The Stubborn Engineer
by Paul McGoldrick
One of our cats died last week. He had been deteriorating for a while but
he suddenly started breathing through his mouth and the vet came to the
house and took him back to the clinic to put into an oxygen tent; he didn't
make it through the night. This week two sheer black kittens, already house-trained,
came into our lives. It is fascinating how easy it is to forget what the
thunder of tiny paws is like and how all young animals bounce their way
through the first year of life. They kind of teach our other adult cat what
fun is all about -- and that is something that sometimes is missing in our
industry, too, as we fail to learn from the bounciness of youth
Most digital engineers should now stop reading this Editorial...because it does not apply to you. It applies to us analog engineers who must be some of the most stubborn people on the planet. Unless we are forced into it by budgets, direct orders or the threat of a pink slip, we get really difficult about bringing the latest techniques and systems into our lives. I find it astonishing how many designs are still being made with components like 741s, 555s, and 2N3904s, components that I have seen in museums! We are also very slow to accept new construction techniques, except in very high volume productions where the consumer price dictates that we must.
Mentoring programs at the semiconductor vendors, with a few exceptions, work poorly. The older, more experienced designers tend to come across as arrogant and inflexible, carrying the torch that they can stop the younger designers from going along the same dead-end paths that they went along in the decades before. But the recently-qualified designer may come with other ideas that should be listened to, ones that don't deserve to be squashed by the design manager or the company Fellow who doesn't want to either accept or encourage change.
We are beginning to see genuine improvements in design techniques for analog circuits, systems that allow the average designer the ability to much more closely achieve the results that are wanted in the first pass. Companies like Analog Design Automation and Anadigm are showing that you can size things properly from your starting point (ADA) and that you can package designs using preset design configurations (Anadigm.) When these techniques are presented to digital engineers thrown into the place I call the "analog pit," they embrace them readily, gratefully. When they are presented to analog design managers, they seem to shrug a lot. When the techniques are brought to the table by junior design engineers, they are scoffed at.
Take a lesson from kittens. They have never been there, they have never
done that. They will bash into things, they will taste things they should
never have anywhere near their mouths, they will chase anything that moves.
Do you suppress that joy of life? No, you can't and you shouldn't. Take
a look at what that fresh designer is saying and trying to do. Do you suppress
that? Too often that is the case but you should instead keep an open mind
and listen and watch and maybe learn. The young may not present solutions
that are universal, but they have been closer to the "here and now"
than we have been, and they really are worth playing with. So, get that
ball of yarn out right now…