Comments from a member re signal bandwidth proposal: I have a few concerns about issues likely to come before the ARRL board that I'd like to present here. I was gratified to see many of my suggestions incorporated in the description of the revised proposal for regulation by bandwidth instead of by mode. However, the proposal described on the ARRL website still contains serious internal inconsistencies that detract from it's credibility. These inconsistencies also preclude some likely future technical developments. Unless and until these inconsistencies are satisfactorily resolved, the proposal should not be adopted. In July 2002, the ARRL Board of Directors directed that (Minute 64): "At the next practical opportunity the ARRL shall petition the FCC to revise Part 97 to regulate subbands by signal bandwidth instead of by mode." My concerns are that making a special dispensation for AM/ISB mode is inconsistent with this direction and not including wider bandwidth segments below the 10 meter band will preclude future technical developments. Petitioning the FCC with this inconsistency detracts from the credibility of the proposal. An approach that is consistent with the proposal would be to assign spectrum in additional HF bands for signals of 9 KHz or less bandwidth. However this is not the best approach to take in this case. The current proposal has no bandwidths greater than 3 KHz allocated below 10 meters. Given that one of the goals of this proposal is to preempt possible requests for future FCC action, it is inconsistent to preempt the future development of wider bandwidth modulation schemes, presumably digital. The advantages of wider bandwidth on HF include, for comparable modulation constellations, faster data rates. Given the apparent intent of the ARRL to provide the capability for emergency bridging of discontinuities in the Internet via WinLink2000 and the like, precluding the use of wider bandwidth signals is inconsistent. It is also inconsistent with the goals set in 97.1 of the commissions rules. In a similar context, HF modulation constellations of wider bandwidth can be more robust during periods of poor propagation than a narrower signal for the same effective data rate. While the modulation schemes are still to be developed, greater immunity to noise, multipath and interfering signals can be obtained at the cost of greater bandwidth. Presumably an adaptive scheme will be able to trade data rate for noise immunity over a continuum, depending on conditions. However this option will not be available if no accommodation for wider bandwidth signals is made in the HF bands, as is the case in the current proposal. In order to provide for these potential developments and also to provide for existing AM/DSB activity in a manner that is consistent with the remainder of the proposal, it is imperative that there be allocations for wider bandwidth signals on the HF bands below 10 meters. It seems wise to make these allocations as general as possible. I would suggest that 20 KHz of spectrum be allocated in the 160m, 80m, 40m and 20m bands for signals whose transmitted bandwidth shall not exceed 20 KHz. In other words, the widest allowed signal will just fill the subband. It seems prudent to select these subbands so that they include the frequencies currently used for AM/DSB "by gentlemens agreement". This will allow for three simultaneous AM/DSB QSOs (assuming that AM/DSB requires 6 KHz of bandwidth) and allow the future development of wider bandwidth digital HF signals. In summary, the revised proposal for regulation by bandwidth instead of by mode is internally inconsistent, inconsistent with the boards direction for its creation and less supportive of the goals in 97.1 of the FCC rules than it might be. These issues must be satisfactorily resolved before the proposal can be adopted by the board.