Common problems in log submission

I just sent in my log for last weekend’s IOTA contest. I only made 38 contacts, mostly for the Prefix Hunt. Here’s something you won’t hear me say about any other contest: I don’t like IOTA. Anyway to find the address, I was browsing the rules and I found it funny to read this paragraph:

8.4 Entries can be emailed to … and should be sent as a normal attachment to the e-mail. Send as an uncompressed file. Do NOT send more than one attachment. In the “subject” line of your e-mail message, please include your contest callsign. Please take a moment to check your log via a text editor before sending it, to avoid potential problems. […]

Of course you think: obvious stuff, nothing funny. Well it is to me. It is the proof that there are operators all over the world in every contest making the log checking progress a tough job. I wrote code to strip the attachement from an email on the server. What if the log is not sent as an attechement but copy/pasted in the mail’s message body? My code won’t find a log. So write more code to intercept copy-pasted logs. OK, then what? I have nice smooth running code to parse Cabrillo logs. But what if someone sends a DOC? A PDF? An XLS? It happens! Or a ZIP file? My goodwill ends before that. No ASCII, no parsing. Period. Some guys don’t know what Cabrillo is or how to make it and send a bunch of files. Then my code detects a dozen files in the attachement – but which one is the acual log?

It all sounds easy in theory: send a mail with a Cabrillo-formatted ASCII file. The code for that is easy and written quickly. As always, intercepting all exceptions caused by ******* operators is hell and makes for more complex programming.

And that’s only to receive the logs. Once you start processing the logs, the saga continues. “Cabrillo” is a standard of which many interpretations exist. A long time ago I wrote some code to parse my own logs generated by N1MMLogger. That was easy: code written in a couple of hours and the code ran very fast. Then I used this code on logs coming from a broad variety of loggers. Code to the thrash bin, write code running at snail-speed (compared to first version) and try to catch all possible exceptions in Cabrillo formatting. And then, there’s the guy mutilating his log by hand…

To paraphrase my friend’s ON4CCP’s “80/20” rule: 80% of the time is spent to catch 20% of the bad logs, but I think the actual ratio will be 95/5 or even more skewed… I think I’m repeating myself.  🙂

I gave up on this. And I also should give up trying to expose / explain the problem to the contest community. It is not appreciated. I offered to give a short presentation about the log checking process and software for the UBA contests on the occasion of the annual UBA meeting last May. One of the people in the organizing committee declined my offer because there probably isn’t a big interest in this (granted) and moreover that no one is waiting to be ridiculized for their own stupidities. I agree on the stupidities, but would I ever ridiculize someone?

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.