Software vs Hardware productivity (was: Re: Clause "with and use")

[Marin David Condic <nobody@noplace.com>]

Well, that's just a matter of "features" if you will. Building a larger 
"device" - whatever that device may be - is always going to be more time 
consuming than building a smaller one. Microprocessors have become 
bigger and that probably means they take more time to design than 
smaller ones. But if you say "Last month, there was a chip with 100_000 
transistors on it and this month the same team built one with 1_000_000 
transistors on it..." does that make the design team 10 times more 
"productive"? Not necessarily. Did they have to design each of those 
transistors individually, or did they just redesign *one* transistor to 
make it smaller, then wire up ten times as many of them into a bigger 
device?

To be fair, a parallel might be that you write a subroutine that is 
called in a loop 100_000 times and this is as much work as it can do in 
the alloted time span. Then you do some redesign of the subroutine to 
chop out some instructions and suddenly, you can execute it 1_000_000 
times in the same time span. Or your program had an array that held 
100_000 elements and, because you got a bigger box, you tweaked a 
constant and in a cocaine heartbeat you got 1_000_000 elements into the 
array. Were you suddenly 10x more productive than the first time? 
Probably not.

There just is no fair comparison of software design to hardware design 
if you're looking at the amount of features on a chip versus the amount 
of features in a software system. The chip designer can *design* one 
feature and duplicate it a million times - creating a million features. 
The software designer has to design each individual feature - otherwise 
its just a generic or a subroutine - which isn't counted the same way as 
a  NAND gate or a barrel shifter or an adder or a memory cell is in the 
silicon world.

Actually, I suspect that if software designers could take credit for 
"features" in the same way that hardware designers do, we'd end up 
showing similar, if not way more productivity gains. When I pop a window 
up on a screen, can I take credit for the number of pixels I've just 
disturbed? That's no different than a memory chip designer taking credit 
for the number of memory cells he created in his design. Programs I 
build years ago used to occupy less than 64k of memory and because of 
improvements in technology, etc., I can now build a program that exceeds 
a megabyte without hardly breaking a sweat. Isn't that more "productive" 
in the same sort of units a hardware designer might get measured?


MDC


Stephane Richard wrote:
> 
> I hear ya.  If you take that to the programming world,  the easier you want
> an application to be for it's user the more design (and cosequently
> programming) it needs.  Plain and simple. :-)
> 


-- 
======================================================================
Marin David Condic
I work for: http://www.belcan.com/
My project is: http://www.jsf.mil/NSFrames.htm

Send Replies To: m   o   d   c @ a   m   o   g
                    c   n   i       c   .   r

     "So if I understand 'The Matrix Reloaded' correctly, the Matrix is
     basically a Microsoft operating system - it runs for a while and
     then crashes and reboots. By design, no less. Neo is just a
     memory leak that's too hard to fix, so they left him in... The
     users don't complain because they're packed in slush and kept
     sedated"

         --  Marin D. Condic
======================================================================