Vint spoke on this very topic as the dinner speaker for the TCP/IP
Interoperability Forum in Monterey (the pre-Interop interop conference). As I
recall from his comments, the decision to split the two came about as a mutual
epiphany between him and Dave Clark as they were trying to figure out some way
around some of the stupidity in NCP. As I recall, the document cited below had
a lot of this history... I seem to also remember that it was published as an
IEEE or some such document in addition to being an IEN. I believe TCP v3 was
the last version that was unitary and that was published in Jan of 78 (IEN21).
In Feb of 78 Jon Postel published the first draft of an IP (IEN28)
IEN48 Cerf Jul-78 The Catenet Model for Internetworking
I'll leave to Vint to clarify and correct...
Dave Crocker <dhc2(_at_)dcrocker(_dot_)net> wrote:
At 02:19 PM 7/17/2001, Radia Perlman - Boston Center for Networking wrote:>I
think TCP and IP were all one layer at one point. It's possible
the old IEN's, (which I'd heard would be on-line "soon" if
That is my understanding, too: In the beginning there was only TCP.
During on-going discussions between the ARPA research and the Xerox PARC
research folks, the need for separating out IP functionality became apparent.
I don't know the dates for this architectural transition.
Transport (TCP, UDP)
Internet (IP, etc.)
Network Interface (LAN and WAN technologies).
The idea of intermediate "upper" layers has periodically been
played with in the Internet realm, such as with XDR and RPC, and again more
recently with BEEP. A range of efforts that standardize on an application
infrastructure, such as with XML, encourage viewing this as a real
architectural enhancement. Hence we probably need to wedge a layer between
transport and applications. I prefer to call it session; others will
likely prefer presentation.
tel +1.408.246.8253; fax +1.408.273.6464