[Search for users] [Overall Top Noters] [List of all Conferences] [Download this site]

Conference koolit::vms_curriculum

Title:VMS Curriculum
Moderator:SUPER::MARSH
Created:Thu Nov 01 1990
Last Modified:Sun Aug 25 1996
Last Successful Update:Fri Jun 06 1997
Number of topics:185
Total number of notes:2026

156.0. "Programmers' Task Analysis Report" by SUPER::MATTHEWS () Thu Jul 23 1992 10:30

    [not specific to VMS, but...]
    
    Programmers' Task Analysis Report
    
    Commissioned by Andy Sadler and Bill Simcox
    
    ABSTRACT: This report summarizes the tasks performed by programmers in
    Digital's customer base. The results are based on one-on-one interviews
    with 12 programmers in 7 customer sites in the U.S.
    
    The report is available in:
    
    		{SUPER,HARDY}::ES$PUBLIC:PROG_TASK_ANALYSIS.PS
    
T.RTitleUserPersonal
Name
DateLines
156.1NITTY::DIERCKSWe will have Peace! We must!!!!Tue Aug 04 1992 10:4227
    
    
    Interesting report.
    
    Let me preface my comments/thoughts by saying that in "life before
    Digital" (if there really was life before Digital) I taught computer
    science, mainly programming courses, at the college level.  
    
    Our "programming courses", regardless of the language of focus, do NOT
    teach programming skills -- they teach language skills.  There is a
    SIGNIFICANT different.  Assuming that a programmer already has a
    relatively complete knowledge of the skills that it takes to construct
    an application, it should be a relatively simple task for them to take
    their current knowledge and apply it to another language.  I would even
    say that most of a language CAN be learned in one week.
    
    However, to attempt to teaching PROGRAMMING SKILLS in one week would be
    folly, at best.  The type of skills involved require assimilation
    absorbtion time.  I do not see how a programming course could possibly
    be taught in one week.
    
    Maybe I'm reading between the lines, but I'm slightly concerned that
    this survey is causing people with decision making power to shift how
    our "programming" courses are being constructed.  Let's continue to do
    what we do best and deal with the language side of things only.
    
       GJD
156.2We're all in this togetherSONATA::SIMCOXMon Aug 24 1992 17:5470
    I agree that the focus of our programming efforts should be language
    related.  What we are trying to do remains the same, though--Within a
    "languages" framework, is there training that VMS programmers need that
    we are not currently meeting, and that would fit in a logical sequence
    that could be packaged as an additional course offering?
    
    The current VMS programming curriculum looks something like this:
    
    
    
                                             - Vector Proc
                                            |
                                            |
                                             - RMS Structures
                                            |
                                            |
                                             - Parallel Proc
                                            |  Apps
                                            |
    VMS for    ----------------- Utlzng  ----- Util II
    Programmers  |            |  VMS Feat   |
                 |            |  From <>    |
                 |            |              - VAXcluster 
                  - Language -              |  Apps
                                            |
                                            |
                                             - POSIX Programming
                                            |
                                            |
                                             - Internals "String"
    
    
    
    If you look at performance of our most popular Utilizing courses, ie.
    C, Fortran, Cobol, the numbers are as follows:
    
    		    U.S. Revenue	  Worldwide
           		($K)              Students
    		    FY91    FY92      FY90   FY91   FY92
    
      Util C         270     203       735    720    494
      Util Fortran   121     100       849    577    462
      Util Cobol      91      59       428    393    300
    
    If you look at our most popular Languages courses, ie., C, Fortran,
    Cobol, the numbers are as follows:
    
    		    U.S. Revenue	  Worldwide
           		($K)              Students
    		    FY91    FY92      FY90   FY91   FY92
    
           C         725     783             2838   2587
           Fortran   165     152              497    403
           Cobol     166     134              367    298
    
    As for the "backend" courses, apart from RMS, POSIX and the Internals,
    the remaining courses are not even taught (UTIL II is taught once in a
    while in Europe).
    
    VMS for Programmers was introduced in Q3FY92 in the U.S. and
    performance over the last 2 quarters was $274K in Q3 and $360K in Q4.
    
    Thus, given this list of courses and what you know about changes in the
    VMS market and emerging trends within a "languages framework", can you
    offer ideas for additional training needs.
    
    I thank you for all the help you can provide.  
    
     Bill
    
156.3selling instead of buildingUTROP1::EIKELBOOM_JTue Aug 25 1992 05:3989
    Bill,
    
    Donot think of programming course as VMS courses.
    
    People these days want platform independence. So if we teach some C he
    should be able to apply the skills on VMS U*IX or even PC's.
    Positioning courses as VMS courses as you here is cutting you off from
    future markets, as the figures show.
    
    Think NAS, Think of the applications as leading. People have to plan
    for, design and implement and maintain these applications.
    Applications talk to APIs. Yes some of the courses above are API's.
    
    APPLICATION programming is the term. VMS programming will be done a lot
    but is only a very specialized flavour.
    
    The corporate curricula keep reflecting individual developpers
    perceptions, and corporate organization structure.
    
    I doubt however that you will be able to come up with a curriculum
    based on a central case study that shows people how things work
    together.
    
    This year I did a major project for the a new customer coming from an
    IBM background. I did this long track that covered VMS programming
    tools, RDB ACMS DECforms etc.
    
    I find corporate materials very hard to use. WHat I did with these
    people is develop a simple but complete application. 
    RDB material use the personnel database, DECforms uses a bank account
    example ACMS works with a Motel application.
    VMS materials are completely unrelated to any real life situation.
    
    So I had to use my own case study, and use that a flight simulator.
    Suppose they trained pilots the way you train programmers. Would dare
    to fly. here a CBT on engines. here a TBI about air traffic control,
    here is a lecture on instruments, here a video on approach procedures.
    
    NONONONO You put them in a flight simulator and really simulate a
    flame-out on the final appraoch. Powerfull this is really an execise
    you like to do in a simulator, because of its resistance to failure.
    
    The same way with my CASE Study. here is this small project. ANd it was
    run as a real project. The customers manager came in the first day and
    we did a bidding session. The students had to sell their project.
    
    then I delieverd a couple of weeks training and we setup development
    environments applications and in the end there was the running product.
    Small but complete. We did a demonstration for the rest of the EDP
    department including the EDP manager, and senior programming staff who
    had just that week signed the contract with DEC. 
    
    This demo assured them that they had taken a wise decision. they were
    amazed to see our automatic testing and release management procedures
    etc..
    
    There is no need to sit down and become depressed, but the above
    figures just demonstrate what is gonna happen to a company that finds
    engineering multiple overlapping unsellable product more interesting
    than orientate on the market.
    
    THis is exactly what gets us all those this is the syntax and these are
    feature dumps where we need decent training on how to achieve the
    customers goals.
    
    Why dont you just start to sell your stuff. You dont even probably know
    how much of this stuff is running in here in Europe. 
    This company make far too much stuff that is just unsellable.
    Why dont you sell something before you develop it. 
    
    
    That was all the preaching what can we do?
    Lets Find answers to the following questions:
    
    What is the added value of the courses above?
    what are the unique selling points?
    Why would I take C-programming with DIGITAL if I have other platforms?
    WHy is your C or Pascal course better then anyone elses?
    
    
    Responses from all please.
    
    Nice selling
    Johan