Heinrich Rudolf Hertz (22 February 1857 – 1 January 1894) was a German physicist who clarified and expanded the electromagnetic theory of light that had been put forth by Maxwell. He was the first to conclusively prove the existence of electromagnetic waves by engineering instruments to transmit and receive radio pulses using experimental procedures that ruled out all other known wireless phenomena.

You could say he is one of the godfathers of modern radio. For more information check out the Wikipedia article here: http://goo.gl/bwcjm

OK I know, its starting to sound a little like a stuck robot but hey its what I am working on right now, more breakthroughs on the robot, its now doing a bunch of checking on the submitted log, works out the total time frame that the log contains (this could be a very handy step later on for scoring in a tie break situation, most calls logged in the smallest time frame could take a decider etc). Also checks your callsign against the one you want to submit the score for, so that should help for miss-logs.

One thing I do want to do for the sake of clarity is to save every log file uploaded, so that if there is any issue later with a score not matching up etc I will be able to see what exactly happened, I should also track the IP address the upload came from too I guess, I am sure I will find a way to work that in there somewhere.

So in short the basics are coming along well, there is plenty to do, sadly so far the process doesn't do anything exiting, just allows you to upload a log, checks the content and thats about it - from this point on it should get more exciting!

Well after some mucking about, the robot is now able to take the uploaded file, do some checks to make sure its the right file of the right kind, store it according to contest and callsign, make sure it really is cabrillo format and then output some of the data from the file to prove that it arrived intact and also give you a QSO count too.

Well its not time to party yet - there is still LOADS to do, firstly I need an SQL database to load all of the logs into, write the code to handle pushing the data I want into SQL in the first place and then there is a load of work to do creating a leader board based on claimed score and then the actual scoring. Plenty to do then...

So the first part of the robot system works, it will collect logs and store them by contest and by call sign - great, but that's only part of the first part, I also need it to read the logs - confirm some of the data from the cabrillo header and then read that back to the person who uploaded their log for confirmation - interesting - shouldn't be too hard, with luck I will have some time to look at that soon - those confirmations have to happen before the log is committed to the database.

Not that I want to run before I can walk but once the logs are collected, it would be possible to do more data processing on them afterwards - QSO/hour rates as a chart maybe, even a single chart with all the logs on it so you can see how you compared - so many options - I should really get step one done first...

Have you ever wondered how those log robots work - yes me too, I have some ideas forming in my head about some local contests I'd like to setup, but I don't fancy the paperwork afterwards to work out who won, by how much etc and so on. What would really help is a log robot - but how do they do it, are there any available online? Well how they do it could be reasonably simple, are there any available online - doesn't look like it.

So there are two parts to the problem, log submission, the robot must collect (and store) the logs. It should also do some basic data collection to make sure the log is complete and post a winners table based on claimed score. This part isn't too hard (I don't think).

Now the tough part, log analysis, this cannot start until all the logs are in, so there must be a deadline, the robot needs to check all the QSO's against both stations where both logs are available (so you also need check log submission) - UBN flagging for no match QSO's must also be done - the logic for that lot sounds like its going to be all kinds of fun.

The two main challenges here are going to be getting the data ordered so that processing is simple, then getting the processing done and forming some kind of output. Its a good think I like a challenge this looks like it will certainly be an interesting ride.