IPP> PRO,MOD> User model for clients

IPP> PRO,MOD> User model for clients

rdebry at us.ibm.com rdebry at us.ibm.com
Wed Jan 29 14:19:10 EST 1997


Classification:
Prologue:
Epilogue: Roger K deBry
Senior Techncial Staff Member
Architecture and Technology
IBM Printing Systems
email: rdebry at us.ibm.com
phone: 1-303-772-2479


---------------------- Forwarded by Roger K Debry/Boulder/IBM on 01/29/97 12:14
PM ---------------------------
Alex Bochannek wrote:


Also fair, but you overlook one important aspect: You already have
designed a brand new protocol on a clean sheet of paper! It's called
IPP. All we are arguing about is that you think it should be
transported in another application protocol while I say you should use
a transport protocol with more than 15 years of use behind it. Not
only your favorite, HTTP, uses it, but also a whole range of
others. Look at RFC821, dated August 1982. It describes the mapping of
SMTP onto the TCP transport in about one dozen lines of text. And it
is still working today.


RKD> I have updated the protocol paper to show both an HTTP
RKD> mapping and one using TCP sockets, per Alex's suggestion
RKD> of a few days ago.  I hope this allows us to carry on
RKD> a reasonable debate on the guts of the the IPP protocol
RKD> without getting bogged down on the transport. In the
RKD> meantime we seem to be running our prototype very nicely
RKD> on HTTP. I'd like to see more implementation experience
RKD> before we cast the transport mapping in concrete.


RKD> I'll post the updated version tomorrow after some
RKD> internal review on clarity and readability.



More information about the Ipp mailing list