An Open Letter to the SP100 Community
cc: the entire automation community
Since everybody seems to be writing "open letters" I thought I ought to get in on the game too. Let me take a moment to set out the situation in the automation wireless standards activity for everybody to reflect on. As a result of having started several months earlier than SP100, it came as no surprise to nearly everybody when the HART Communication Foundation board voted with one negative to approve the release of the HART 7 specification, including the portion known as "WirelessHART." This is a real watershed, because we have gone from having NO open standards for wireless in automation that have not been borrowed from the IT world (IEEE 802 family, including 802.11x) or the commercial world (IEEE 802.15.4 and its companion, Zigbee standard) to having ONE that is approved, real, and open. Regardless of what appears to be a desire on the part of some members of this community to put the cat back into the bag, you can never make stuff unhappen. Recently, a member of this community asked for a use case for combining the WirelessHART standard with the work that we, the members of SP100, will do going forward. Rather than just do that, I'd like to provide some illustration of use cases for WirelessHART and also for the path SP100 appears to be going down. Let's face it. WirelessHART has happened. That's a fact. The first WirelessHART products I know about personally will be introduced in December for shipment in January. WirelessHART's use case by itself is compelling. There are now 24 million installed HART devices, and the number is growing by approximately 2 million devices per year, according to the HART Communications Foundation. As of right now, the only way to wirelessly get the "stranded information" we've all been talking about for two years now out of those devices is via WirelessHART. Beside that, it is a simple marketing case. You use devices you are completely familiar with, and you put this simple device on them and connect up this gateway to your control system's Ethernet LAN. End users understand this, it resonates well, and in the absence of SP100, there'd be no argument, either. And there's the rub. There IS NO ISA100.11a standard, is there? There may be, if we as a standards body don't screw it up. But there isn't one yet, just like there wasn't a WirelessHART standard a week ago either. The HART Communication Foundation has been accused of approving a competing standard, as if there WAS an ISA100.11a standard already. This is wrong, and wrongheaded. And it is entirely counterproductive. WirelessHART isn't everything to everybody, and we, as an automation community need a standard that provides a "bigger tent" so to speak, than WirelessHART. There are, for example, around 12 million Profibus devices installed, and by the end of 2008 there will be close to 1 million Foundation Fieldbus devices installed in the field. Not only that, but there are thousands of DeviceNet, ControlNet, CIP and even BACnet devices installed in process automation environments. We need the other half of the standard. ISA100.11a could become the other half of the standard, complementing, cooperating, and interoperating with WirelessHART. It should become the other half, and there are no technical reasons why it could not incorporate WirelessHART easily and painlessly. All that is required is a dual-stack model. That would make it entirely transparent to the end users, and they wouldn't even have to worry about which standard they were using in the field. That's what the end users want. I've decided it is time to do another survey, since several members of even the End User committee seem to think I'm not correct about that. Next week, I will be sending out, exactly as I did before, a survey to over 20,000 end users. I will, as I did before, ask ISA, and the other automation media, to mirror the survey, in the hopes that, as we did with my last survey, we'll get a statistically significant number of responses in relatively quick time. Here is the survey: 1. Should SP100's standards be compatible and transparently interoperable with the new WirelessHART standard?2. Should SP100 be a competing standard to WirelessHART?3. Should SP100 establish standards for communication between the wireless gateway and the control system?4. Should SP100 establish standards for communication between the control system and the enterprise?Why the last two questions?
The use case that has been developed for the SP100 path to the future rests on those two questions. Some members of this committee have taken the users' insistence on "one wireless standard" to mean that they want the following scenario to happen. First, we want the users to decide between WirelessHART and a competing standard, called ISA100.11a. End users have voted with their feet for nearly 20 years now on fieldbus, and we're not listening, again. Then we want them to go to their management and tell their management that they either need to take plant wireless networking away from IT, or make IT share responsibility. That will, of course, happen when the last member of every IT department in the world slumps to the ground cold and dead. If the hapless end user survives that, we then want him to go to management and tell management that he wants to install a single wireless network that ISN'T IEEE802.11x, because "it's better." If the much diminished pool of end users that survive that are still numerous enough to make a market, we are then going to give them the business case for doing that. I'm not exaggerating. For SP100 to do a lot more than be the complement to WirelessHART is, frankly, silly. I don't care if you are a devotee of Apprion's or Cisco's or 3ETI's, or anybody else's 802.11x systems, but the fact is, they are already there, they are a worldwide open standard and it is highly doubtful that we have a prayer of dislodging them. Just go look at what the end users, like PPG Lake Charles, for example are doing! WirelessHART and hopefully ISA100.11a for the sensors, 802.11x for the WiFi and backhaul. Several people are saying that ISA100.11a is a done deal and will be finished by December, or perhaps just a bit later. Anybody who has been active in this committee who really thinks that needs to look long in the mirror, and anybody who's going around saying that needs to examine his or her conscience. As one of the founding members of this committee, and as a professional with more than thirty years of experience using wireless in automation, and as an informed industry analyst, it is my professional opinion that if the SP100 standards body follows the rules governing ANSI standards bodies,we'll have a standard sometime in the next 10 to 18 months. Not by December. I am sick and tired of having this standards body waylayed by private agendas. The HART Communication Foundation has provided a copyright license to SP100 so we can begin the process of making the standard we are creating compatible and interoperable with the only automation wireless standard that actually exists. Two weeks ago, many prominent members of this committee were complaining loudly that they didn't know what the HART standard was, and by not providing it before it was officially adopted, the HCF was somehow not playing fair. Now, in an extremely expeditious manner, especially since the standard was only adopted last Friday afternoon, the HCF has licensed it to SP100. Did this quiet the furor? Of course not. Now, many of those same players are saying that it isn't good enough, because the agreement that Chip Lee worked out on behalf of the SP100 committee says that its distribution would be limited to "active members." Once again, the HCF "isn't playing fair." Nonsense. Not only did Chip Lee agree on that wording on all of our behalf, it is pretty clear to me that it makes sense. If you don't think we can't tell who the active members are, dream on. And just so it's clear what I think I am, I have asked to change my status to voting member. I chose not to be, when we formed this committee, but I intend to further increase my level of contribution from this point on. And now I want to again declare that I am a completely disinterested third party, working on behalf of the profession and the end user community. I do not work for any automation vendor. I do not prefer the products of any automation vendor. I am speaking for the end users. Nothing more, nothing less. Unlike some members of this committee, I am listening to them, too. No, it isn't the HCF that isn't playing fair. It is some of the members of this standards body that aren't playing fair. According to Mr. Bolick's letter, the only way HCF will be considered to play fair is if it withdraws the WirelessHART portion of the HART 7 standard. In my humble opinion, and with apologies to Mr. Bolick, that's nonsense. Yet Mr. Bolick made no effort whatsoever to explain what the dozen or so companies who worked on the WirelessHART standard and have already developed prototype WirelessHART devices, stand to gain by doing that. What is the business case that would cause those companies to give up all the work they've done, and eat all the NRE they've spent already? There is none. And therefore, Honeywell was alone in voting against HART 7. The first WirelessHART devices will begin appearing by December. It is doubtful that the first ISA100.11a devices will begin to appear in 2008 at all. Let's face reality. The WirelessHART standard is here to stay. SP100 is now burdened with the obligation to cooperate with WirelessHART. That's a fact. Unless you want to pretend that it doesn't really exist, and that on the close order of 11 companies, not just Emerson, will have products on the street within the next four or five months, all you are doing is playing games for private agendas. I implore you all to stop playing these silly games. You are doing the profession, the end users, and all of the automation vendor companies a disservice. Do not lose sight of the goal, here. The goal is to provide a game-changing, paradigm-breaking technology that is simple, easy, and uncomplicated for end users to work with.In the final analysis, the end users will vote with their cash. And they CAN and WILL vote "no."
Walt BoyesMember, SP100
