall Atkinson)
Newsgroups: comp.protocols.tcp-ip,comp.unix.wizards
Subject: Re:  TCP/IP on System V/386 & STREAMS
Message-ID: <296@uvaarpa.virginia.edu>
Date: 5 Aug 89 19:45:18 GMT
References: <8908051908.AA12163@ucbvax.Berkeley.EDU>
Reply-To: randall@uvaarpa.Virginia.EDU (Randall Atkinson)
Followup-To: comp.protocols.tcp-ip
Distribution: inet
Organization: GE-Fanuc North America
Lines: 19

In article <8908051908.AA12163@ucbvax.Berkeley.EDU>,
	 dcrocker@AHWAHNEE.STANFORD.EDU (Dave Crocker) writes:
>There is no standard for mapping TLI calls into TCP operations.  Each
>supplier of SVR3(Streams) TCP is likely to have somewhat different
>interface functionality or calling sequences.

I understand that AT&T UNIX System V Release 4 will provide
tcp/ip support.  I have heard that a Berkeley-style sockets
interface implemented via STREAMS will be provided and I'm
sure that a TLI/STREAMS interface will also be provided.

I think that the System V.4 mapping of TLI calls to
TCP will become a "de facto" standard.  Certainly, I
hope so (in large part because I believe that sockets
'know too much' and want to see the cleaner TLI/STREAMS
approach be more widely used).

What I've heard about V.4 is from what I consider reliable sources, 
but if I'm wrong, I'm sure someone will correct me.