[Moon] Do you use a callsign list? ( shortened version)

[Peter Blair]

"The paper lists make me a little uncomfortable, but as they seem to be in widespread use I guess it is ok."  
Russ,  I wonder if you can substantiate the above statement that lists are "in widespread use".  If you do have any evidence please produce it.  I dont know when it was that the last list appeared in the 432 and above NL, I suspect it was before JT came on the scene. I do remember Al making the decision. But there are other lists, on PA0PLY ‘s site for example.

Lets try and get some facts, we will of course depend on honesty here but will anyone who has NOT used a callsign list for assisting CW EME contacts in the past 10 years please email me OFF LIST just with a simple message " NO LIST" with the frequency band "2m and/or 70cm and/or 23cm" and your call  and we'll add them up. I promise that no callsigns will be revealed and I will delete all the replies at the end of the exercise. 
Remember, OFF LIST only to g3ltf at btinternet.com and no flames please, I am just genuinely trying to get some facts.
73 Peter G3LTF  

Apologies, I did not realise how long the tail on this email was so I am re-sending it in case it got stopped for being too big.

-----Original Message----- 
From: Russ K2TXB 
Sent: Saturday, August 23, 2014 4:56 PM 
To: 'Bernd DF2ZC' ; moon at moonbounce.info 
Subject: Re: [Moon] The Swedish 4Th 432 & UP EME meeting CW 

Hi Bernd, of course I agree. I was trying to simplify my argument but went a little too far I guess.  Here is the thing.  I have noticed many times that, On CW, when I think I know the call, then that is what I hear.  Assuming the signal is in the noise and we are really only able to reliably hear 40-50% of the dots and dashes, when we have decided what call it is then it is easy to say to oneself, ok I only heard the dah space dash, but it was long enough to be dah dit dah, so I KNOW that was a K.  Soon we have convinced ourselves that we are copying well and we send back a 539 report!  In  reality, if we had no idea what the call was we would not be able to send more than QRZ.  This is where the lists (both paper and in the mind) become so useful.  There is nothing we can do about lists in our minds.  They are there and we can not stop ourselves from using them.  The paper lists make me a little uncomfortable, but as they seem to be in widespread use I guess it is ok.  But I have often seen where a decision based on scanning a list was wrong.  And once one has decided that he 'knows' who it is, it is much harder to get to the point where he tries again to copy the call carefully.

As you say, the deep search algorithm does accurately decode the call based on the number of bits received in the correct bins over a period of time.  In this it is much more accurate than a human can ever be, but it still does make a mistake now and then.  When it is not sure it adds qualifying information to the decode line and it is up to the operator to do whatever is necessary to make sure the computer got it right.  Usually that means waiting for a subsequent decode to make sure it comes out the same.  This is simple and usually works well.

The people who are calling this cheating are really only upset because they feel it is an unfair advantage, but it is not a cheat and certainly much more accurate than copying by ear, with or without a paper list.

73, Russ K2TXB

-----Original Message-----
From: Bernd DF2ZC [mailto:df2zc1 at googlemail.com] 
Sent: Friday, August 22, 2014 4:46 AM
To: moon at moonbounce.info
Cc: Russ K2TXB
Subject: Re: [Moon] The Swedish 4Th 432 & UP EME meeting CW

Dear Russ,

I am aware I run the risk of warming up this soup again but there is one paragraph in your correct statements I can't leave without
comment:

>I disagree about the comparison of knowing or having a list of stations  
>versus WSJT deep search.  I am very familiar with how these lists are used.
> I have sat and watched operators of some big stations hear a partial 
>call  and grab the list for the band and look through it for the 3 
>letters they  got, until the find a call that fits.  Then they answer 
>the station sending  OOO.  That is exactly the same thing as deep 
>search, and in fact is often  less reliable in terms of figuring out the correct call.

This use of a list is not at all the same as the ds process in JT65.
While people using lists only copied for instance "K" and "TX" they then guess that it is K2TXB since that c/s is in their list but have no real basis for a guess. It could have been also an off-list callsign such as "W4KTX" for example.

As you know, JT65 however does a correlation of the signal form in the noise with calculated signal forms based on possible combinations with callsigns in call3.txt. It then displays the result of that correlation with a confidence value of the result. It is no guess as above but a mathemactial method to compare two signals. The operator is then left to decide whether he trusts the result of that correlation or not, based on the confidence value attached to that result.

Therefore the JT65 DS method is not to be compated to the theat guess into the blue some or many used in the pre WSJT era.

Now I'm looking forward to the flaming by the handful of CWisKINGS that of course what I wrote here is wrong, including the notorious personally insulting comments of someone in HB9. For the latter applies the rule "if you get personal it shows you have no point factwise".

73 Bernd DF2ZC




-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.moonbounce.info/pipermail/moon/attachments/20140823/04a6b5fb/attachment.html>