From: Joel Kolstad on Wed, Feb 22 2006 7:54 pm

Hi Len,

You have a lot of interesting history in there!

I feel fortunate to have experienced (first-hand) the
electronics technology revolution of the last 60 years.
Truly remarkable times!


I think it's ironic how the cool fashion accessory today is something like a
Motorola Razr phone with its associated millions of transistors buried in
numerous ICs running software that even relatively few BSEE's would fully
understand without a fair amount of additional study (turbo codes, direct
sequence spread spectrum systems, psychoacoustic codecs, etc. not usually
being a large part of the undergraduate curriculum...), yet 30 years ago
anyway hauling around a brick-sized ham radio or CB was quite the nerdy thing
to do... Incredible how times change...

Don't forget that MARKETING has been responsible for setting
"the public's" definitions of "cool" things. Lots of the
products on today's market are the work of specialists together
in one product family. It is difficult for anyone to be a
guru in more than one discipline within electronics (and radio).

Take the general MPEG4 compression that enables HDTV to work
within a 6 MHz bandwidth (and enabling 4-channel sound and
assorted whatnot to coexist there). Having been slightly
involved in some test equipment design for same, the concept
is mind-boggling to me. "Tylenol time" to attempt under-
standing it. I'd rather sit back and watch, enjoying the
MUCH better picture of HDTV as compared to the old 4:3 aspect
ratio analog TV. As far as I'm concerned, the compression
techniques are complicated enough that they don't seem to need
any IP protection...


If they want to TEACH students what goes on in industry of the
day, absolutely. They've gone insular amongst themselves in
the last few decades...in the teaching part of their activity.

IMO one of the biggest problems universities face is that everyone is
"expected" to get a 4 year degree these days, and yet the reality of the
marketplace is that relatively few jobs truly require anything approach that
level of "hard core" education. Hence, engineering courses get watered down,
and a lot of BSEE of BSCS students end up performing straightforward
programming or digital design using techniques that a 2 year technical college
could have easily provided them with. Industry has often contributed to this
problem, requiring even technical sales people to now have those 4 year
degrees... sheesh!

I would "blame" personnel departments for the "degree required."
Hey, it's hard enough for them (now called "Human Resources") to
try to sort out applications in all the myriad specialties typical
of an electronics house in the industry now. Over 30 years ago
RCA Corporation used to give two-level interviews to applicants;
initial screening in personnel, final screening by engineering
department folks (not necessarily liking the task). That was a
better, quicker way to separate the good from the bad and ugly.
I got hired by them that way and later wound up as one of those
who did the second-level final screening. It works.

In practical matters, academia just doesn't have the familiarity
with "industry practice" so it can't get a good handle on what
they really need to stuff in a curricula. That's academia's
problem..."industry" is going to go ahead and do its thing
anyway, regardless of title-rank-status blinged into minds by
academicians. "Industry" exists to stay in existance, make
products, make profits. "Industry" ALSO winds up as the general
inventor-innovator-creator of new technology...they just don't
allot a lot of time to write up much of it a la academicians,
only for marketing purposes to get folks buying their new stuff.


I think we're just about there -- I've seen chipsets that'll provide, e.g.,
some tens of bytes of data once every second or so and consume mere tens of
microwatts (on average) to operate; that seems like the kind of thing some
clever person can generate just from the rotation of the tire itself using
some horribly crude & dirt cheap implementation of a "generator."

Well, I'd point out the simple little "fob" transmitter for
the "keyless entry" system. A single IC in there that both
generates the "rolling code" digital sequence AND the tiny
UHF transmitter. The code sequence is flexible enough, simple
enough to set up for tens of thousands of different
combinations, low-power enough to last for years without a
battery change, can work in sub-zero and desert temperatures.
And be relatively cheap to make as well as small enough to
carry in a trouser pocket. One heckuva set of specs there!

The "keyless entry" receiver is as complicated, yet is only a
small part of the overall electronics in a modern auto. My
wife's (she picked it out, I okayed it) new Malibu hatchback
has ten kinds of bells and whistles on its dash display, a
relative drop in the bucket compared to its normal housekeeping
computerized chores of checking things, adjusting carburation,
alert warnings, light control, etc., etc., etc. It's made
possible by the ubiquitous microcomputer (now just thirty-
something old). Wonderful stuff in my view! Takes some of
the worry about cross-country driving as well as everyday
driving.


Well, in my view, there's too much ham emphasis on transmitters
and power and mechanical aspect of things.

What would you like to see more of?

Receivers, subjects that don't emphasize the new buzzwords
of the industry. Oscillators, some practical skinny that
doesn't mirror the current industry need for "low phase
noise" (applicable to FM and PM yes, but not many DSSS
high-speed data communications going on in amateur radio).
Practical simple test equipment for Rx and Tx that doesn't
need some specialty component part or unobtainium. Most
authors use what they had (leftovers from work or other
projects) and don't mention what could be substituted.

Some better theory and practical uses of complex quantities
that involve impedance and admittance...especially how to
measure them good enough for a home workshop environment.
A vector impedance meter and vector analysis system is just
now (in the last few years) appearing in the ham scene.
Some practical stuff on making inductors, Q affected by
what, how to measure approximate Q with workshop gear.

Some APPLICABLE (not product-specific promotional-based)
information on making things in the home workshop, whether
it is "manhattan-style" PCBs or even better soldering
techniques. [RoHS is with us, like it or not, an tin
solder without the lead doesn't handle as the old solder
did] How to work with materials properly, metal to
fiberglass-epoxy sheet or whatever. Even something on
standard screws, fasteners, things that can hold something
together. Avoid the TV craft-show techniques of most of
the show content concerned with promoting some latest
maker's new stuff...while it is momentarily interesting
it is also obviously a part of marketing pushing this new
craft stuff. ["product reviews" are a gross example of
this kind of PR]

A "tuna-tin two" two-transistor Tx is cute, but crammed
into an Altoids box? QRP is kind of a specialty niche
and probably deserves a special issue of some magazine
for that, but the lure of "simplification" (in size and
complexity) for simplicity's sake doesn't TEACH much of
anything except serve as a real test of dexterity on the
part of the builder. "Simplicity" in ham radio was very
popular from the 30s to the 50s because ready-built was
expen$ive then. Simple things could be built in a few
weekends, provided some enjoyment, but not a great deal
of real learning (other than assembly) into theory. Today
it is possible to build a "simple" single-conversion HF
superhet receiver with zilch problems of image response
yet have fine selective IF response for AM voice. Takes
all of six ICs on a PCB size less than 6" x 8" in size.
Frequency control in tuning can be simple or complex as
desired.



I did mean "true" SDRs, such as GNU Radio, Flex Radio, etc. I think it has
plenty of application to the HF world (indeed, the development of digital
modes for HF seems much more active than on VHF/UHF, which has always struck
me as kinda bizarre given how much less bandwidth is available there in the
first place... but of course the fact that you can get a signal to the other
side of the planet on 100W in good condition is always a big motivator...)

IF - and only IF - the ionosphere is kind to you... :-)

In digitized "data" modes for amateur use, there's only
ONE new, innovated-for-ham-use mode: PSK31. All other
modes involving "data" are adaptations of commercial-use
modes. One mountain-states small company had their first
product as a "beginner's HF PSK31 transceiver." It didn't
sell despite having some neat, modern microcomputer
and display screen, keyboard included. It wasn't
"traditional ham" gear, wasn't pushed by a big company.
Peter Martinez (G3PLX) had his PSK31 up, running, and
being tested by others using it in Europe for a few years
without any real details of it appearing in USA ham
magazines.

There just isn't much of anything else in "data" comms
standards here in the USA unless it is already developed
and debugged by "the industry." Digitized voice IS
possible in narrowbandwidths on HF but there don't seem
to be any amateur experimenters trying it out. What would
SDR do without any standards to adapt to?

One can PROCESS an IF today and do things like make a
digital spectrum analyzer (I don't see much effort spent
on that). That isn't really a Software DEFUNED radio
system. Firmware-software already CONTROLS most of the
functions in modern ready-built amateur radios. That isn't
really Software DEFINED radio either. OK, the quibble is
a minor thing. Software/firmware/digital-control is
definitely on the scene today.


One problem with digital voice is that there isn't
even a hint of a standard protocol or of many experimenters
yielding any results on same.

Have I mentioned how some of the best low-bit rate CODECs are proprietary
and/or patented and not licensable by a single lowly hobbyist? ;-)

Yes, you did. :-) I also mentioned that there's not much
of a hint of anyone in the USA experimenting on their own to
make a trial or two at their prospective standard. Someone
must BEGIN, be the flag-bearer, the amateur pioneer. What is
bizarre (to me) is that a modern PC has much more computing
power than big mainframes I connected to 30 years ago to do
ordinary synthesis-analysis of circuitry. The tools for
experimentation on coders-decoders EXIST NOW.

Perhaps the biggest bottleneck to all this possible pioneering
is the current mindset of radio amateurs. If it doesn't "look"
like "radio" (of their time), it "isn't real radio." :-(

The editors of the few remaining ham publications have the
paper-version control over what everyone sees for new
things. THEY are the ones needing to look into the future
and do gauging on what THEY think is "good" for readers.
Meanwhile, the Internet is busy, busy showing lots and lots
of new things that ARE done, can be done, to anyone who
bothers to search. [there's quite a bit of material on
CODECs and general DSP out there once you get the right key
search words] Most of what we see on the Internet is not
locked up in truly enforceable copyrights.