I could imagine that we are only at the very beginning of understanding the real power for Internet of Things in its industrial markets, We look forward to learning more specifics -- what needs to be done on the industry level -- on the Industrial Internet framework.
When the world started to use internet for e-commerce and all kind of financial transactions , no special framework was needed except may be the security features like digital signature , which is not part of the core internet protocols.
So why do we feel the need to have separate stnadards for the so called "Industrial Internet"?
Prabhakar, I would assume that it has to do with the timeliness of communications on a factory floor. Normal TCP/IP transfers will get sent and reliably but not necessarily in a timely fashion. A delay in sending/receiving an important message could be expensive or at worse deadly. I am guessing that the Industrial Ethernet may have (or need) guaranteed delivery times or added message protocols to allow for higher priority packets.
@Robotoics Developer: We will find out in January, but i suspect you are right. This group may well want to bring certain real-time and reliability mechanisms needed in industrial markets into mainstream IoT standards.
I think there's too much hype and/or expectations over this industrial Ethernet. IP has been used increasingly on the factory floor, to control machines, and in other control system applications. Having been involved in this for the past many years, my perception of Internet Protocols is that they have been used for "things" right from day 1. Even "things" that require very tight time tolerances. IP is not just about e-mail or web browsing. And there are companies out there specializing in building very rugged, shock-hardened, EMI-hardened, even radiation hardened Ethernet nodes already. This is not a brand new thing.
This industrial Ethernet push will probably introduce some new features, like isochronous Ethernet perhaps, but whether they find a lot of use or not is another matter. And btw, work on timing accuracy over IP networks, like NTP and IEEE 1588, have existed for quite some time. See also IEC 62439 (not to mention proprietary schemes that achieve the same goals). this stuff has been going on for some time.
Perhaps the biggest need is to keep making Ethernet nodes smaller and more power efficient. That way, networking everything in sight will be affordable even for the average consumer. I think too much is being made of this industrial Ethernet push, as if it alone would open up huge new possibilities.
I agree Bert...this looks to me like PR/marketing re-packaging of the existing technologies and existing developments in this area...I guess if you want to be perceived as an industry leader this is what you do ;-)...Kris