"Roland Perry" wrote in message
...
In message , at 20:39:08 on Sat, 14 Jan 2017,
tim... remarked:



I had this discussion very very early on in my career when real time
development was done on specialist development systems and moved to being
done on IBM PC's

Recruiters would say nonsense like you don't have PC experience, my client
can't use you - to which I could respond "but I do have development on PC
experience" (though I was pretty unique in that at the time), but if I
hadn't they would have been saying to a 28 year old graduate "you can't be
retrained to use a simple computer interface" (that all our, left school
at 16, admin staff are being retrained to use) - Utter Bull ****!

I agree that in the environment you describe, the skills are portable, and
attitudes of employers and recruiters had not yet moved on. The "whole new
way of working" (which was probably coming in when you started your
career) was to program in high level languages rather than assembler.

Thus it doesn't matter whether you know how to optimise assembler routines
on a pipelined TTL mainframe but people suspect you might not be able to
readily do the same on an 8086; having mastered C (or whatever) that skill
is indeed portable to any computer with a decent C compiler.

All my working experience post dates assembler

I wrote high level code in a variety of languages

BCPL, RTL/2, Coral 66, PL/M and even some Fortran XX (was it 77? CBA to
check) (FTAOD each language does not represent a new employer).

All written using either a specialist (and relatively very expensive)
development systems based upon the same processor that the final software
was to run on or cross compiled on a larger machine, usually a VAX (I'll
refrain from calling that a mainframe, but wont complain if you do).

Now almost everyone that wants microprocessor based solutions uses C/C++,
cross compiled on a PC.

Of course, in between, some people used C compiled on Sparc workstations,
but I managed to miss that in my career.

tim