"mark" wrote
802.11x over long distances seems to be a pretty hot topic right now,
and I've been thinking of experimenting with it myself. I'm wondering
what constitutes "breaking" part 15 FCC rules which would put me under
part 97 rules

What you are suggesting will cost you about $500 for each end.

Power, antenna gain, and content. Under Part 15 there is no content
restrictions, and the total power gain is determined by whether you are
putting up a point to point, or a multi-user system.

I am aware that certain channels in 11b fall under the amateur radio
spectrum, however, it seems that as soon as you step into part 97, any
and all means and methods of encryption are suddenly illegal, no?

You probably want to stay away from encryption, unless the ARRL
publishes a password that everyone will use.

If this is the case, my goal is to stay in part 15 rules so I can
maintain a shred of security. My goal is roughly 9 miles, which is
just over 14km.

You don't need Part-97 for that. Just remember, no coax unless you
have an Amp. Put the radio in a weather box, and mount it to the
antenna. You will need a very good parabolic at each end, and if
ice is a problem, you need to use a radome.

There are two ways of increasing the distance of your link:
#1: boost your power (up to 1 watt) - this puts you inside part 97,
and therefore precludes the use of encryption.

If you increase the power, you ruin the reason for spread spectrum.
Like CB radio, everyone goes out and buys an amp, and then the
band is "broken."

#2: Use a high gain, directional antenna. This one I'm not sure about.
Would something like a cantenna or yagi of some sort also push me into
the amateur radio realm? If not, is it possible to get a connection at
that distance?

No. You will need more dB gain, and a parabolic dish or yagi is the
way to go. The rule is, you have to reduce the power as the antenna
gain increases, but most WiFi is low power. You would need to get
up around 500 mW before this would take effect.

11a attenuates more quickly than 11b. Therefore, I'll use 11b to
increase the chances of success. How badly do obstacles (such as
trees) affect the signal? (my guess is pretty badly.

Trees and microwaves don't work. 11a uses OFDM, and that's not
really spread spectrum. 11b uses a true Direct Sequence form of
spread spectrum (The older Symphony and RangeLan stuff used
frequency hopping), and DS provides a coding gain.

My main question, ultimately, is whether using a high gain,
directional antenna puts me under part 97.

High gain, no problem, until you get to 1 Watt Effective Power, You
won't get to 1 Watt unless you use an amp.