Re: [digitalradio] Busy detect screenshot for Winmor
Thanks Skip. primary mission is for regional access. So far that has turned out to be the case, stations within a few hundred miles. I'm also paying around with 10M ground wave to see what results are. 6M may also be tested. Andy K3UK On Mon, Jun 28, 2010 at 7:00 AM, KH6TY wrote: > > > Andy, the reason there are multiple stations is that every station on a > frequency is not on at the same time. This is why busy detectors can work as > a sharing mechanism. The busy detector make you wait until the traffic on > the frequency has been passed and you can use the frequency. The reason > there are multiple bands is for a similar reason, and also to accomodate > propagation. If you want to have your mailbox always reachable, you > obviously need to scan multiple frequencies and multiple bands. > > The whole point of busy detectors is SHARING frequencies on a > first-come-first-served basis. Otherwise, if there is always a clear > frequency, all the time, the busy detector is not needed. > > A Winlink station may start out at the highest speed level, but usually > cannot maintain throughput and has to drop down, which releases the top half > of the channel, because the bandwidth decreases at the same time. At least I > think that is the way it works. I stand corrected if it is not. Stations > using 500 Hz Winmor should use narrow IF filters so a Pactor-III station on > the same channel does not block your access. > > I have not counted the number of US and Canadian stations in Winlink > recently, but there used to be about 50 stateside, and the idea, and we use > this for MARS, is that if propagation is not favorable for a local server, > one farther away may be accessible. It does not matter, since the Internet > ties all of the servers together and you can retrieve your email from any > that you can access. > > My experience with Winlink was that I almost always had to connect with a > station in New England (from Charleston, SC!) because of propagation or more > locally PMBO's being busy (or not answering because they were busy on a > secondary frequency or band). > > I assume you are not trying to be a Winlink network with your single > station in New York state, so this eliminates the need to be accessed > internationally, or by yachts far offshore. You probably need to first > define what your station mission is going to be and start from there. > > 73, Skip KH6TY > > > >
Re: [digitalradio] Busy detect screenshot for Winmor
Skip et al, Settling on a sensible list of frequencies will take some studying FYI.. Here is a list of specific frequencies used by Winlink HF stations, May 2010 list. Many have multiple stations using the particular frequency. The list is world-wide. 3565 3569 3580 3583.5 3587.2 3589 3590 3591 3591.5 3592.5 3593 3593.5 3595 3595.9 3596 3598 3603 3604.5 3605 3608.5 3611.9 3613 3613.5 3615 3617.5 3620.2 3624.3 3627.7 3643 7035.4 7036.9 7037 7038.7 7040.9 7043 7043.5 7043.9 7046.7 7049 7050 7051 7051.4 7051.5 7052.5 7053 7063.9 7065.9 7066.9 7067.9 7068.3 7068.9 7069.5 7070.9 7071.9 7074.9 7075 7075.4 7076.9 7090.5 7091 7092 7094 7096.5 7098.5 7101.2 7101.7 7101.9 7103.5 7103.7 7104.4 7107 10110 10116.2 10118.5 10122.9 10127 10127.9 10133.9 10135.4 10136.9 10138 10139.5 10140 10141 10141.2 10142 10142.7 10143.4 10143.7 10144 10144.5 10145 10145.5 10145.9 10146.2 10146.5 10147.5 10147.7 10148.2 10148.5 14062 14064 14064.9 14065.9 14066.9 14068.9 14069.4 14074.9 14075.3 14075.9 14088.2 14089 14094.9 14095.9 14096.2 14097.5 14098.5 14098.7 14101.7 14102.4 14102.7 14103 14104.2 14105 14106 14106.7 14107.4 14108.5 14108.9 14109.2 14110 14110.4 14111 14111.9 14112 14112.4 14112.5 14113.5 14114 14115 14115.5 14117.9 14124 14127.5 18075.4 18097 18100.9 18101.9 18102.9 18106.2 18106.5 18106.7 18106.9 18107 18107.9 18108 18111 18113.8 18116.5 18124 18126.5 21074.9 21075.4 21091.2 21098 21098.7 2 21117.9 21122.5 21126.5 21183 21298.7 24939 28133 On Sun, Jun 27, 2010 at 9:16 PM, Dave AA6YQ wrote: > > > > +++ More AA6YQ comments below > > -Original Message- > From: digitalradio@yahoogroups.com [mailto:digitalra...@yahoogroups.com]on > Behalf Of KH6TY > Sent: Sunday, June 27, 2010 7:02 PM > To: digitalradio@yahoogroups.com > Subject: Re: [digitalradio] Busy detect screenshot for Winmor > > >>>Its my impression that the WinMOR busy frequency detector has been > well-characterized as effective (going back to its original deployment in > SCAMP), so its not clear to me why more evaluation is required.
RE: [digitalradio] Busy detect screenshot for Winmor
+++ More AA6YQ comments below -Original Message- From: digitalradio@yahoogroups.com [mailto:digitalra...@yahoogroups.com]on Behalf Of KH6TY Sent: Sunday, June 27, 2010 7:02 PM To: digitalradio@yahoogroups.com Subject: Re: [digitalradio] Busy detect screenshot for Winmor >>>Its my impression that the WinMOR busy frequency detector has been well-characterized as effective (going back to its original deployment in SCAMP), so its not clear to me why more evaluation is required. More evaluation is required simply because it has not been tested in general use, so it may have been "characterized" as effective, but a full-blown use (in the subbands) will confirm that is characterization is correct. It will also show if there are people disabling the busy detector for reasons they deem necessary or convenient. +++ "in the subbands" is by definition not full-blown use. The safest way to find that out is to use it in the automatic subbands. This way, if it needs improvement, or people are disabling it, the least amount of harm to the busy detector reputation will be incurred, and potentially many less people will be angered that might otherwise retaliate and intentionally block. +++ WinMOR servers have been operational for months; not a single report of a WinMOR busy frequency detector failure has appeared here. Contrast that with ROS. There is no incentive NOT to keep the Winmor busy detector active - yet. +++ Are you saying that there has been no intentional QRM of WinMOR servers? If so, does the WinMOR community agree with you? So, there is no need to demonstrate to the broader community that it is already "safe". PROVE that it is safe first, with a wider use (in the subbands), and if it is, then turn it loose into the wild. Just characterizing it as such on a limited beta test program with a few beta testers does not prove what will happen with wider usage. Keeping it in the unattended subbands can serve just as well as having Winmor mailboxes everywhere immediately, and if it turns out it truly is a good neighbor, then the use can be wider. I think Andy also feels it is too soon to operate his Winmor mailbox outside the unattended subbands. +++ As I've already said, individual operators should apply their good judgement; if they aren't yet confident in the busy frequency detector's effectiveness, then running their server within the "automatic subbands" is entirely appropriate. But when experience leads such operators to become confident, they should be free to venture out onto other frequencies to which they are by regulation entitled to use. 73, Dave, AA6YQ
RE: [digitalradio] Busy detect screenshot for Winmor
>>>AA6YQ comments below -Original Message- From: digitalradio@yahoogroups.com [mailto:digitalra...@yahoogroups.com]on Behalf Of KH6TY Sent: Sunday, June 27, 2010 2:07 PM To: digitalradio@yahoogroups.com Subject: Re: [digitalradio] Busy detect screenshot for Winmor Dave, I realize you have championed the idea of a busy detector for a long time, but unless it cannot be switched off, it will eventually be switched off, and those mailboxes will be spread over the bands, since they are allowed to go anywhere RTTY can. >>>I would be happy to see servers incorporate busy frequency detectors that cannot be disabled. However, adoption by server operators will require the elimination of intentional QRM. What is wrong with keeping narrow bandwidth servers with busy detectors operating at the high end of Winlink Pactor-III channels, since Pactor-III seldom reaches the highest speed level for very long and decreases bandwidth to suit the lower speeds? >>>There are two reasons to encourage servers with effective busy frequency detectors to utilize available frequencies: 1. it provides an incentive for server operators to incorporate busy frequency detectors 2. it demonstrates to the broader community that servers with busy frequency detectors are as polite as human operators, which should reduce the rate of intentional QRM >>>If a server operator is not yet confident in the effectiveness of the busy frequency detector included in his or her server, then using frequencies within the "automatic sub-bands" is good way to monitor the busy frequency detector's effectiveness and either gain the confidence that the detector works well enough to operate outside those sub-bands, or not. Your assumption is that Winmor servers and clients will always keep busy detect activated, but it has been shown that mailbox operators grow impatient to retrieve email, and if a channel is busy too often, will transmit anyway in an attempt to override the traffic already on the channel, even among servers of like kind. >>>As I've said, it would be best if busy frequency detectors were permanently enabled -- but there will likely need to be progress on all sides before this happens. Just getting an effective busy frequency detector into every WinLink PMBO would be a huge positive step. Why not try the busy detector/busy operators in a place designed for other automatic stations and see how well the whole system works. That is my suggestion. >>>Its my impression that the WinMOR busy frequency detector has been well-characterized as effective (going back to its original deployment in SCAMP), so its not clear to me why more evaluation is required. >>>The longer we keep digging our hole deeper, the longer it will take to escape. 73, Dave, AA6YQ
RE: [digitalradio] Busy detect screenshot for Winmor
At 12:52 PM 6/27/2010, you wrote: >This has generated enormous frustration over the years, to the point where >some operators now intentionally QRM such servers. This intentional QRM is as >disgusting as running a server without a busy frequency detector, and provides >a convenient excuse for server operators to continue avoiding or disabling >busy frequency detectors. So so true ! But not only severs. Many times I have come up on a clear freq for a keyboard to keyboard on time QSO just to be QRM'ed because it was pactor. Way to many have this thinking it's a MBO *just* because it's pactor. Wish I knew a way to help those with that thinking. John, W0JAB
Re: [digitalradio] Busy detect screenshot for Winmor
I can't disagree with your points Dave. However, until I have more experience with the busy detect from the server end, I think I'll feel more "comfortable" in the auto-sub band...at least for now.. Then , if I do transmit on a busy frequency despite the busy-detect, I'll feel less guilty. Despite Field Day, the band has not been busy enough for me to really test the busy detect at the server end. On the client end, the busy detect rarely fails to warn that the frequency is busy and halts a transmit until over-ridden. Andy K3UK On Sun, Jun 27, 2010 at 1:52 PM, Dave AA6YQ wrote: > > > I disagree. Being able to operate outside the "automatic sub-bands" is an > incentive for operators to preferentially choose servers that include an > effective automatic busy frequency detector and to keep that busy frequency > detector enabled. > > We're in a deep hole dug by those who ran (and continue to run) servers > (e.g. WinLink PMBOs) without busy frequency detectors. This has generated > enormous frustration over the years, to the point where some operators now > intentionally QRM such servers. This intentional QRM is as disgusting as > running a server without a busy frequency detector, and provides a > convenient excuse for server operators to continue avoiding or disabling > busy frequency detectors. > > The first step in escaping from a deep hole is to stop digging. In our > case, this means that > > 1. servers with effective busy frequency detectors enabled should be > welcome across the full range of frequencies available to them as specified > in the applicable regulations > > 2. the intentional QRM must stop > > 3. servers without busy frequency detectors (e.g. WinLink PMBOs) should > immediately be retrofitted with effective busy frequency detectors -- a > possibility that Rick KN6KB stated here a few months ago that he would > investigate > > 73, > >
RE: [digitalradio] Busy detect screenshot for Winmor
I disagree. Being able to operate outside the "automatic sub-bands" is an incentive for operators to preferentially choose servers that include an effective automatic busy frequency detector and to keep that busy frequency detector enabled. We're in a deep hole dug by those who ran (and continue to run) servers (e.g. WinLink PMBOs) without busy frequency detectors. This has generated enormous frustration over the years, to the point where some operators now intentionally QRM such servers. This intentional QRM is as disgusting as running a server without a busy frequency detector, and provides a convenient excuse for server operators to continue avoiding or disabling busy frequency detectors. The first step in escaping from a deep hole is to stop digging. In our case, this means that 1. servers with effective busy frequency detectors enabled should be welcome across the full range of frequencies available to them as specified in the applicable regulations 2. the intentional QRM must stop 3. servers without busy frequency detectors (e.g. WinLink PMBOs) should immediately be retrofitted with effective busy frequency detectors -- a possibility that Rick KN6KB stated here a few months ago that he would investigate 73, Dave, AA6YQ -Original Message- From: digitalradio@yahoogroups.com [mailto:digitalra...@yahoogroups.com]on Behalf Of KH6TY Sent: Sunday, June 27, 2010 9:25 AM To: digitalradio@yahoogroups.com Subject: Re: [digitalradio] Busy detect screenshot for Winmor Thanks, Andy. Unless it is not impossible to disable busy detect, to answer your previous question about where to operate with Winmor, I personally think that Winmor frequencies should ALL be kept within the automatic subbands, since the tendency is going to be to disable it due to the uncertainty if there is malicious blocking or not. This way, busy detect can still be useful in enabling frequency sharing with other Winmor stations, and if someone disables busy detect, the effect on the rest of the hams will not be significant. This brings to mind the edict by Winlink that busy detect must not be enabled because of others trying to harm Winlink. It is highly unlikely that any malicious blocking will be done in the automatic subbands, because there is no reason to do so. The only blocking will be if the frequency is already in use by another mailbox. The recently reported problem with a PSKmail server still interfering with JT65 points up to another reason that ALL mailbox stations need to be in the same area, regardless of bandwidth. The more narrow the bandwidth, the easier it is to find a clear frequency there, so there is still an advantage to using a more narrow bandwidth. The frustration of being blocked too often if operating in the general use areas is, sooner or later, going to result in operator deactivation of the busy detection, especially as more and more Winmor mailboxes are set up. Before things get to that point, I think that it would be wise for early adopters, such as yourself, to set a good example by operating Winmor only in the automatic subbands and using the busy detection feature to more efficiently share frequencies there. 73, Skip KH6TY On 6/27/2010 8:46 AM, Andy obrien wrote: Skip (and anyone else interested), see the attached screenshot showing the Winmor server busy detect Andy K3UK
Re: [digitalradio] Busy detect screenshot for Winmor
I agree Skip and have been studying the "unattended" sub-bands for suitable frequencies. Andy On Sun, Jun 27, 2010 at 9:24 AM, KH6TY wrote: > > > Thanks, Andy. > > Unless it is not impossible to disable busy detect, to answer your previous > question about where to operate with Winmor, I personally think that Winmor > frequencies should ALL be kept within the automatic subbands, since the > tendency is going to be to disable it due to the uncertainty if there is > malicious blocking or not. This way, busy detect can still be useful in > enabling frequency sharing with other Winmor stations, and if someone > disables busy detect, the effect on the rest of the hams will not be > significant. This brings to mind the edict by Winlink that busy detect must > not be enabled because of others trying to harm Winlink. It is highly > unlikely that any malicious blocking will be done in the automatic subbands, > because there is no reason to do so. The only blocking will be if the > frequency is already in use by another mailbox. > > The recently reported problem with a PSKmail server still interfering with > JT65 points up to another reason that ALL mailbox stations need to be in the > same area, regardless of bandwidth. The more narrow the bandwidth, the > easier it is to find a clear frequency there, so there is still an advantage > to using a more narrow bandwidth. > > The frustration of being blocked too often if operating in the general use > areas is, sooner or later, going to result in operator deactivation of the > busy detection, especially as more and more Winmor mailboxes are set up. > Before things get to that point, I think that it would be wise for early > adopters, such as yourself, to set a good example by operating Winmor only > in the automatic subbands and using the busy detection feature to more > efficiently share frequencies there. > > 73, Skip KH6TY > > > On 6/27/2010 8:46 AM, Andy obrien wrote: > > > > Skip (and anyone else interested), see the attached screenshot showing > the Winmor server busy detect > > Andy K3UK > > >