So I get to give a fair number of industry presentations, and one of the recent things I have been trying to explain to people is that it is a myth that the IEEE makes decisions regarding standards. As a long-term IEEE'er, it might surprise some that I would say something like this, but I get it, as do a number of my other colleagues in the IEEE. The IEEE is a forum for the industry to make decisions. Now, it should be pointed out that the real challenge is that those decisions require 75 percent or better approval to move forward (unlike the simple majority or super majority requirements one hears about for Congress and the Senate).
Therefore, the importance of opportunities for the industry to gather and drive toward consensus cannot be overstated.
In my last blog I talked about the Ethernet Alliance's Technology Exploration Forum (TEF), coming up on June 14 in Santa Clara. Given the IEEE 802.3 100Gb/s Backplane and Study Group's progress, where it will be seeking the approval of its project in July, and the call for interest on next generation 100GbE optics, the timing of this event is perfect.
For those of us involved in backplanes, it is easy to understand the debates. What is the channel? What is improved FR-4? How about the cost? And this will go right into the signaling and encoding. Oh, and don't forget about energy efficiency Ethernet, or EEE, and how it applies.
Next generation 100GbE optics? Same story here. How can we reduce power and cost while increasing port density? Million dollar question, eh? 25Gb/s signaling? Re-timed versus non-retimed versus partially retimed interfaces! How about 25Gb/s over multi-mode fiber? A new signaling specification for single-mode fiber? In the 10GbE project there were two reach objectives--one for 2km and one for 10km. One solution was chosen to satisfy these two objectives. How about if the reach was 500m? Something new then?
I'm an IEEE'er and an engineer, and I have to agree with lcovey about cutting edge. IEEE 802.3 standards have been technology followers, borrowing from other technologies such as FDDI, fibre channel, etc or from other organizations such as the ITU, OIF, etc. Even the IEEE Std. 802.ba specification borrowed a lot from its own IEEE Std. 802.3ae. As a matter of fact, that is one of the requirements for the development of an IEEE 802.3 standards project: technical feasibility. That in itself removes the ability to be cutting edge.
That's not a bad thing though because many end users like being able to know that their networking equipment is based upon "tried and true" technologies and that if multiple companies support the standard that it can also mean there will be multiple vendors to choose from. For those that want cutting edge technologies, then there are other choices and associated risks.
I'm not an IEEE'er. I'm not even an engineer. But I know that the IEEE doesn't set standards. It merely provides confirmation of what the engineers worldwide have already determined. Lots of companies want to control what the IEEE does, specifically to benefit their own technology, but the organization routinely shrugs off all sales department pressures and follows the dictates of the membership. As a result, nothing identified as a standard in IEEE can be considered cutting edge. That fact is what always surprises me about companies trying to influence the voting. they always like to say their stuff is the cutting edge of technology, but if it is a standard, it can't be.
Join our online Radio Show on Friday 11th July starting at 2:00pm Eastern, when EETimes editor of all things fun and interesting, Max Maxfield, and embedded systems expert, Jack Ganssle, will debate as to just what is, and is not, and embedded system.