• The Four Hundred
  • Subscribe
  • Media Kit
  • Contributors
  • About Us
  • Contact
Menu
  • The Four Hundred
  • Subscribe
  • Media Kit
  • Contributors
  • About Us
  • Contact
  • The AS/400 at 19: Predicting the Future–Or Not

    June 21, 2007 Timothy Prickett Morgan

    As a young student of math and engineering a few decades ago, I was drawn to these topics because they offered what seemed to my eyes a calculable and assuring degree of certainty in the answers they provided. I didn’t think about it at the time all that much, but math was particularly fulfilling because it had answers, and given a set of data and conditions, you could predict future sets of data and their conditions. In that idealized world, you can predict the future.

    Of course, the more I studied complex systems–either physics or aerodynamics–the more I got this sick feeling that some of the key assumptions that people put into their equations to make the numbers work out were, at best, educated guesses and, at worst, circular reasoning that did not bridge the gap between the whiteboard or the computer program to reality. Now, many years after abandoning my engineering career and living life in a much more messy world than the young engineer I wanted to be would have thought he could cope with, here I sit on the 19th birthday of the AS/400, the machine that has, more than anything, defined my knowledge of the world, provided my peers and partners, given me a sense of purpose, and put food in my mouth and a roof over my head, and I am convinced that it is not possible to predict the future of the System i platform in a way that will be fulfilling to most of us.

    I don’t know about you, but I find the inherent unpredictability of the world a little disconcerting. This is not what I thought the world was, but it is what the world clearly is, based on the empirical data. And when I say unpredictability, I don’t mean the kind of new-agey, quantum unpredictability that Heisenberg was going on about in his Uncertainty Principle, because at large scales, this unpredictability doesn’t seem to change the hardness of objects nor their tendency to remain objects that seem to follow the rough equations of Newtonian physics–unless you can throw a baseball at near the speed of light, of course. Then, things seem to break down a bit and go all Einstein on you. But, even relativity and string theory and lots of other things I read about when I get a spare moment don’t seem to help much. The grand unified theories are not particularly satisfying. They are almost unprovable, and as for their predictive ability, I am not impressed.

    And last week, after reading a marvelously un-academic book called The Black Swan, by Nassim Nicholas Taleb, a former Wall Street “quant” who seems to know more about doubt and the weakness of predictive models than anyone I have ever heard of since studying a little Hume and Locke back in college, I am convinced that my own sense of history, learned from the school of hard knocks mostly, is probably correct. The things you can’t predict will change everything, and the things you thought you predicted correctly will probably turn out differently than you imagined.

    When it comes to the future–and I bring this up because the OS/400 and i5/OS community has endured two decades of prediction and prognostication–it is very hard to be sure of anything. And that means the safest position is not in a bunker somewhere, but in keeping an open mind and keeping an eye out for opportunities.

    The fact that the AS/400–call it what you will–has been around for as long as it has is not a function of just prediction, but being open to new possibilities and chasing new opportunities. Those of us in the OS/400 and i5/OS community tend to idolize the founding technologies that were embodied in the System/38 back in 1979, the System/36 back in 1983, and the AS/400 back in 1988, much as Americans tend to worship the Founding Fathers. We all know the innovative technologies that are woven into the AS/400 and that are still carried on by the System i platform: virtualized hardware, an integrated relational database management system, an integrated set of programming languages that have native and fast access to the database information. RPG is a great language for business logic, and the 5250 protocol for letting applications feed data to end users was extremely efficient. The whole shebang of processors, IOPs, and other co-processors inside an AS/400 allowed some modest machinery to do a lot of useful work.

    And I don’t think for a second that even IBM thought the AS/400 would do as well as it did. And I don’t think IBM has any better skills than you or I to figure out how the System i will do in the future. IBM gets to try technologies and tactics, it gets to be in control of the initial conditions of each System i product cycle. But that is a far cry from being able to predict what more than 200,000 customers worldwide will do or not do. So far, they seem to be investing in lots of other technology besides i5/OS. There are reasons for that, not the least of which is the perception that other computer platforms are less expensive. (I could argue that they are not, and have from time to time.) The cheapest tool to use is the one you know how to wield well–provided other tools are not ridiculously more useful or less expensive. If anything has hurt the AS/400, iSeries, and System i market in the past decade, it has been the shift of programming from the back office to the Web browser, where there are lots of different ways to get an application screen to fill up with data.

    So as the AS/400 turned 19 last Thursday and prepares for its final teen year, it was tempting, so very tempting, to make predictions. If you look at the essay that I usually write this time of year, I haven’t indulged in too many predictions, although I did argue vehemently last year that it was time for some enterprising private equity group to buy the System i business and remove it from IBM’s control, just so we could change the initial conditions in a way that would be competitive. I make similar arguments all the time about how IBM could do this, that, or the other to change the conditions to better meet the market. That’s because I am a tinkerer at heart. You try things to see if they work, and you keep an open mind.

    That’s why I am encouraged when I see IBM put out user-priced entry System i5 machines when it was pretty clear that the company was perfectly willing to sit on its thumbs and not to anything until Power6-based machines are expected in early 2008. While I have argued in the past few weeks that IBM’s pricing models do not match those of competitors in terms of the price per unit of work or the price per user, it is a step in the right direction. Whether or not this helps the System i business grow or not remains to be seen. I am hopeful, as always. And the fact that so much private equity is coming into the System i market in the past year–hundreds of millions of dollars that we can see, and heaven only knows how much more we haven’t heard about–is also encouraging as well as cause for concern because any big change is always cause for concern.

    I am, like you, both excited and maybe just a little nervous by the possibilities for the future of the AS/400. But, then again, I have been in this state for many, many years. No matter what, I plan to keep an open mind.

    RELATED STORIES

    Happy 18th Birthday, AS/400; Time to Leave the Nest

    The IBM Systems Agenda: iB(M)

    Happy 17th Birthday to the AS/400!

    The AS/400: 16 Years of Bending, Not Breaking



                         Post this story to del.icio.us
                   Post this story to Digg
        Post this story to Slashdot

    Share this:

    • Reddit
    • Facebook
    • LinkedIn
    • Twitter
    • Email

    Tags: Tags: mtfh_rc, Volume 16, Number 25 -- June 25, 2007

    Sponsored by
    DRV Tech

    Get More Out of Your IBM i

    With soaring costs, operational data is more critical than ever. IBM shops need faster, easier ways to distribute IBM applications-based data to users more efficiently, no matter where they are.

    The Problem:

    For Users, IBM Data Can Be Difficult to Get To

    IBM Applications generate reports as spooled files, originally designed to be printed. Often those reports are packed together with so much data it makes them difficult to read. Add to that hardcopy is a pain to distribute. User-friendly formats like Excel and PDF are better, offering sorting, searching, and easy portability but getting IBM reports into these formats can be tricky without the right tools.

    The Solution:

    IBM i Reports can easily be converted to easy to read and share formats like Excel and PDF and Delivered by Email

    Converting IBM i, iSeries, and AS400 reports into Excel and PDF is now a lot easier with SpoolFlex software by DRV Tech.  If you or your users are still doing this manually, think how much time is wasted dragging and reformatting to make a report readable. How much time would be saved if they were automatically formatted correctly and delivered to one or multiple recipients.

    SpoolFlex converts spooled files to Excel and PDF, automatically emailing them, and saving copies to network shared folders. SpoolFlex converts complex reports to Excel, removing unwanted headers, splitting large reports out for individual recipients, and delivering to users whether they are at the office or working from home.

    Watch our 2-minute video and see DRV’s powerful SpoolFlex software can solve your file conversion challenges.

    Watch Video

    DRV Tech

    www.drvtech.com

    866.378.3366

    Share this:

    • Reddit
    • Facebook
    • LinkedIn
    • Twitter
    • Email

    What Is SMIOSTCPGT and Why Is It Eating My System? IBM Kills Off System i ServerProven, Standard Edition Rebates

    Leave a Reply Cancel reply

TFH Volume: 16 Issue: 25

This Issue Sponsored By

    Table of Contents

    • IBM Offers Virtualization-Friendly Pricing for RHEL 5 on Power
    • MPack Hacker Tool Claims 10,000 Compromised Web Sites
    • Lawson Expects Better Results for Fiscal Q4 Than Anticipated
    • Database Sales Grew By 14.2 Percent in 2006, Says Gartner
    • IBM Previews Virtualization Management Tool for Power-Based Boxes
    • IBM Offers Virtualization-Friendly Pricing for RHEL 5 on Power
    • The CIO Is the Hammer, and Everything IT Vendors See Are Nails
    • As I See It: Dare to Be Rich
    • VoIP and the Search for Single Points of Failure
    • IBM Kills Off System i ServerProven, Standard Edition Rebates

    Content archive

    • The Four Hundred
    • Four Hundred Stuff
    • Four Hundred Guru

    Recent Posts

    • Meet The Next Gen Of IBMers Helping To Build IBM i
    • Looks Like IBM Is Building A Linux-Like PASE For IBM i After All
    • Will Independent IBM i Clouds Survive PowerVS?
    • Now, IBM Is Jacking Up Hardware Maintenance Prices
    • IBM i PTF Guide, Volume 27, Number 24
    • Big Blue Raises IBM i License Transfer Fees, Other Prices
    • Keep The IBM i Youth Movement Going With More Training, Better Tools
    • Remain Begins Migrating DevOps Tools To VS Code
    • IBM Readies LTO-10 Tape Drives And Libraries
    • IBM i PTF Guide, Volume 27, Number 23

    Subscribe

    To get news from IT Jungle sent to your inbox every week, subscribe to our newsletter.

    Pages

    • About Us
    • Contact
    • Contributors
    • Four Hundred Monitor
    • IBM i PTF Guide
    • Media Kit
    • Subscribe

    Search

    Copyright © 2025 IT Jungle