Middleware Matters

Last weekend was amazing! (Yes, I know it's almost a new weekend, but I've been traveling and have been unable to write about last weekend until now.)

I played in the Narragansett (Rhode Island) Beach Jam, a flying disc (frisbee) freestyle tournament, and my teammate Carl Emerson and I managed to take 4th place. Not bad, considering the fact that 5 of the top 30 and 10 of the top 50 players in the world were also competing. A very cool aspect of the tournament was that Joey Hudoklin, widely considered to be the greatest player in the history of the sport, was competing (his team took 2nd). He won his first world championship in 1979, and was a member of the Bud Light Frisbee Team in the 1980s. Yep, he basically had my dream job: getting payed to play frisbee! He was an amazing player then, and is still an amazing player now. I had never met him before, so it was quite an honor for me to hang out with him.

But right up there with taking 4th place was the fact that Joey and Rob Fried, another fantastic jammer, are both also amazing guitar players, and I happened to have my harmonicas with me, so we played a couple sets -- blues and a lot of Hot Tuna -- for the party after the tournament. We'd never played together, but you wouldn't have known it. It was a blast, and judging from the compliments we got, everyone really enjoyed it.

Just returned from a nice ten days touring the Irish countryside with my family. In the nearly eight years I've worked for IONA, this is the first time I ever went to Ireland for something other than business. We first attended the wedding of our friends Brian and Nataly, which was most excellent. We then drove around to see places like Connemara and the Ring of Kerry, and went to meet my Irish relatives in County Cork for the first time (both my maternal grandparents came from there). Nice people! The weather most of the time was surprisingly good, with lots of sunshine and warmth.

We finished with a day and a half in Dublin, where among other things I visited my favorite Dublin pub, Toners, and also sat in on harmonica with "The Business," a great Dublin blues band that plays every Wednesday night at Bruxelle's (on Harry St. just off Grafton St.). Some of my IONA colleagues came down to listen -- thanks guys! The only drawback was that I couldn't hear myself at all (no monitor), and so I had to play by feel, which meant keeping it conservative. It was fun nevertheless, and getting to watch Pat, their great guitarist, really close up was most excellent. I can't wait to join them again on my next trip.

A recent webinar given by my colleague Peter Cousins, the technical director for Artix, is currently featured in an ebizQ article about SOA. The title of the article is "Integration on the Edge" (which I borrowed for the title of this blog entry). Definitely worth the read.

Why is it that IT shops seem to continuously try to convince themselves that it's possible to create and maintain homogeneous systems by simply buying into the "right" technology du jour? Peter explains how traditional integration approaches are always oriented around gluing services together after the fact (typically in a slow and costly hub-and-spoke fashion), when it's much less feasible and much more expensive, rather than just making them integration-ready themselves. He explains how moving integration "to the edges" results in systems that avoid technology du jour lock-in and more easily deal with inevitable and continuous technology and business changes, thus making it easier for you to build and maintain a true service-oriented system that actually delivers real business ROI.

My latest "Toward Integration" column from IEEE Internet Computing is now available. This one, entitled "Dark Matter Revisited," takes a look at three different middleware frameworks and applications written in dynamic languages such as Python and Perl. (I've provided the usual PDF version as well.) As always, any and all comments and feedback welcomed.

JAOO is the best technical conference I know of. All the speakers they invite are known to be working on cool stuff and known to be able to give good talks. They throw an amazing conference party, with the famous Danish chef "Chili John" Rasmussen cooking a wonderful buffet for 750 that's better than many restaurant meals prepared for only one. The party also features a rock supergroup called Absolute Girls made up of some of the finest female rock musicians in Denmark. How mint is that? In the conference schedule, they mark any talk that looks like a product pitch or marketing presentation with a red mark so you know to avoid it. The conference organizers are always on top of everything.

I was hoping to get invited to speak again this year, and I'm happy to say that I recently got my invitation -- yes!! I haven't come up with a title or abstract yet, but I'll likely speak either about multi-middleware services or about web services notifications.

Are you going? You should!

Last night I had the pleasure of seeing Paul Rishell and Annie Raines at Johnny D's in Somerville, MA (outside Boston). All I can say is WOW!! I believe Annie might be the best harmonica player I've ever seen. She was like Little Walter and Big Walter and Sonny Terry and Sonny Boy Williamsons I and II all rolled into one. Her control, her versatility, everything was just perfect. I've seen many of the greats -- e.g., James Cotton, Charlie Musselwhite, Rod Piazza, Jerry Portnoy, Carey Bell -- but I believe last night Annie was better than any of them.

About 15 years ago I went to see James Cotton at Harper's Ferry in Allston (also outside Boston). James, of course, was Muddy Waters's harmonica player for many many years, and is a music legend. I clearly remember a young woman standing right up front, intently watching every move James made. I thought it was a bit unusual, since most of the audience was male. At one point, though, he called her up on stage, seemingly out of the blue, and she got up there, took his harp, and just started wailing. That of course was Annie Raines. Back then she was just getting started, but now she's in the same league as James Cotton. And that's saying a lot!

Paul Rishell is also incredible. He does some amazingly difficult guitar work and makes it all look oh so easy.

Go see these two if you can!

I noticed that over on Loosely Coupled, Phil Wainewright has included Artix in his list of top ESBs. That's cool, especially given that it appears that he made his choices based on seeing actual products shipping to actual customers. Customers who use Artix are generally quite happy with it.

It's also good that Phil so clearly explains the fact that the term "ESB" is somewhat loose, since I'm not sure I would describe Artix as just an ESB. The "bus" part of the term ESB seems to imply a single transport and protocol, but Artix is carefully and uniquely designed to focus service/application interactions at the WSDL logical contract level rather than at the WSDL physical binding level. And keep in mind that WSDL does not always imply SOAP. In other words, you write your app to the service contract, rather than instantly making it obsolete by tying it to particular transports, protocols, or message formats, and instead let multi-transport-multi-protocol-multi-format-but-still-fast-and-efficient Artix worry about the best way to bind and communicate.

Another flying disc (frisbee) freestyle tournament this past weekend, and another nice result! This time, my son Ryan and I teamed up with veteran jammer Steve "Boston Pops" Scannell -- nicknamed "Pops" because he's, umm, a bit older than most jammers ;-) -- for the New England Overall Flying Disc Championships. This event involved a number of disc sports, including disc golf which I try to play about once a month, but we entered only the freestyle competition.

Steve and Ryan played really well, and I did OK, and our overall effort was good enough for a 2nd place finish. Pops was especially outstanding, hitting a number of difficult moves in both the preliminaries and in the finals. I haven't seen him jam that well in competition in awhile. It was definitely an honor for Ryan and me to team up with a player of his caliber. He may be a bit older, but it sure doesn't show! And I don't have to be his father to say that Ryan is without question one of the hottest young freestylers around, hitting very advanced moves after only 3 years of playing, and yet he's still getting better each and every week -- kinda scary!

That might be it for freestyle competitions for this year, unless Ryan and I go south for the Tennessee State Championships in September.

I really like the Daily Python-URL website, as it's an easy way to keep up with everything people are doing with and saying about Python. I was surprised to find that my latest Internet Computing "Toward Integration" column on middleware "dark matter" -- distributed computing frameworks and tools written in dynamic languages -- appeared in the July 28 listing there. Cool!

If you use Python, even just occasionally, I highly recommend a daily dose of Daily Python-URL.

Radovan recently posted a compelling argument regarding the shortcomings of ESBs. The main reason I like it and largely agree with him is because of what he says about the power and utility of moving integration capabilities to the endpoints.

So many integration projects have taken a centralized approach that it's led many to believe that centralized integration is the only viable integration solution. After all, with a centralized/bus approach, the integration problem is obviously greatly reduced, since each endpoint need only integrate with the bus rather than integrating with each of the other endpoints. In other words, the bus affords an O(N) solution rather than an O(N^2) solution.

Hypothetically, the reduction in connections offered by the bus approach is certainly true, but in practice, the centralized/bus cure is often worse than the disease, for several reasons:

• Centralized approaches force the existence of canonical protocols, transports, and data formats, which adversely affect efficiency. Everything being integrated must be able to convert its own protocols, transports, and data formats to and from those of the bus. This means that to integrate System A with System B, in the best case you require two conversions: A to canonical, and canonical to B. Worst case is when A and B have protocols, transports, or data formats in common, thus requiring no conversions, and yet the canonical/centralized approach forces the conversions to occur regardless.
• Since everything has to connect through the bus, the overalll system capabilities are limited by the capabilities of the bus and its canonical formats, protocols, and transports.
• Buses are typically proprietary, though with the advent of web services, this is improving.

If you instead move integration capabilities directly to the endpoints themselves, they can speak directly to each other. If they have protocols or formats in common, there's no need for any conversions -- they can just talk directly. If their data formats are different but not by much, then just a little bit of data mapping is required, rather than requiring potentially heavy-handed conversions into and out of a canonical format in the middle. They're free to use whatever protocols and data formats they choose, rather than being limited by the capabilities of the bus. Likewise, the speed and efficiency of their connection is not limited to that of the bus.

"Ah," you say, "but connecting endpoints directly to each other brings us back to the N^2 connection problem." Hypothetically, yes. In practice, no. In practice, nobody ever needs to connect everything to everything else. The N^2 problem just doesn't arise. Installing a bus to solve the hypothetical problem and trading away all possibilities for speed, efficiency, and flexibility in the process is often an extremely poor choice.