When I first attempted to capture the history of the working group in the
introduction to P1394.3 I knew that I was omitting details that predated my
intimate involvement. I thereefore asked Ueda-san, Shimura-san and
Isoda-san to fill in some of the gaps. They have done so, and sent me some
proposed text. I've reworked it a little to integrate it with the existing
narrative, and suggest that what is currently the last paragraph of the
introduction be replaced with the two paragraphs below:
"Once the selection of an underlying protocol was agreed, work progressed
fairly quickly on what was eventually dubbed PPDT: peer-to-peer data
transport protocol. The prototype efforts expended by some early adopters,
both in Japan and the United States have been invaluable in the resolution
of detailed technical questions that arose as this document was refined. In
particular, the essential mechanisms first proposed as the Simple High
Performance Transport (SHPT) were ultimately adopted. SHPT took advantage
of the SBP-2 unordered execution model to provide independent,
bi-directional data transport within the context of a single SBP-2 login.
SHPT also optimized error recovery operations after bus reset by using
execution context information known by both initiator and target.
Towards the end of summer 1999 it was apparent that the work neared
conclusion; at this time the working group discussed possible homes for the
draft standard and agreed upon the IEEE as the best choice."
Absent objections, I will place this material in the draft to be balloted.
Regards,
Peter Johansson
Congruent Software, Inc.
98 Colorado Avenue
Berkeley, CA 94707
(510) 527-3926
(510) 527-3856 FAX
PJohansson at ACM.org