OK, I can't keep my mouth shut anymore. Below is the link to an interview with the OPC Chief Architect, Rashesh Mody, which Alex sent. Mr. Mody also happens to be the WonderWare Chief Technology Officer. Wonderware is a sister company of Foxboro and Triconex, (owned by Invensys)! He is talking about OPC UA which is the next generation of OPC that is built on Microsoft's .NET instead of .COM/.DCOM. Is Mr. Mody the father of the Invensys developed Archestra framework which is also built on .NET? Foxboro has already let everyone know that all future application development on IA will be done using Archestra and thus .NET. http://www.automationworld.com/cds_search.html?rec_id=1169 Part of the discussion focuses on what this might mean to people that have already invested a lot of effort into using the OLD OPC that was built on .DCOM. The really bad thing about all of this is that it still uses a vendor/OS specific transport medium that is very likely to change whenever Microsoft wants to move on to the next OS and transport. Rashesh Mody would obviously have a short term vested interest in seeing OPC move to a .NET solution because it will plug and play easily with the Archestra framework used by Wonderware/Invensys. It is so vendor specific that it will be unusable on any other OS! If we buy into it we know that it will mean constant migrations that will be required every time MS decides to change their OS, which is, at best, every two years. From a visionary standpoint, (longer term functionality), wouldn't it make sense to move OPC in a direction that is able to be used by everyone. http was developed years ago but it continues to provide the backbone to display and access data from any OS without the user knowing or caring what OS the server or client resides on. .NET has the capability to ride above and utilize all of the standard transport protocols such as http, TCP/IP, etc. but requires that both the server and client be running .NET/MS OS and that is where the real problems start. Below is a link to the best article I could find that described .NET in a way that even a dummy like me could almost understand. http://www.developer.com/net/cplus/article.php/10919_1479761_1 Invensys and Foxboro are pretty far down the road in making a very important decision that will determine their future. They have already committed to Microsoft but now many users are voicing their concerns. Control system users have already gotten a small taste of what they are in for if they migrate their systems to a total MS operating environment and many are quite concerned. At the top of the heap and growing alarmingly fast is the issue of network and application security. The recent rollout of XP SP2 and all of the issues that were raised is a red flag for the future as MS tried to plug security holes with a "Quick Fix" that broke many of their own and third party applications that were built to run on their OS. Microsoft would have to be ten times better at providing security for their networks, applications, and transports, than any other provider in the business, if for no other reason than they are hacked 20 times more than any other vendor! They are not close to being that good and as they try to please more and more business sector and private customers they are growing the number of potentially exploitable applications and protocols at an alarming rate. I've read a lot, and listened a lot to all the discussions and have come to my own conclusion that there is currently more risk running MS on a networked Control System than with any other OS. I think the MESH network is a good direction for us to move in but it, too, will require that new skills be adopted by control system administrators and technicians. Foxboro is starting to straddle the fence a little more and it appears that their support of UNIX workstations and AW's will make it possible to use them on the MESH in the traditional way, but won't be able to take advantage of the new Archestra/.NET applications. Thankfully, the new controllers and IO are not directly affected by an MS OS and we do have a path forward that will allow us to replace aging, less capable hardware with new, higher throughput, equipment without being forced into using a very vulnerable OS. Tom VandeWater Control System Developer/Analyst Dow Corning Corporation Carrollton, KY USA _______________________________________________________________________ This mailing list is neither sponsored nor endorsed by Invensys Process Systems (formerly The Foxboro Company). Use the info you obtain here at your own risks. Read http://www.thecassandraproject.org/disclaimer.html foxboro mailing list: //www.freelists.org/list/foxboro to subscribe: mailto:foxboro-request@xxxxxxxxxxxxx?subject=join to unsubscribe: mailto:foxboro-request@xxxxxxxxxxxxx?subject=leave