The General Approach to Computing - A Ramble

[Tony Duell]

> 
> On Thu, Apr 23, 2009 at 1:04 PM, Warren Wolfe <lists at databasics.us> wrote:
> > That many
> > orders of magnitude of complexity have come and gone since the IMSAI days
> > are good, in terms of what one's money will buy today, but bad in that one
> > is much less likely to be able to fix one's own equipment to component
> > level.  It's not as satisfying to say "Hmmm.  You need a new video card" as
> > to hold up the offending transistor and cackle.   Or is that just me?
> 
> Maybe I'm seeing this from a different angle... I don't see a lot of
> difference besides the complexity of the devices.  A transistor is a

The difference to me is that I know I can't repair a transistor. I 
suspect I could make a transistor, although it would be a very poor one 
(The book 'Instruments of Amplification' covers making point-contact and 
copper oxide junction transsitors at home). But I couldn't repair one. I 
can't open the package without damaging it, so the package wouldn't e any 
use once I'd repaired the insides. And the failure is very likely to have 
damaged hte semiconductor 'pellet' inside, so I can't repair that. 

But I can repair a video card. I know I can take it apart into separate 
componetnts and put them back again without damage. I know that I can do 
electrical tests to find out just which of those compoentns has failed.

What I don't like is not having enough information to be able to carry 
out those tests, or worse still, once I've found the problem not being 
able to get the part.


-tony