Thursday, April 09, 2015

Could I have saved my IT career and have better karma now? Probably so.


After my “career cardiac arrest” at ING-ReliaStar in Minneapolis (at age 58), in the post-9/11 (and post dot-com crash) world of December 2011, I did find that the culture of the “interim” job market was indeed a big shocking. There was a great deal of hucksterism, of companies trying to hire people to knock and doors, phone them and email them (and now, use their social media “friends” and “followers” lists as leads) to sell them things.  Companies seemed oblivious to the idea that people don’t want to be pimped.
  
Actually, that’s all part of the cultural change that has gone on with the Internet, the “alone together” idea (Sherry Turkle’s book, in the Book review blog Nov. 2, 2011).  In my parents’ era, it was more socially acceptable to use your friends for business. 
  
I can remember being taken out to lunch by a life insurance agent in Merrifield VA (while it was being redeveloped) in 2007, and wondering, is this the life I was supposed to have, pampering people?  I don’t do that.
  
Could I have avoided all this by remaining more technically competitive as I grew older?
  
Consider how the mainframe market behaved back in the good old days, the 70s and early-mid 80s.  Companies did their own in-house systems in COBOL, mainly (but in Assembler and on older platforms like DOS more often than you think).  In time, purchased systems from application vendors became more prominent.  These ranged from Infonational (accounting), which NBC used in the 1970s, to CFO, then VLN and finally Vantage in Life Insurance, to Hogan for banking, and MSA (DB) for accounts payables.  Slowly, the market shifted to these vendors.  Originally, the market wanted basic IBM skills (COBOL, JCL, VSAM, CICS, and database which used to prefer IMS before IDMS and relational DB2 became more important in the 80s).  People who came from non-IBM mainframes (like Univac in my case) had to “get IBM”.  In time, the market tended to want deeper specific experience with these packages, especially in health care (like Medicaid MMIS). 
  
 Then, the market started to contract in the late 1980s with the wave of hostile takeovers and data center consolidations.  Curiously, after Reagan, defense was getting stronger again, especially in the mini world of Vax/VMS.
  
In the 1990s, I worked for a smaller life insurance company in Arlington (USLICO, to become ReliaStar and finally ING and Voya).  USLICO had “everything”, up to a point, although most of the IDMS was through VSAM transparency.  So did ReliaStar in Minneapolis.  In the 1990s, the expectations of retrofits for Y2K kept the mainframe market strong.  But the culture became even more specialized.  We used DB2 and Vantage, but tended to use outside consultants to write all “the modules” needing deep technical expertise.  The inhouse people were slowly getting “less technical”.
  
After Y2K and then 9/11, the market had changed.  Mainframe “gigs” tended to demand much more specialized or concentrated skills and be short term, often involving temporary relocation and living in a corporate apartment on a per diem.  (That doesn’t help “family values.”)  MMIS jobs tended to require five years of MMIS itself.  Had I not left Bradford in NYC and moved to Dallas in 1979, maybe I would have built up the five years and stayed in MMIS even until today (merged with Obamacare, which would give MMIS projects a lot of work).
  
I probably blew a chance to move into Vantage in the early 1990s, had I not been so preoccupied with what I perceived as an improperly done elevation on an in-house system in 1991.  The details are somewhat interesting, and are mentioned in my DADT-3 book, and I may give more background later on  my DADT notes Wordpress blog.  In the early 1990s, Vantage ran slow, and VLN (from a Texas entrepreneur) was perceived as a legitimate competitor.  But VLN failed as a company, and Vantage took over the race, and by the late 1990s was a powerhouse in mainframe legacy life insurance, annuity, and new business processing, batch and online (CICS), with a choice of DB2 of VSAM databases.  By the late 1990s, Vantage cycles ran efficiently.  After Y2K, the job market specifically for advanced Vantage skills would be strong.  “Vantage rules the world”, at least in life insurance legacy systems. It’s now CSC (link ).

I may have also had an opportunity to learn java programming in 1999, and worked on the mid-tier team, and actually gotten away from Y2K work, had I been more “alert”.  At the time, I was recovering from the hip fracture accident (which actually went very well; the University of 
Minnesota was the best place in the world to treat something like this).  I was very interested in matters related to my book, to the website that put it online, and then to COPA.  I was also seeing early distractions from my mother’s health issues, which by the spring of 1999 were a big-time problem. 
  
In managing my own websites (at the time, it was “hppub.com”, mirrored on AOL Hometown) I kept the technology very simple.  I used simple HTML, and found I didn’t even need to bother with Metatags to get it indexed in search engines.  So I didn’t learn the scripting that commercial sites use (and that is becoming more necessary for mobile sites today).  However, at the same time, at work, the effort in 1999 was mounted to code a “mid tier” data access layer in java.  Had I been more alert, I might have gotten myself on that team.  There is a big advantage to learning an OOP language when putting up a new system that you code yourself.  It is very difficult to learn a language maintaining something in retrospect.
  
Getting real proficiency in OOP coding would have avoided some of the “problems” I wound up with.  That is, being pushed into hucksterism, or into areas that were overly personal in the demands of others (as I found out as a substitute teacher), or having the questionable “karma” of dependency on “inherited wealth” now (after mother’s passing) which could become the target of criminal or political (or a combination of both) expropriation.  My circumstances could have been much healthier and I could be much more knowledgeable about matters like security and moving quickly to newer content-management platforms for my own work.
  
I did move over to “client-server” in early 2000, winding up in “CSW Support” and on a 2-person team to “fix bugs”, a third tier that became redundant when the layoff came.  But it is difficult to learn Powerbuilder and java fixing work of others (I did take a one week course in each.)  In the intervening years, java (once very popular in retail systems) has become somewhat suspect over security concerns, despite its amazing rise in popularity from 1997 to 2002.
  
We can also ask, how to kids learn to become “supercoders” like Mark Zuckberberg?  (Remember the line “Kids’ Stuff” in “The Social Network”?)  It seems to be like playing piano, it is much easier to become proficient and quick when you are about 14 than when you are 60.  Lots of kids overseas have become supercoders, but the economies in places like Russia don’t give them legitimate jobs. Instead, Vladimir Putin (was well as the Chinese government) seem to look at their teen hackers as their new soldiers.
  
Yes, having the skills of a coding prodigy is a good thing in the US, Canada and the West today.  If you’re in college, it can pay your way, avoid debt, and let you start your adult life at 18, on your own.  Of course, the skills in the workplace aren’t quite the same as what academics require.
  
In the meantime, I’ve noticed that the mainframe job requisitions by email from recruiters in the past two years have become more generic.  That seems to be cause of the demands create by the cavernous complexity of Obamacare.   The Supreme Court this spring could have an unpredictable effect on the future market. 

No comments: