Monday, January 25, 2010


John Harper AE5X's blog isn't in my blogroll, partly due to the fact that he declined to publish my comments to a couple of his earlier posts. In fact I see that comments are off for several posts that I might have liked to comment on, suggesting that John isn't interested in anyone's opinions but his own.

In a recent post John has a dig at propagation indicator gadgets including my WebProp, claiming they give indications of conditions that are incorrect. John makes his point using a screenshot of the DX Cluster showing a lot of spots for 12m contacts, together with shots taken at the same time of WebProp and another gadget from Paul Herrman N0NBH claiming conditions on the higher HF bands were "poor".

In defence of WebProp I would say that the 24MHz contacts John saw on the DX Cluster were probably short skip sporadic-E or trans-equatorial propagation, two modes that to the best of my knowledge are completely unpredictable. WebProp and the other propagation gadgets base their estimates solely on the effect of D-layer and F-layer ionization which is affected in a fairly predictable way by the solar flux, A and K indices.

But upon further reflection I realized that John had made a good point. WebProp and co don't come with any warnings or disclaimers that they don't show the whole picture. It may be obvious to some that you can't predict the unpredictable, but there are a lot of people who believe that if a computer says conditions are good or poor then that's what they must be. A propagation indicator that can't reliably show what conditions are actually like is about as much use as a weather forecast that says it's raining when it is sunny (though in the UK we're already used to that.)

With hindsight it would have been better - or at least more accurate - to have made WebProp just present the solar indices received from WWV and NASA and leave their interpretation to the end user. However I can't just withdraw WebProp as a lot of people are now using it on their sites. Nor can I easily remove the suspect information as that would make the box shorter and leave an ugly gap beneath it.

What I have done is add an option "condx=no" that can be used to suppress the band condition indicators, as shown above. This will make the box shorter so anyone who decides to switch to using this version will also need to adjust the height they allowed for the box.

It would be better if the condition indications used the presence of DX Cluster spots for their assessment of band conditions. I have had the idea to do this for a couple of years, but it is easier said than done and it is quite possible that I will never get around to it.

In the meantime, the "condx=no" option should make the information WebProp provides acceptable to those who think propagation gadgets should stick to the facts. Perhaps AE5X might even like to include the new version on his blog site?
Post a Comment