Posted by: wiltjk | January 3, 2011

OLD IT, not SKYNET, will Kill the Enterprise

It’s a new year and now the time for the oversell of health club memberships, optimism for a better year, and predictions. I have to say, I will only offer one prediction this year:

We will begin to see the eminent fall of large organizations, not because über intelligent neural systems will end them, but because their dated/stagnate IT that is now on life-support will fail them, incapacitating them to be easily overtaken by smaller, more agile competitors.

Please don’t mistake me for a mainframe-hater. I have my career to thank for their existence. Upon graduating with a degree in Physics, I moved to Pasadena, CA to work on the Data Communications module of the Burroughs Medium Systems (B27/8/900-B47/8/900) Master Control Program (MCP) operating system.


It was great working with some of the most impactful technologists of our time building OS modules for what later would be known as the villain of the initial TRON movie (the MCP) and developing CANDE, the precursor to today’s user interfaces and IDEs.

But, let’s be honest here, that all should have been history by the end of the 1980’s. Right?

Well, by the end of 2010, I met face to face with more mainframe based 3090 systems than I ever was exposed to in the 1980’s

  • Flying home from an engagement in October, I was stranded in Amsterdam due to high winds. The airline system being accessed from a high-tech flat-screen PC to rebook my flight and select my seat was via a 3270 terminal emulator complete with half-duplex transactions using command line keywords.
  • A few weeks ago in Bangalore, I went to five (5) different book stores looking for Switch: How to Change Things When Change Is Hard by Chip Heath & Dan Heath. Not finding it on the selves, I had each store look it up on their inventory system. Again, it was a PC running a 3270 emulator with a keyword based UI that displayed results in pages of results. 5 different stores, 5 different inventory systems, all 3090 based applications!!!
  • Recently, I was purchasing light bulbs at a major hardware store chain, and noticed that the cash registers all ran 3270 terminal emulators. When supplying information, the tell-tale segmented underline fields were used.
  • Earlier this year, my wife and I were ordering new kitchen appliances at a major retail chain. Again, the system used yielded screen after screen of 3270 data input, ALL IN UPPER CASE!!!

I had no idea that IBM maintained such a large infrastructure presence in business!

These large organizations are all literally relying on their inventory, POS, scheduling, delivery, and many other business systems to be run from hardware and operating systems that literally are on life-support.


Why life-support?
(a) The physical hardware has not been manufactured in years
(b) Few exist who can support them (most retired, many no longer around)
(c) Nobody with hopes of improving their career will ever desire to work on them

Organizations that wear these archaic systems as a badge of strategic IT efficiency (i.e., impervious to viruses) will be the ones that later make headlines like:

1024 Flights Grounded Indefinitely as Airline transfers back to Manual Bookings While New System is Built 

After 128 Years of Service, Another Retail Giant is Forced to Close its Doors After IT System Crashes

Last COBOL Developer Leaves Retirement for $125M Salary Eclipsing Tiger Woods


The New Old-Mainframe
I have not only seen the typical old-mainframe systems in use around the world, but also the new old-mainframe systems that are headed down the same path:


I was recently at a research facility for a leading technology firm. The technologists had the latest must-have hardware to be envied by any Geek, and yet they all were running Windows XP.

When I went down the path of “Why???,” it was quickly pointed out that they had at least one mission critical application in their portfolio that could only run on XP. This is a primary reason enterprises avoid getting out of the rut of everything Windows XP with the next most popular excuse being they have tens of thousands of units that make supporting multiple OS versions too difficult a task to even attempt.

To this I ask:

  • Microsoft has given the special capability to run specific applications in XP-Mode through virtualization in Vista and Windows 7, why is this not an acceptable means to this requirement?
  • How many resources and how much time does it take to make the latest hardware run Windows XP counting in 64-bit hardware and all the drivers required?
  • Is the weakest link – that one mission critical XP application -or- is it really an IT department incapable of continuous IT improvement such that it cannot embrace and incorporate modern technology?

When I surmise all these antiquated systems that literally “run” the enterprise, I find myself realizing that in many organizations, IT is a major cost center. It not only is a financial burden to the enterprise, but it also is fuelling the stifling of innovation and agility that may contribute to the ultimate demise of the organization.

Who really needs such a self serving IT department?


Is your IT the real SKYNET?
(or, Is it time for new IT for a business to survive & be competitive?)
If SKYNET is smart machines taking control of your world, I sometimes wonder if IT organizations might be the same to the enterprise, but instead of an intelligent take-over, it’s more from the perspective of their limiting or dumbing down of capabilities to choke the business rather than to make it easier for the business breathe.

If your IT department is not moving toward becoming a profit center, then yes, it may be time to replace them.

If your new projects are not developed with an upgrade/advancement plan three years after deployment, then it may be time to clean house.

If products like the new tablets and phones emerging are banned from the enterprise because there is no IT support or they are viewed as a threat, then it may be a good time to find IT support that embraces it.

If your IT development staff is not begging to build solutions using the new devices emerging over the past couple years, then you may be at risk of stagnating the business to fit in the bounds of your IT department’s limitations.

If you think (or are told) your business is so unique that it needs to have a very specialized IT organization, then consider the following:

  • Every business considers themselves totally unique; its what gives them their competitive advantage.
  • Every business that is unique has been able to train their IT staff enough about the business to get them to their current level of efficiency.

What-if there was a fix?
What-if the IT staff took the initiative to become savvy not only in technology, but also in business practices and the actual business domain in which they serve?

What-if they could clearly articulate the the business of the business and communicate the strategy and value of a solution in strictly business terms?

What-if, due to their unique intimate knowledge of the business processes in which they have built existing systems, they were to become innovators and identify improved business strategies around these business processes – outside the normal bounds of IT?

What-if your IT did all this, could they then be considered a profit center?


Peter Till’s artwork (above) does a great job depicting what we expect of our portfolio of enterprise solutions – namely to think on behalf of our business – perhaps so that we don’t have to.

This is wrong! People solve problems, computers execute algorithms. Our IT organizations need to be in a position of performing the former and not blaming the later when they fall short.

My work has brought me in touch with organizations where they are seriously training their IT staff in the business domain as they recognize them as uniquely qualified problem solvers that can actively contribute to the business in this capacity. This enhances their ability to align the technologies that will support people-solved problems making technology the tool, not the solution.

These organizations are looking at this investment to turn their IT into a profit center. But it goes much further than what appears on the surface. If their IT is now an internal profit center through innovations and greater efficiencies in the business, what if they sold this expertise to other organizations who are burdened by their cost center based IT organizations? Now, this business/technology knowledge pairing becomes a product that can be sold as a service. Now, IT definitely achieves profit center status!

Ironically, the best way to avoid the real-world SKYNET through dilapidated IT hardware and personnel is to embrace the initial motivation behind the movie entity where continual investment in, improvement of, and advancement of our IT people and systems is paramount.



  1. Excellent post!! Very informative and easy to understand. Looking for more such posts!!

  2. Excellent post, Jim. I’ll point out that the danger of “Legacy” is not just confined to IT. Enterprises have all kinds of legacy, from IT to processes to business models. Consider, for example, the record industry, who fought change to the point of suing their own customers!

  3. Dear Sir,

    Please consider your misleading or incorrect statements.
    Why life-support?
    IBM in 2010 release a new Line of Mainframe systems Z10 (1.5TB of memory, 64 Cpu’s – multi-core) It also has VM capability for running several OS’s: MVS, AIX, Linux(RedHat, SUSE), and Windows is coming soon. Please educate your self on the the Architecture ( IBM Mainframe’s are currently unsurpassed in I/O operations, something all other systems to date have a issue with. If you are talk about 3290’s Terminal they have been replaced with software solutions now for PC’s or Thin WinTerms. (It’s an easy switch, however companies are cheap and will keep stuff forever. )
    There are people around that do support this that are not retirees. (We have 100 of them. ‘Fortune 500’)
    Yes it’s not as glamors as building the next facebook, but you do get payed well for your skill set. (Plus facebook has not turned a real profit, remember the internet 90’s. They also count on mass cash infusions to stay a float)

    • Your (A) comments on the Z10 are absolutely correct and there is no argument IBM maintains a leadership role in the enterprise.

      Likewise, you have valid observations in (B) and (C). However, with 9,007,346 Java developers (based on and presumably as many.Net developers, I would surmise your 100 CICS and VM/CMS resources are both a rare and highly sought after commodity well worth the premium they should receive for their skills.

      When I revisit my own PL/I code (never did much in COBOL), I shutter at the thought of losing the OO constructs, workflow frameworks, garbage collection, and integrated IDE/debugging environments I today take for granite 😉

  4. […] This post was mentioned on Twitter by James Sparenberg and howard pyle. howard pyle said: antiquated IT infrastructure is more likely to kill the future than evil skynet type systems. great article: […]

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s


%d bloggers like this: