[CQ-Contest] A Plea to Cabrillo Contet Robot Writers

[Mike]

Vladimir Sidorov wrote:
> Gents,
>
> Before taking decisions on Cabrillo it would be nice to find out:
>
> - what contesters would gain with an implemention of a new standard
> - would it be worth extra expenses for updating the commercial contest 
> software
> - would the new standard allow a non-computer geek accessing the log to 
> eventually edit it manually. So far one can easily edit the Cabrillo file, 
> like correcting weird call-signs or just adding a couple of log lines in a 
> matter of seconds using a simple text file editor
>   
Any changes should be made in the log and then generate the 
file(whatever its format).
> - would such an access require extra software. Please don't forget, for 
> example, the M$ Excel mentionned here costs quite some money and it is not 
> supplied just in the Windows pack...
> - how much would it affect the Internet traffic. Some prople still use 
> dial-up or so, and uploading 10 k or 1 MB file matters...
>   
It's still a text file. Larger because of the tags.
> And finally, or perhaps first of all, tell us non-geeks what you want to 
> fix? Is the "problem" of non-existing RST in a log worth such a headache?
>   
Because each field has a name tag associated with it ( <call:4>NF4L in 
ADIF or <call>NF4L<end_call> in XML, and
a length (in XML the closing tag), there are NO ambiguites. (sp?).
When you read a Cabrillo, or any other delimited file, you have to know 
where in the record the field is, and how long it is,
and/or what the delimiters are..
You and I know 599 is a signal report (but is it sent or recieved), but 
the computer doesn't. It could just as easily import
or export it as call. That of  course would  be a programming error on 
one or both ends.

The problem of the non-existing field is a different one, and either the 
author of the checking software needs to fix it, or
the authors of ALL the programs that export the file need to fix it. 
It's a no-brainer to figure out which is easier, but
probably more difficult to assign blame. Either the sponsers mis-defined 
and published the wrong specs, or the author
of the checking program made an error, or all the other authors made the 
same error. This type of error would exist
no matter what the format. Actually I can visualize a solution, but not 
one which would be widely accepted.

Respectfully, Mike, NF4L<end_of_soapbox>