Plowing on through

Busy day today as the plenary session got underway! Committee chair Dave Robin reversed the usual practice of starting with the involved proposals and pushing out the low-hanging fruit at the end. So today we started with the "easy" or quick proposals to reduce the size of the backlog of change proposals.

He led off with a VERY easy one I had drafted last year when a discussion on the BACNET-L mailing list last year revealed that the datatype of array element zero was nowhere defined in the standard! It was pretty obvious that it had to be Unsigned, and we don't think anyone has implemented it any other way, but such things are best not left to chance. So my proposal merely added, in the approriate place in the standard (Clause 12, p.129), the sentence "The datatype of array element 0 shall be Unsigned."

We polished off a number of relatively simple ones: Bernhard's examples of context-tagging with a context number greater than 14 (I posted earlier about BACnet tagging rules, but an example is a Very Good Thing to help implementers get it right.

We polished off one short, well-done but ultimately troublesome one ("What BACnet Is" was the title, I think) by referring it to the BACnet International and, at my request, the BIG-EU Marketing organizations. Thanks, David!

Then another of my proposals came up. I've already exercised my "taunting rights" reserved these past 13 years (so I won't name it), ever since the committee took a nice, clean little service I'd drafted for a specific purpose and festooned it with a lot of general application capability that nobody ever used. I admit that the new, further extensions I had reluctantly taken would have some useful application (albeit limited) -- but it appears nobody has ever used the service for anything but its original applications. So, that proposal is dead.

Another of mine came out of an observation by BIG-EU's Testing and Conformance Group that the Loop object's behavior when "out of service" was not as completely specified as, say, the Analog Output object. Cut, paste, edit and it was there. Passed. Note to my BIG-EU colleagues: Mission accomplished.

Frank's first proposal, to state the requirement for the DM-DDB-A BIBB (initiates "Who-Is") to be supported by BACnet Advanced Application Controllers because it is actually required, sailed on through with change or question. Good job, Frank!

Then we started getting into more potentially troublesome proposals.

Carl drafted one on the fly to strike the unused ReadPropertyConditional service. Rather than push this on towards a public review addendum we held it back for review in the summer meeting next month so that more folks will have an opportunity to weigh in on it. (Consider this fair warning.)

And there was more -- much more. We'll be back at it tomorrow morning for the last session of the meeting, ending at noon.