Dust Network CEO Rob Conant also gets it...

"Something happenin' here...what it is ain't exactly clear..." For the second time in a week, I've interviewed a major wireless CEO and they've both said much the same things. Here's what I asked Rob Conant, CEO of Dust Networks: WB>It is now an open secret that after some significant flirtation with Millennial Net, Emerson has decided on Dust for the radios for its new wireless offerings. They cite "numerous" test sites. Some of your competitors are loudly doubting the existence of those "numerous" sites-- would you care to discuss your relationship with Emerson, and tell me a bit about the actual test sites? RC>No, I can't talk about our relationship with any of our clients. Actually, we feel like it is our competitors who are spreading the FUD (Fear Uncertainty and Doubt --ed.) Our job is to eliminate that FUD. We do have many test installations, including things like a dual delayed coker, but we can't yet talk about them. Our first deployment, back in September of 2004, has been running continuously since then, with no problems. That's uptime of 99.97%. And that's with our first generation system. WB>You, and Emerson, have made a huge noise just before the HART Wireless working group is to vote on the final choice for radio communications technology. I've asked Emerson's leadership this, and now I'm asking you. Your competitors say this is a bold-faced attempt to hijack a standard. Is this your intent? RC>Heck no! Look, it is bad for the industry not to have a standard. As you have said in your blog, it is important to making the market for there to be a single standard for industrial uses of wireless (not including RFID and plant floor WiFi, of course) just as soon as possible. We're a small company, and we're putting a lot of effort into wireless standards activities with HART and SP100. That's why we're working with more than half of the equipment vendors who are involved with HART Wireless and SP100. WB>Many of your competitors have commented that they don't believe in the scalability of your solution. What do you say to that? RC>TSMP (Time Synchronous Mesh Protocol) is our answer to scaleability. It is much more scaleable and deterministic than CSMA networks. We have proven that we can scale up to 10,000 points. Access is limited by the available bandwidth at the gateway, of course. But we have shown that we can produce hundreds of nodes per gateway. WB>John Berra announced that reliability was a problem that has been solved. I don't believe that for a minute, based on over 30 years' experience with radios. Do you agree with him, and why? RC>Reliability is being taken care of by smarts in the network. We produce redundancy enough that the signal will get through. WB>Yes, but doesn't that redundancy make the latency problem real? You can make something reliable by making is so slow that it can't be used for control, can't you? RC>Reliability takes time. Low Power Self Organizing networks are not able to do realtime control. The first applications we'll see once a standard is in place are monitoring, data collection and supervisory control. WB>So you're saying that you won't be able to do loop control, for example? RC>No, no. Look. It all comes down to latency. There is a direct trade-off between Power Consumption and Latency. The more the radio is on, the lower the latency and we can maintain the same level of reliability. If you can give us more power, we can do loop control and other control functions, in non-critical, non-safety control systems. This is why CSMA won't work...it uses too much power. In cellphones it works fine, but it isn't a reliable low power network. WB>Are you in favor of multiple silicon vendors? RC>Yes, absolutely. We expect there to be multiple vendors for the silicon. We believe there is plenty for everybody. WB>Will we get interchangeability, or simply interoperability? RC>Interoperability is the bare minimum, the lowest bar for a standard. The end users want interchangeability, and the only way they are going to get that is if the automation vendors agree to a standard, and how agressive the automation vendors want to be in determining that standard. It is ultimately going to be a vendor decision, not ours. WB>What do you think should happen next? RC>What needs to happen is to get all the vendor companies and end users committed to getting a standard to the market ASAP, so that the industry can make a market. All the equipment vendors want a standard to emerge, and all the companies need to buckle down and agree on one. WB>It seems to me that some companies are still looking for individual competitive advantage here, instead of recognizing (as the telecoms industry has done) that a common standard out soon will create the market need that end-users are too scared to talk about now. Do you agree? RC>Well...yes. And you misjudge your ability to sway end-users. People read your blog, your editorials, and you ought to be saying stuff like this. Dust is committed to getting out a standard.