There are some content regulations so transmitting "in the clear" for all to read is generally demanded by the .gov. Groundwave HF paths technologically don't

There are some content regulations so transmitting "in the clear" for all to read is generally demanded by the .gov. Groundwave HF paths technologically don't have much of a symbol rate limit. You could talk from ND to SD using a quite fast symbol rate as long as its a groundwave path under 25 miles or so. There are limits, but they're huge. Skywave "ionosphere bounce" paths have massive multipath issues although slow enough symbol rates can make it thru without too much intersymbol interference. These paths are worldwide. Even with existing limits, its quite trivial to transmit a modulation that has too fast of a symbol rate for skywave / international paths. I guess I'm just pointing out the existing rate limits are "too high" often for international communication. You could turn 10M into some kind of short range local wifi I suppose, but all anyone more than a hundred miles away would hear would be noise / interference due to intersymbol interference and multipath. Supposedly, in my grandpa's day, the government regulated modulation method to a precise detailed level so if you're only allowed to transmit 45 baud ITU2 encoded rtty FSK, specifying a dozen of one or 12 of another is the same thing, just depends how you say it. Then they removed detailed mode regulation "do what you want, grandfathering in existing users". Regulation can take awhile to change with the times. Again, supposedly in my grandpa's day, wattmeters were not accurate enough to be useful so the government regulated DC input power because voltmeters and ammeters are accurate enough for a ballpark guess. See also peak power vs "peak envelope power" aka PEP measurements for analog single sideband voice. So bringing it back around, at one point symbol rate was a reasonable proxy for old fashioned rtty fsk bandwidth regulation.