[IP-SFS] IP over Semaphore Signals for April 1

[Bob Braden]

Authors,

Your April 1 submission is in the running for possible publication,
but we need some more changes to make it acceptable.

1. Is the reader supposed to know that the encodings
	shown in Figs 3 and 4 are actually official navy flag language,
	or derived therefrom??
	YOU NEED ONE OR MORE CITATIONS AND CORRESPONDING REFERENCES
	TO THESE FLAG SIGNALS.
	
	A couple of minutes in Google found www.sacdelta.com/semaphore.htm
	and navy.memorieshop.com/Signaling/Flags.html.  It would be
	good to have both one of these URLs and also a reference to some
	published work containing the same thing.  And you should briefly
	explain what changes/additions you had to make to this navy
	semaphore standard to create your encoding.

	I see a rich set of semaphore literature in wikipedia.

	(Who is Jim Croft, referenced in the Abstract?)

2. The layer diagram in Fig 1 is WRONG, which will upset many of
	your potential readers.

	IP (Internet Protocol) is the Internet layer.  It is layered on
	top of SFS, which is what should be in the lowest box of Fig 1.
	What RFC 793 called the "communication layer" is now called the
	Link Layer (or link layer).

	I guess Europeans might call IP-SFS an adaptation layer, since
	it adapts IP to SFS.

3. An RFC is required to be logically complete without its Abstract.
	It should start with an Introduction section, whose contents
	are allowed to be redundant with the Abstract.  Your
	Abstract is too detailed at the end; a much shorter
	Abstract would have more authority (and what you remove
	from the Abstract should move to the Introducetion).

If you search the RFC index for "IP over", you will get some 93 matches,
including RFC 1149 (Avian Carriers)

You need section/subsection numbers

In the description of a block (btw, a more standard terminology here might
be "frame" rather than "block"), use "5-15" rather than "5...0xf".
The RFC convention is to give such fields in DECIMAL, but in any
case, mixing decimal with hex is bad form.

"nibble" -> "4-bit nibble"

The description is confusing... Fig 2 refers to octets of data, but
then "Signal Coding" talks about nibbles.  And what is a "block"?
Is an IP datagram fragmented to fit into a block if it is larger
than 255 octets?

THe sentence under "IP-frame encapsulation" does not make sense as written.
And what is an IP frame?  An IP data unit is called a packet or datagram.

Let me know if any of this is unclear, and I hope you can respond quickly.
Otherwise, April 1 will arrive without you.

Bob Braden for the RFC Editor

The game here is to make it look almost exactly like a "real" RFC.
You need to look at more examples.






	

	


----- End Included Message -----


----- End forwarded message -----

-- 
Jogi Hofmueller |m| phone: +43 316 821 451 55
                |u| 
                |r|            http://mur.at/