Published: February 24 2009, 10:37 AM | no comments
by Reg Harbeck

Wow. I just found out that comments posted to CA's blog site have been "going missing" until yesterday, when this bug got fixed and the comments all showed up.

My apologies to everyone who posted a comment and felt ignored. We're now able to make those comments publicly visible, which is what I've now done to those which were posted in the past, so comment away!

Share this post:  Email

Share

Published: February 24 2009, 10:23 AM | no comments
by Reg Harbeck

I'm going to Austin!

Yes - that Austin: the one in Texas, U.S.A. where the Winter 2009 SHARE conference is being held.

Yes - that SHARE conference: the one where mainframers from North America, and some from other parts of the world, too, come to talk, learn, teach and connect about the mainframe.

Yes - that mainframe: the one that turns 45 years old this April 7, and has been acting as a solid foundation of business computing since its birth.

I'm going to Austin, going to SHARE, going to meet mainframe colleagues that I haven't seen since CA World or last SHARE or even last Winter SHARE. And I'm going to present on a few topics, and moderate a discussion, and help out at CA's station at the SHARE Technology Expo, and participate in some events, and learn from a lot of interesting sessions.

Can you tell that I'm looking forward to it?

I hope to see you at SHARE too - if not this one, then the next one, or even the one after, because CA is one of the biggest supporters of SHARE, and is 100% behind my support of this critically important organization and event, so I plan to continue attending every SHARE for as long as I'm able, supporting it in whatever ways my ability and position allow me to.

BTW, if you're not able to get to this SHARE, you can still get some value from it. There are some webcasts and proceedings being publicly posted at SHARE.org. Check it out!

Share this post:  Email

Share

Published: February 18 2009, 04:00 PM | no comments
by Reg Harbeck

Just over a year ago, I enjoyed a book by Chris Anderson (editor in chief of Wired magazine) entitled "The Long Tail". In it, the author asserts that we're entering a world of more and more customization or broad variety of available consumer goods and a reduced commerce in commodities. Key reasons for this include the Internet, the availability of e-commerce and just-in-time manufacturing, electronic items (such as digital music) which take negligible space to store and can be digitally copied on demand, and the diverging tastes of the new digerati.

This got me thinking and, deliberate dinosaur that I am, over a year later I reached the conclusion that it's worth blogging about how the mainframe was in many ways the original example of this.

Now, you're probably wondering where the expression "long tail" fits in to all of this. Basically, traditional commerce in consumer commodities, predicated on predictive ordering and storing of what the majority of people would want, can be likened to a bell curve that is truncated where it narrows in order to keep things cost-effective.

But, if the majority can now have their uniqueness affirmed and their unique tastes catered to, suddenly the "truncated tail" of the bell curve becomes where the majority of commerce takes place - profitable even with a single-digit number of items of a given type being sold, due to the sheer number of different, individual items and configurations available.

So, what makes the tail of the "mainframe dinosaur" the original long one then (aside from some spectacular fossil analogies available in museums such as the world-famous Royal Tyrrell Museum in Drumheller, Canada)?

Simply this: that from the very beginning, every mainframe environment has been unique and custom.

Of course, the nice thing is that, unlike the unique and custom hardware and software configurations of every individual PC in most large organizations, each of those organizations usually has only one or a few mainframes, so they're able to build a stable and reliable configuration within their organization - at least until they acquire another company that has a mainframe; then the process of merging to a new stable configuration is needed.

But it's a good thing to remember when thinking such leading-edge thoughts: when you're looking for a good example or insight into where we're going in the digital world, where we've been with the mainframe is often a very good place to start.

Share this post:  Email

Share

Published: February 06 2009, 09:31 AM | 2 Comment(s)
by Reg Harbeck

I’ve been involved in a lot of discussions about SOA and the mainframe recently, and it occurred to me that you might enjoy this perspective on it.

Of course, you’ve probably read my blog about “How to Talk Like a Mainframer” at http://community.ca.com/blogs/execio/archive/2008/11/24/how-to-talk-like-a-mainframer.aspx that I wrote this past November.  But one TLA I didn’t mention in that blog was SOA.

Now, over the past year, I’ve discussed SOA in presentations I’ve given around the world, and one question I’ve asked my audiences is, “How do you pronounce SOA?”

And I always get a majority voting for either “ess oh a” (i.e. spell it) or for “so ah” (i.e. rhymes with “boa”). But I can never predict which they’ll choose.

I used to think I knew how to pronounce it, after getting a copy of the book “Service Oriented Architecture for Dummies” (ISBN 0-470-05435-2), a very useful guide, which at the outset prescribed the “rhymes with boa” approach. Life was good.

Then I went to Poughkeepsie. Now, as you may remember from my November blog, the folks in Poughkeepsie like to spell things (like “CICS” for example). Well, a bunch of CA mainframers and I were invited by IBM to get an update on their mainframe strategy, and one of their presentations was about SOA. And I was shocked by what the presenter told us.

He said that we must spell it, because “SOA” (rhymes with “boa”) is a word that means… er… VD (sorry – couldn’t think of a sufficiently obvious euphemism) in a European language.

Now, for some reason, I thought he said a Nordic language (you know: Swedish, Norwegian, Danish, Finnish or Icelandic). So, I went around spreading the word (just the word, mind you) that we were to spell it, not pronounce it, to avoid a certain discomfort.

Then I went to Sweden.

So, here I am in Sweden, presenting to a Nordic audience, and I get to the topic of SOA. And I have to open my big mouth and ask them, “Is it true?”

“No.”

Uh… not in any Nordic language?

“No.”

So, now I’m in Warwick, England, talking to my buddy Marcel from the Netherlands and I mention this embarrassingly rash assertion I’d been making, when he corrected me. In fact, “soa” does mean that – in Dutch! But, don’t worry, he says, we don’t mind. Go ahead and spread the word that pronouncing it to rhyme with “boa” is OK.

So, I’m listening to the industry experts and various executives present about it, including people from both IBM and CA, and guess what? There’s no predicting whether a given presenter will spell it or pronounce it.

Which, by the way, makes “SOA” the perfect TLA.

Why? Because TLA has two meanings: Three Letter Acronym if the three letters spell a word that you can pronounce, and Three Letter Abbreviation (see my July 24, 2008 blog on the topic at http://community.ca.com/blogs/execio/archive/2008/07/04/tla.aspx) if you just spell it. Well, apparently, SOA is both!

So, the last question then is, do you care? In other words, is SOA something your organization is looking at seriously enough that you’ve evolved a corporate standard way of pronouncing it, or at least a debate about it?

Share this post:  Email

Share

Published: February 02 2009, 07:10 PM | no comments
by Reg Harbeck

Today is Groundhog Day, and I’m reminded of the movie with that name, in which Bill Murray plays an arrogant reporter who gets trapped on that date, waking up each morning back at the beginning of the same day, until he clues in and becomes an honestly caring person.

Another reason I’m reminded of that movie is because of something else that goes round and round: the wheel. And I don’t just mean that it rolls in its natural motion. I also mean that we keep rediscovering it – or reinventing it, as the case may be.

Let me give you an example from my own experience. Just over eleven years ago, when I joined CA (back when it was Computer Associates), I began to travel regularly. My laptop case had a shoulder strap, my small suitcase had a handle, and my large suitcase may have had some very small wheels, but they didn’t work very well. So I had to do a lot of carrying.

Then, a revolution of sorts occurred: it turned out that everyone, myself included, was tired of lugging our luggage. Suddenly, a new sort of laptop and carry-on case rolled out, with wheels! I can’t tell you how much wear and tear this has saved me, but I’m sure it’s substantial.

But here’s the funny thing: the wheel has been around since the dawn of human history – maybe longer. Yet we’re still discovering new, obvious uses for it.

Which brings me to what I’ve been really writing about all along here: the mainframe. After all, every time a new innovation in quality computing comes along (for example, virtualization), it turns out that the mainframe has already been doing it for quite some time. And that it’s doing it better than any other platform.

Which is not meant to mock those who put their workloads on other platforms. Just to remind those of us who use mainframes to rediscover it, just like we can rediscover the wheel even if we use wheels for all sorts of things every day.

And then, once you’re thinking about all the useful things that can be done with something so tried-and-proven, make sure to tell the non-mainframers you know about it. You could start something that will benefit all of us.

After all, one good turn deserves another!

Share this post:  Email

Share