Wednesday, January 10, 2007

MeetUps Apple Computers Inc. is no more. It's now Apple Inc. And they played a Beatles song at the keynote. Must have patched up things with Apple Corp. This reflects big changes both at Apple and the computer industry. Growth and cool cutting edge reside in consumer applications--songs, movies, entertainment centers, TV to go and on and on. I was almost thinking that Apple has thrown in the towel on basic computing when I read Om Malik at GigaOM talking about some things to come to enhance the Mac and I read about a new Microsoft Office coming sometime for the Mac (to make it compatible with Office enhancements with Vista). Apple was touting iPhone and it's Mac TV offering at MacWorld. Microsoft was touting home entertainment at CES this week. Huge crowds and lots of buzz in that market. Small, quiet crowds these days at most industrial gatherings. Who is going to step up and put the buzz back here? Ever wish you could meet with people who have similar interests once in a while to chat? Try Meet Up. It's interesting. A group organizer pays $19 / month to establish a group. He/she may or may not charge others to attend the meet up. There are Tech meetups (none around Sidney, Ohio though), dog owners meetups (specializing by breed, of course), even a Pagan meetup (near Sidney, but you won't catch me there!). Speaking of meetups, anyone going to the ARC Forum in Orlando and want to meetup Wednesday evening? Or a meetup at Universal during the annual reception? Imagine it. A bunch of automation geeks hanging out, drinking beer (or soda) and discussing the merits of Ethernet protocols or database programming  ;-) 9:20:26 AM    comment []

Ethernet revisitied Well, Carl, I ignored the Ethernet ribbing in your first post. I like your Profinet brochure, but it is, well, a Profinet brochure. I hate to mention it (actually, no I don't), but there are other Ethernet-based industrial networks. I asked Kenna to do a piece getting a little to the nitty gritty of Ethernet--and actually not Ethernet, but the TCP/IP stack and some protocols. I didn't ask for a roundup of industrial Ethernet-based networks. Since I don't have an "Automation World Labs" like PC Magazine has its lab, I can't run comparisons and report on those. So, I stick with reporting a piece at a time. One trouble we have with the term "Ethernet" is that no one (possible exception would be network engineers on standards committees or in academia) uses the term for what it really means. In popular usage, it has come to mean an entire suite of hardware and protocol layers. It can even mean different chips. So, if one by one you look at the different aspects and try to force people to talk more specifically, then perhaps you can avoid gamesmanship and specmanship in the game. Therefore, I like taking a look at piece by piece and not trying to describe the world of Ethernet in one place. It is complicated if you try to be all-encompassing in one spot. And it would take an encyclopedia. For instance, take chips. Ethernet in typical use defines connectors, wires, and the protocols on a chip to make the interface. Then there are the protocols that define how packets will be sent. Then there are protocols that find other chips on the network. Then there are applications. Most of these protocols (plus other useful ones) are found on inexpensive commercially available chips. But then, there are chips (some call proprietary--as a pejorative term) that are, shall we say, special purpose that include all those above commercial, popular and somewhat standard things and extend them in the interest of a special use--for example determinism. Some network applications use these, some don't. Some say they are absolutely necessary, some say they are useful for many applications in industry, and some say they (those special chips) are unnecessary. Some view this as a technology discussion. Actually, it's a market discussion. They all work. The question becomes, how well they work for the applications engineers use--and what they'll buy/specify. So the market will actually decide. And what the market will decide is that there will be two or three leaders with different technologies. People will confuse the technologies--sometimes on purpose. We'll continue to have material to write about--and have fun doing it. Oh, and as for gigabit Ethernet. Yes, we're starting to see announcements of products. 100 Mb Ethernet is fast enough for almost any industrial application. Gigabit is interesting, but (imho) it'll be some time before there are sufficient products and applications to drive change to it. From Carl's last post: Gigabit Buckeyes. I apparently owe my fellow-Buckeye, Gary Mintchell, an apology for not dragging him down the Ohio State defeat path. Turns out heâo[dot accent]s not an OSU fan (neither am I). Sorry, Gary. I started university down the road a piece from Gary in Springfield, OH and finally got my electrical engineering degree at the University of Akron. I thought my gigabit Ethernet commentary might spark some controversy. So far the only controversy (in the office here, at least) is: Buckeye âo[base "] cookie or candy? Please feel free to weigh in with your comments on either controversyâo[oe] [PTO PROFIblog] 5:04:42 AM    comment []