If you're looking at transceivers, particularly older ones, make sure you look for at least two things: IF filtering and stability.

Good, narrow IF filtering allows you to select only the signal you want.  The important difference between IF filtering and, say, audio filtering is that IF filtering will eliminate strong signals before they hit the sensing point in the AGC loop.  Without the IF filter, strong signals can capture the AGC, driving down the gain of the signal you're interested in.  (You know all those complaints you hear about hams who are allegedly using too much power on PSK31?   In most cases, their accusers are actually saying that they either don't have IF filtering or don't now how to use it.)

Stability is important for PSK31 and even more so for JT65A.  Most recent-vintage rigs will be fine, but if you're contemplating something older, make sure you confirm that other operators are successfully using those rigs on the modes you're interested in.

Sure, why not?  What happens if someone lets their license expire, then renews during the grace period?

Here are a couple of things to try:

1) Is RF getting into your computer?  Transmit into a dummy load; if the problem goes away, it's RF.  (If you don't have a dummy load, lower the tx power by turning down the tx audio going into your transceiver.)

2) Is your computer too slow to handle the job?  If you're just listening to signals, how long does it take for the software to decode received signals?  If the signals aren't decoded by 55-56 seconds after the start of the minute, your CPU is too slow.

Does the FCC ever issue a notice when one of these characters actually pays the fine?  I've seen notices about fines being assessed, or rulings being challenged, or the amount of the fine being challenged because the defendant pleads poverty, but I've yet to see any indication that anyone ever pays.

JT65A operates on a 1-minute cycle.  Stations start transmitting at 1 second after the top of the minute, then stop at 48 seconds.  This leaves a 13-second deadtime in which receivers can decode what was heard.  Then the cycle repeats.  If you're too far out of sync, you miss part of what the other stations are transmitting, or they miss part of what you're transmitting.

I've found that you can be off by as much as a couple of seconds, but it's a good idea to be closer than that.  If you're using JT-65HF, look at the times under the DT (delta time) column.  If you see some that are - (transmitted early) and some that are + (transmitted late), you're probably in good shape.  If all of the times you see are the same sign (and more than a few tenths of a second), it's time to check your clock.  The horizontal red lines drawn across the waterfall also serve as a quick and easy time sync check.

I've found that Dimension4 does a good job of keeping your PC clock accurate.

I'd vote for multiple panels, too.  Currently I'm using three panels for work and three for the amateur radio desk.  That's lots more pixels than I'd get on a single panel, yet all of them are more or less at the same distance from my eyes.  The only problem is that I occasionally have to hunt for the cursor.

By the way, if you're using a laptop and don't have enough video ports, plugable makes great USB-to-HDMI adapters.  I'm using their USB 3.0 model, and it's plenty fast.

You get some additional phone frequencies on 75, 20, and 15 meters:

http://www.arrl.org/files/file/Hambands_color.pdf

Here's a question: why do we usually hear the "you're transmitting too much power" complaint from PSK31 ops, and not from CW ops?  Both modes have modest bandwidths, and both modes are susceptible to the receiver's AGC reducing the gain of multiple signals in the presence of a strong one.  You'd think the CW folks would be fighting mad, too.

But they aren't, because they learned long ago that the way to receive a weak signal in the presence of strong ones is to reduce the receiver bandwidth to eliminate the strong signals before they can capture the AGC loop.  Typically, CW ops use fixed IF filters and passband tuning to knock out all but the signal of interest.  This works equally well for PSK31.

Blaming excess power for one's PSK31 reception problems is a losing battle.  The fact of the matter is, you have no way of knowing whether a signal is strong at your location because of high power, a high-gain antenna, or fortunate propagation.  And unless the person the strong-signal guy is talking to happens to be close to you, you have no idea whether his transmitter power is excessive, i.e., more than needed to sustain communication.

We PSK31 ops should stop complaining about someone else's allegedly high power, and start learning to use rx filtering.  If we want to blame anything on transmitters, it should be the wide signals produced by overdriving transmit audio.  I frequently see PSK31 signals so badly overdriven that they spew garbage over 2-3 kHz, and that garbage can't be filtered out if it's on top of the signal you're trying to receive.  

Memory and scientific calculators are permitted, as long as you can prove to the examiners that you've cleared it.

We always have a supply of calculators, including scientific models, that candidates can borrow for the exam.

You have a great web site!  You've done some very interesting path simulations for various digital modes but, more than that, you've explained how we could do them ourselves.  I'm a fan of Olivia (too slow in your estimation, I know, but to each his own), and am going to see if I can do similar simulations for that mode.

Unfortunately, I can't say, because I have only a couple of data points.  My Pentium 4 3GHz machine running XP takes perhaps 6 seconds to decode, whereas my 4-core, i7 laptop running W7 x64 takes about a second.  Running JT65A on the i7 laptop is easy, whereas running it on the P4 machine isn't a lot of fun.  If I were buying a machine today, I'd want at least an i5 processor, which is not to say that you couldn't get away with less.

I doubt that the number of cores directly affects the operation of JT65-HF itself.  However, most of us are running more than one program on the machine that runs our JT-65 application--I run about a half-dozen--and multiple cores are a big help for that.

One of the things that makes JT65A operation easier and more enjoyable is a processor that's fast enough to decode the past minute's signals in a second or two.  The faster the processing, the more time you have to decide what to do before the start of the next minute.  Make sure that the cheap netbook you have your eye on has enough oomph to decode signals quickly.

Some hams are going to disrespect codeless Extras no matter how well you argue to the contrary.  As a wise person once said, you can't reason someone out of something they didn't reason themselves into.

Here's something that's puzzled me since the advent of the Internet:  In real life, if we see someone mumbling nonsense on a street corner, we give them a wide berth.  On the Internet, if we find someone mumbling nonsense on a forum, we try to reason with them.  Why waste your time?

While I haven't used it, I've heard good things about HamTestOnline.

Another useful resource is KB6NU's free No-Nonsense General Class License Study Guide, which you can download here:

http://kb6nu.com/wp-content/uploads/2007/09/2011-No-Nonsense-General-Class-License-Study-Guide.pdf

I'm speaking only for myself, but, no, it would not.  I became a VE after the code tests were dropped, but I've heard from some of the VE old-timers that administering the code tests was a pain in the neck.  Having taken the 20WPM test at an early VE session back in '85, I can see why.

Beating a dead horse isn't all that much fun, either.

N2EY is right: if someone wants to set up a club for CW novices, issue certificates and awards, arrange skeds, hold contests, administer recreations of old exams, etc., no one is stopping them.  Sounds like a great idea to me.  And, as has been pointed out, SKCC is already doing some of that.