Friday, April 27, 2007

Ethernet Bloggers have various philosophies about comments. Some turn them off entirely due to spam (I've only been spammed a couple of times, so far). Some don't publish comments until they have reviewed them. I allow comments--but there's a caveat. I get an email every time someone comments. I can go back and delete comments that are spam, violate ethics or good taste, or whatever. One thing I usually delete are the (as Slashdot calls it) "anonymous cowards." I've been on the Internet since 1991 and have never stood behind that idea of the anonymity of the Net. Be who you are and stand behind what you say. That's my philosophy. And if I say something wrong, then I'm responsible and will correct it. My blog on Ethernet has generated a few comments--one by an "ac". That person does know the Internet--used a gmail account and masked the IP address so that I couldn't trace it quickly while I was a LaGuardia yesterday. So Carl Henning responded to the post before I was able to delete it. I have no issue with either camp, and maybe I'm beating this topic to death. (Want to suggest a new one?) I was a peripheral member of ODVA in the mid-90s (in the interest of disclosure). I don't remember the category, but it was those people interested in working with it but not a big enough company to pay the full dues. I quit when I became an editor in 1998. And by the way, the people I interviewed on the EtherNet/IP side were very upfront and professional. I wanted the other side of "standard" versus "custom" ASIC simply for all those out there who confuse the issues. And they are still out there. I used to have supplier marketing people come up to me and knowingly say something like "Rockwell (Siemens) runs ODVA (Profibus) and they extract a huge fee from people for using 'their' network that's under the table. And besides, really, the other guy is really proprietary [meaning owned by the supplier for the supplier's benefit]." As I'm writing this it occurs to me I've heard the same stuff about Emerson and Foundation Fieldbus. I had some relatively inside information at one point and have spent countless time investigating since. Guess what, these are all open organizations. Yes, there may be one or two companies who invest more than others in the technology, but that's simply a choice. I used to cover the embedded systems market. I bet I'd run into half-a-dozen manufacturers of Ethernet and embedded TCP/IP ASICs at the shows. One was developed specifically to meet the needs of industrial device manufacturers (Net Silicon at the time in the late 90s). Each manufacturer had some feature or other that beat out the competition. I've seen the benefits of the special features for motion using the Profinet ASIC, and I've talked with engineers who have built applications using CIP and EtherNet/IP for motion--and were happy. Custom versus commercial off the shelf (COTS) arguments usually revolve around price and availability. There is a technical argument also to be made as to where you want the algorithms to run--in silicon or as software on the controller. That one is best answered by actual application. Price and availability? Just check with the device manufacturers and see which costs more and are they on the shelf. Just as long as all the users out there understand that these solutions are all open and not proprietary. That's my agenda. 10:53:28 AM    comment []