[CQ-Contest] The Skimmer Rule Challenge

[Dick Green WC1M]

My response to W4TV:

> No, the real reason the cluster spawned a new category was because
> those who used the cluster did not want to compete against tag-team
> operators.

I don't think so. It was obvious that using packet is fundamentally
different than not using packet. I think a lot of people feel that way about
Skimmer, too.

> If you want to base decisions on "how competition is effected"
> then you need to reexamine all categories.  If "how completion
> is effected" is your criteria high power/low power/QRP should
> be based on an EIRP (effective isotropic radiated power) and
> not just power.  If "effect" is the criteria, the vertical/dipole
> and amplifier station belongs in the "low power" category and the
> barefoot/yagi station belongs in the "high power" category.  As
> the categories currently stand, the only thing that matters is
> transmitter power and the involvement of other individuals.

Well, CQ WPX has the Tribanders/Single-Wire category. It's a crude attempt
to do what you've suggested.

It's true that the categories for most contests don't take antenna
differences into account. In part, I think this is because there are far too
many variations. A yagi isn't just a yagi. It has a certain pattern and EIRP
that depend on the design, height and terrain. It would be nearly impossible
to normalize the thousands of different antenna patterns in use by amateurs
around the world so they could compete on level playing ground, even within
a category like Tribanders/Single-Wire. Just look at the hoops the WARC
sponsors have had to jump through to even approximate equality (and ask the
participants how close they came.)

Besides, it's no accident that contest sponsors have settled on rules that
reward antenna builders. Antennas are a huge part of amateur radio. Nothing
gets your signal out and the other guy's signal in as well as a better
antenna. We encourage people to develop more effective antennas by rewarding
them with contest and DXCC awards. I believe this is in the best interests
of Amateur Radio. We also reward people who develop and use technical
innovations, and we have special categories for that (yes we do: let's not
forget that packet is a technology without which spots from other humans
couldn't get to you.)

As the rules for SO-Unassisted stand today, those who develop more powerful
and flexible antenna systems and those who develop better skills for
listening, running pileups, tuning, understanding propagations, etc. are the
ones who win. Most of those who compete in the Unassisted category have
signed on to that philosophy. 

That's not to say there's anything wrong with the so-called Assisted
categories. They simply reward a different mix of contesting tools. You
choose your category based on which tools you want to use.

> Skimmer is simply a repackaging of technologies that have been
> available for many years (SO#R and CW decoders) with a new
> interface.

Not really. We haven't had the capability to simultaneously monitor an
entire band. Advancements in computer speed and DSP processing have made
Skimmer possible. And it's going to get better.

> The top operators are where the are because they optimize all
> their resources - not because they copy CW better than the
> other entrants.

This is simply not true. Better ability to copy may not be the only reason
top operators win, but it's a crucial skill. Ask any top op.

>  They know which band to be on at a given
> time, they are more skilled at moving multipliers, they have
> better knowledge of short term openings, they know the habits
> of the regular but "rare" stations, etc.

Yes, they have these skills, too. I don't disagree that the same group of
top operators will win if Skimmer is allowed in the Unassisted category. But
if one Unassisted top op breaks the ice and uses Skimmer, then they'll all
have to use Skimmer, and that's not how they want to compete. Like I said
before, if they wanted to compete that way they'd operate Assisted.
 
> Skimmer technology is not going to change any of that - it will
> be just one more source of information for the operator.

And it's almost identical to the kind of information you get from packet. It
doesn't really matter that one source of this information is human and the
other is mechanical. It's the type of information you are receiving that
affects the competition. For a long time it's been accepted that the type of
information provided by packet is of sufficient value that a different
category is needed. Skimmer provides essentially the same information, and
in the not-too-distant future will be more effective at doing it than
packet. If it looks like packet, talks like packet and smells like packet,
why wouldn't we put it in the same category as packet? Because it's not
"assisted" by another human? That's just semantics. The information provided
is nearly the same: the callsigns and frequencies of stations calling CQ.

> I think the Luddites are overestimating the impact of this
> technology.  Skimmer is being given skimmer too much credit
> and too much value is being placed on copying CW by ear.

It's not just a matter of copying by ear. It's a matter of taking 20-60
minutes to tune through a band looking for new stations to work, versus
having a bandmap with all the calls in front of you, available for instant
point-and-shoot. Let's have a race to see who can work more new stations in
a fixed period of time.

> For the single radio operator skimmer is not likely to do much
> beyond what can be accomplished by effectively tuning the second
> receiver (in band) while running.

This isn't true. In SO1R, there's a big tradeoff between time spent running
and time spent doing S&P. This is why skilled SO2R ops have the edge. They
still have to make the tradeoff at times, but less often. An SO1R op with an
external SoftRock can have the bandmaps populated by Skimmer while running,
substantially reducing the time required for each S&P excursion. A pure SO1R
op (literally using only one receiver, no Softrock) can let Skimmer run
between CQs, or take a few minutes to let Skimmer monitor the band before
starting to point-and-shoot. Let it run again when the bandmap empties. This
may or may not be more efficient and time consuming than tuning, but my
sense is that it will be.

> Still, SO2R is a far bigger determinant of
> the outcome than skimmer will be.

Possibly for the highly-skilled SO2R operator, but not necessarily for the
majority of SO2R ops. Depending on the contest, about 5% of my QSOs are made
on the second radio. The top operators usually run around 10%. Not all of
these QSOs would be lost in SO1R, so the net gain is in the single digits
percentage-wise. In a contest where 500 mults is a top score, it's not hard
to imagine Skimmer adding 25-50 mults or more, which will make at least as
much difference as SO2R.

If you compare SO-Unassisted and SO-Assisted scores, what you often see is
that the top Assisted operators have very high mult numbers, but their QSO
totals tend to lag the top Unassisted ops. This may be because they're
spending too much time grabbing packet spots, but I think it's more likely
they can't run as fast or copy as well.

> The skimmer effect is similar to computer logging - if the
> operator continues to use a single radio, the reduction in
> time spent on logging is not going to significantly increase
> his score.

Hard to believe. There's little doubt in my mind that an SO1R operator using
computer logging with CW encoding will run rings around an SO1R op logging
by hand and sending with a key. Just look at how scores have increased since
the introduction of computer logging. Some of that has to do with increased
participation (computer logging is more fun), but a lot has to do with
greater efficiency.

> It is not the technology ... skilled operators without skimmer
> will continue to beat lesser skilled operators with skimmer just
> like there are single radio operators (e.g., K3ZO) who regularly
> beat lesser skilled SO2R operators.

But what if the competition is between two operators of roughly equivalent
skill level? My example was K5ZD and K1AR (apologies if one or both think
the skill level isn't equivalent :-) If one uses Skimmer and the other
doesn't, the game has changed. If one uses Skimmer, the other will be forced
to do it, too.
 
> When skimmer is given a fair analysis and in context of all the
> other technologies, this issue not one of "assistance" - it is
> nothing more than "not in my back yard."

Once again, the term "assistance" is obscuring the issue. Help from other
operators is certainly a differentiator, but it's not the only one worth
considering. 

Regarding "not in my backyard", category decisions should be influenced by
demand as well as how various options affect competition. I'm not hearing a
lot of wailing about Skimmer being allowed in the SO-Assisted (or whatever
we call it) and Multi categories. Could it be that Skimmer is so much like
packet that no one cares? But the polls currently suggest that the number of
people who want to keep Skimmer out of the Unassisted category far outweighs
the number of people who want to allow it. We have to take that into
account.

[Yes, some people are worried about robo-contesting if we allow Skimmer in
the SO-Unassisted and Multi categories, but I think that can be addressed in
the rules (and some maintain that, unattended, it isn't legal in the USA
anyway.)]

The question is, do we want to preserve the essence of SO-Unassisted
contesting as it stands today? If so, it would seem that Skimmer isn't
appropriate because it's like packet. We recognized long ago that packet
would change SO-Unassisted and created a separate category for it.

>  Skimmer, with all its
> limitations, is likely to have a far smaller impact than SO2R.
> It may provide the opportunity to get a small handful of extra
> multipliers but not necessarily more than could be found by
> improving the operator's "second receiver" utilization.

As someone who has been doing SO2R Unassisted for almost 10 years, and
getting decent results, I don't agree. I think Skimmer would substantially
improve my mult performance.

I use packet, by the way, when I compete on the KC1XX M/M team. There is
absolutely no doubt in my mind we'd get our clocked cleaned by K3LR and
W3LPL if we didn't use packet. I fully expect Skimmer to eventually be
adopted by the super-M/M stations and that it will provide a significant
advantage to those who implement it first. 

For example, Skimmer might be more effective than a human second op on each
band. As it stands now, all three super-M/M stations have interlocked second
radios on most of the bands. In some positions at KC1XX, the second op has a
separate antenna and can hear (more or less) to tune for more stations while
the primary op is CQing. At other positions, the alternate antenna isn't as
effective, and the second op ends up using the main transmit antenna to tune
the band between CQs. You can imagine how tedious and fatiguing that is. In
both cases, the second op has to use hand signals to let the primary op know
he wants to transmit. It's a major hassle and not always fun. Itmay actually
be more efficient for the primary op to let Skimmer fill the bandmap (it
won't get tired from the constant interruptions from the primary
transmitter), and take a moment to point and shoot when the rate slows down.
Even if it's less efficient, it'll be quite useful for contests where we
can't field a complete set of ops for all the bands.

I'm all for Skimmer being used this way, and expect few if any Multi
stations will complain about it. It's something of an arms race in the Multi
categories, and the attitude seems to be that any edge is worth pursuing,
regardless of its relationship to antenna building or skill building. Great.
That's why we have categories. 

I've proposed packet-less operation at Multis a few times, but no one has
ever taken me up on it. We certainly couldn't interest our top competitors.
Looks like the main priority in Multi contesting is to run up the biggest
score possible, using all the available tools. That's fine, and sometimes I
find it fun to do that. If I want the pure, packet-less experience, I can
operate SO-Unassisted. I'd like to keep that option open, and not be forced
to use Skimmer to stay competitive.

> To those who ask, "why should I have to compete with someone
> using skimmer?"  I ask, why should I have to compete with someone
> using SO2R or big antennas?

Because, as I said before, the contest sponsors have decided to reward those
who build bigger antennas and develop better skills. To me, that's an agenda
worth supporting. I can tell you from experience that the commitment in
time, money and effort required to do those two things is far, far greater
than the commitment required to buy a Softrock, download Skimmer and get
them working. We should recognize and reward those who make the larger
commitment.

> The operator who rejects SO2R, or big antenna, or skimmer does so with
> full knowledge that he has chosen to limit his options.

Sure. SO-Unlimited ops have long chosen not to use packet. It seems from the
numbers that they're now choosing not to use Skimmer. Could it be because
the effects of Skimmer and packet are basically the same?

73, Dick WC1M