I would also like to see a commonly accepted symmetrical transport layer
... but I have problems with what are currently called "IP" and "1284.4" ...
1) (the simpler one) 1284.4 was designed, if I understand correctly, as a
transport that would run across a variety of links, some with rather
restrictive natures (i.e., 1284 parallel port, minimal options) ... so
1284.4 on top of 1394 is a little bit like running the rts-cts serial port
protocol on top of IP (and I've see *that* done, too).
2) IP is a fine, well tested, and (in the form of v6) fairly simple network
protocol ... it is *not*, however, a transport protocol ... you still need
TCP on top of IP to get reliable two way communication. So you really need
TCP/IP ... and then you need to learn the IP addresses, so you need the
full IP address locating protocols of interest (DNS, maybe DHCP ....).
Which is fine if you already have it, but ....
1394 already has all the infrastructure in the local environment that is
needed. All that is needed is the transport itself. A paired SBP-2
initiator/target may provide this function, as may the "thin" transport
proposed by the PWG-C, as may a version of the transport being proposed by
IP1394. Before you get too excited about using 1284.4, or even pure IP,
take a look at the conversations in the IP1394 group. I suspect that you
will find it enlightening, if a bit frustrating.
The key is looking to *throw away* a protocol or layer ... don't add a new one.
Michael D. Johas Teener, Chief Technical Officer & Chairman of the Board
Zayante, Inc., 269 Mt. Hermon Rd. #111, Scotts Valley, CA 95066-4000, USA
email: email@example.com voice: +1-408-461-4901 fax: +1-408-461-1394
======================== http://www.zayante.com ===========================