[Date Prev][Date Next] [Chronological] [Thread] [Top]

Re: Get rid of "suggested minimum upper bound"



At 11:23 AM 2002-03-11, Michael Ströder wrote:
>Basically I asked for feedback of implementors. If anyone relies on the "suggested minimum upper bound" (formerly known as "maximum length" ;-) and how implementations are interpreting the meaning of it.

I believe we have multiple independently developed interoperable
implementations supporting the "suggested minimum upper bound" as
detailed in RFC 2252, 4.3.2.  I note that server which enforce the
"suggested minimum upper bound" as their upper bound are compliant
with this section.  As the section places no specific semantics
which clients are to implement, I believe it is sufficient that be
(which I believe there are) multiple client implementations which
recognize and generate values per the grammar.

I think the specification just needs to be more clear in that
this is not a "maximum length".  I don't think it would be
appropriate to alter the ABNF nor to change the meaning
from that described in RFC 2252, 4.3.2.

Kurt