ISA100.11a-- the final, final votes are in-- and it doesn't call for strict interoperability and interchangeability
Well, it's over. Finally. The ISA100 leadership team is packing the .11a standard up and sending it to the Standards and Practices Board for approval...and then on to ANSI, and maybe to IEC.
Wait! Didn't this already happen?
Well, no. Not exactly. The vote to approve the standard came with so many technical comments and objections that the leadership team decided, after they felt they'd resolved all the comments and objections they were going to, to "recirculate" the ballot to see if they could get more people to vote yes.
They did. 47 people voted yes, some with unresolved comments and objections, 10 of us voted no, and 4 people abstained. ISA doesn't count the abstentions, so you can look for PR from ISA saying that over 80% of the voters voted yes.
As you might guess from the way the last paragraph was worded, I voted no. I find that it is now time to explain why. I voted no because of one issue-- the standard does not, at its core, require complete interoperability and interchangeability between ISA100.11a devices. You can, as a vendor, make your devices interoperable with other vendors, and interchangeable as well, but you don't have to.
That's just wrong. And it is going to cause end users so much headache that I believe the standard will have to be revised dramatically, or it will fail.
In Luton, England next week, those who are physically going, and those, like me, who will attend by phone when we can, will be discussing ISA100.12-- the committee that was set up to try to make ISA100.11a play nicely with WirelessHART.
The only thing this wrangorous (yeah, its a word) committee has been able to do is to commit to a standard calling for two radios in the same box-- a dual mode gateway. This isn't necessarily a bad idea, after all IEEE 802.11 does this. If you have an 802.11 gateway or router, it has now got up to four radios in it-- 802.11a, 802.11b, 802.11g and 802.11n(draft). There will be issues about how to use a WirelessHART device in a whole network of ISA100.11a devices, and vice versa, but this is not a bad idea.
Of course, if ISA100 doesn't figure out how to demand and ensure interoperability and interchangeability, we'll also be trying to figure out how to get all the ISA100 devices to play nice. At least one vendor of ISA100 prototype devices has told me that he, personally, doesn't care if his devices are interchangeable with anybody else's.
Maybe this whole thing is why wireless sensor networks that are proprietary keep being designed and introduced. Every survey I've done or seen in the past two years keeps saying that proprietary networks based on 802.15.4 radios or other meshing radios are holding their own at around 20% of usage.
And based on what is coming out in the near future, even major companies are going to continue down that path.
Meanwhile, more companies keep joining the WirelessHART ecosystem. The only problem HART is having is getting devices through the testing process-- and they'd better get their act together PDQ on that.
So, what's a poor end user to do? If you have a lot of field sensors already, they're probably HART. You probably should continue down that path. Especially since HART, Profibus Technical Organization, and Fieldbus Foundation are working together to produce a joint specification.
ISA100.11a doesn't help you with backbone coordination issues-- that part of ISA100 is continuing to be worked on, in the .15 committee.
Neither ISA100.11a nor WirelessHART are as easy to deploy as the vendor propaganda would have you believe, but too many people have now installed either WirelessHART or Honeywell OneWireless (which is the prototype for ISA100.11a) for us to believe that it is incredibly hard to do, either.
As Jim Montague says in the August issue of Control, it is time to walk the walk...wirelessly. The only way end users are going to figure out a bunch of this stuff is to put systems in and try them. Don't believe that the horse you pick now is the horse you'll be riding in ten years. Lots can change by then. There's a whole new set of frequencies available now, and the ISM (Industrial, Scientific and Medical) band, which has been untenably crowded for thirty years, may get some of them. New radio designs are being worked on. New, more intelligent devices will come out.
At this point, the benefits of wireless sensor networks outweigh the problems caused by multiple non-interoperable standards.