Sunday, August 18, 2019

Automated FT8 "FoxBot" in upcoming DXpedition - confirmed

Mk1 FoxBot, 2019
UPDATE: the SV5DKL logo has now been removed and the DX team will be in "full compliance" with the mode. Moral of the story: don't advertise your intentions to use a bot program!

.

Click here for WSJT-Z "bot" (de SQ9FVE) description.

.

A DXpedition to Tokelau will take place from 1 to 11 October and it may be will be won't be your chance to work an FT8 robot operating in Fox/Hound mode.

Stathis SV5DKL has been working on a "FoxBot" for some time now, is listed as a partner to this DXpedition and has confirmed that the DXpedition will be using his FoxBot.

The ARRL has recently announced that, effective immediately, regarding automated contacts:

WHAT is NOT allowed:
Operator sets up fully automated QSOs that automatically answer and CQ again without human intervention. This is possible with external software to control WSJT-X, modified versions of WSJT-X and third party programs that use the WSJT decoding algorithms but offer fully automatic operation.


In effect, contacts made with robots will not count for DXCC credit.

But in the future DXpeditions will use FT8 robots. As they said in the movie Titanic - "it's a mathematical certainty". Efficiency of DXpedition manpower and the demand of getting a new one will combine to move this practice into the New Normal.

DXpeditions will be of three types: those that use FT8 bots openly, those that do so on the sly and those that don't (yet) use automation at all. The latter group will come around after the first two groups normalize the practice.

And, for the second group - how will the ARRL know?

When Joe K1JT invented FT8, he no doubt had an idea of what it would accomplish in allowing stations with compromised antennas to work DX. But now the Law of Unintended Consequences has come knocking at the door.


Interesting times ahead.
.
.

13 comments:

  1. It's at the sole discretion of the Team to use the FoxBot, or not. My suggestion was "Have the SSB/CW operator of the nearby op. position have a look at it, every now and then".

    ReplyDelete
    Replies
    1. It is a technological hobby and FT8 is a mode that begs for automation - it's 90% there already anyway. To expect a worldwide group of people engaged in such a hobby, to not develop the mode further, is unrealistic. We'd have to be Luddites to not push the mode to its next logical step.

      Delete
    2. Agree! And the name is either Stathis or Efstathios! 73

      Delete
    3. Corrected, I'm blaming my word-bot ;-)

      Delete
  2. ARRL DX CW 1991 200+ QSO done by robot. It was agreed with K6STI NOT to launch commercial product in order to preserve ancient hobby. K1JT changed the scene. Hams are conservative preventing HF license without CW, DX Cluster, internet while claiming the hobby technically best for 21st century. Costly DX-peditions have more sense.

    ReplyDelete
    Replies
    1. Interesting, I was underwater most of that year and had not heard of this.

      Delete
  3. If you can't tell what (or who) you worked, it doesn't matter. My bot FT8 exchanges have exactly the same content as my "human" exchanges.

    Note to future DXpeditions: Go ahead and run an FT8 bot with an operator nearby to glance at it from time to time - just don't plaster that you'll be doing it this way on DX forums. No one will be the wiser and - here's the Big News - it won't matter.

    ReplyDelete
  4. Actually, I think I was able to detect folks using the older, buggier version of someone's 'auto' FT8 code, when I was running the special event station N8M at the recent Detroit Maker Faire. I had a couple of stations that kept trying to contact me again right after we had finished a good QSO. The operator can't be that stupid or have such a short attention span and it certainly wasn't a contest. I chalk that up to buggy software, which to a trained operator can stick out like a sore thumb.

    The point is, don't cheat. If the rules state (and they do now) to not use auto-mode for DXCC, then don't. It's not worth the scandal of cheating and getting caught, if that's the only thing you care about. Other folks won't cheat because that's simply the right thing to do, not because they are afraid they'll get caught.

    Set your moral compass accordingly.

    ReplyDelete
  5. Here's another thought:

    The sole job of an FT8 operator on a DXpedition will be to click on callsigns to move them to the "To be worked" window on WSJT-X. That's it - no skill involved. The kid the DXpedition paid to climb the palm tree to hang the dipole could stand in as the team's FT8 op with a 5 minute crash course in how to click a mouse.

    The DXpedition's FT8 op will be consigned to a role of look, click. Stimulus, response. Plankton could almost do it. Furthermore, an FT8 op will be no more skilled in the mode after 2 years of using the mode than he was the week after he started.

    To disallow automatic FT8 operation for DXpeditions is to consign a human to hours of drudgery. The mode - and common sense and the concept of efficiency - begs for attended automation. It was made for it.

    ReplyDelete
  6. Hello dear OMs !

    Today, I released the Beta-1.4 version of the FT8-Helper and uploaded to my homepage www.edtk.de

    Version History:

    Version 1.4 released on 21.08.2019
    Changes:
    - FT-4 is supported with WSJT-X 2.1
    - When started, the RUN and S/P boxes are green coloured
    - Accepts calls from stations in S/P mode
    - In S/P mode, the strongest station is called if more than one station with the same priority colour is calling CQ.

    Bugfixes:
    - Accepts low case characters in callsign.txt
    - Problems with RRR procedures are fixed
    - Stations sending frames out of order should not confuse the FT8-Helper
    - Timing error with fast computers and "Fast decoding" fixed
    - Size of Log-session is set automatically (no need to pre-set it)
    - Blocking of WSJT-X by modal (blocking) dialogs does not lead to unannounced crash

    Version 1.3d
    Bugfixes:
    FT8-Helper v1.3.d works now with both WSJT-X v2.0.1 and v2.1.0 (also 2.0.0 but not tested)
    Problems with short callsign (e.g. A3B) and stations calling "CQ TEST" fixed. Side effect: callsigns like "RAEM" are not accepted.
    Various crashes caused by blocked or nor reachable windows are catched and indicated to the user.

    Version 1.3c
    Problems were solved when using low resolution monitors. In S/P mode no Wide-Graph window required. The minimum monitor resolution is 1024 * 768 pixels. The text size for windows applications must be set to 100%. Read the short guide for hints using the FT8-Helper with low resolution monitors.





    Best 73&DX

    Geza

    DG5LP

    ReplyDelete
  7. I "gone" for a while and missed this post. I read all the "automated QSO" posts on blogs today and yesterday. Thanks for this last "robot" link. Some people will be happy, others will no be happy with this. I think it is interesting to experiment with. Automation can not be stopped in the end, it will happen. We knew that years ago. Just a matter of time... Welcome to the future! 73, Bas

    ReplyDelete
    Replies
    1. Welcome back form OZ-land Bas,

      SQ9FVE is updating WSJT-Z faster than I can write about them, fixing bugs and adding features. The best aspect is not the automatic functionality but the filtering and (now) the alerts. If you're an FT8 op who doesn't like the "bot" aspect, you may very well appreciate the other features in Tom's WSJT-Z.

      Delete
    2. Wow, this certainly will shake something! Will others (JTDX?) follow? 73, Bas

      Delete