[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: snmpconf issue #1: language versioning



>>>>> On Thu, 05 Apr 2001 10:53:07 -0700, Steve Waldbusser <waldbusser@nextbeacon.com> said:

Steve> What we have on our side is that we're defining a subset of a
Steve> very very well known language. So the risk is low that there is
Steve> a hidden flaw in the base language.

No, but we might have hidden flaws in the functions we're defining and
making use of.  I'm not saying the constructs of the language will
change, I'm saying how it is used will.  You've already gotten rid of
"int" in favor of the more generic "var" variable type between the
last version and the new.

Steve> Such changes are backwardly compatible.
...

And there is our philosophical difference.  I'm not convinced we won't
want to make backwards incompatible changes.  You're saying that all
implementations from now on must, for example, implement the
setRowStatus() function from now on, regardless of whether it's
replaced by a more generic "createRow()" function in the future (that
implements both createAndWait and createAndGo).  IE, the
setRowStatus() function will always be around since the next version
of the document is likely to have it.  (sounds suspiciously like
Windows ;-)

-- 
Wes Hardaker
NAI Labs
Network Associates