Re: FCEncap Last Call Comment 40

[Ralph Weber <ralphoweber@compuserve.com>]

Mallikarjun,

> I would have been much pleased if an assertion was made that the
> probability of "bad sync" is very low and can be simply ignored.

I think it is more a matter of, "bad sync" is supposed to be
a very low probability issue as per RFC 2030, and that is why it 
is being ignored. The last time I checked, there was no real world 
data verifying the low probability assertion, so I would hesitate 
to state it as a fact.

I would further note that, even if different implementations
took different actions to detect and correct "bad sync", there
would still be no interoperability problems. There would be
implementations that the market views as robust and implementations
that the market views as crash-prone. Coming from a Fibre Channel
standards background, the existence of market forces to select
behaviors is sufficient, except where interoperability issues
arise.

> The reason I took the time to explain my earlier cryptic comments 
> was also because I didn't want to leave you with an (incorrect) 
> impression that I think the "absolute difference" text change did 
> address the comment.

Since we both agree that "absolute difference" is a bad idea
and since I see no way to standardize good behavior at this time, 
I will revise the comments response to reject comment 40.

Thanks.

Ralph...