[Dxspider-support] badnode filtering

[Jim Reisert AD1C]

On Mon, Nov 16, 2009 at 2:15 PM, Dirk Koopman G1TLH
<gb7tlh at dxcluster.org> wrote:

> What I am trying to achieve, with this simple filter, is to prevent
> obviously invalid dx spots being sent around the system. For the absence of
> any doubt: TO and TX would happily pass through a DXSpider node (as before)
> with the pirate filter applied. Therefore they would *still* be available to
> be rejected by more picky software :-)
>
> So far, I cannot (for the life of me) see a downside to rejecting spots that
> DXSpider identifies as invalid. Since nobody else seems to have such a
> comprehensive list of prefixes (with *your* invaluable help btw), if
> DXSpider reckons it's a pirate call, then no other package has the data to
> gainsay that.
>
> Of course if you have the full filter (including dx spot calls) it does stop
> the likes of 1B1AB :-)
>
> If DXSpider nodes act as a primary filters for the rubbish coming from the
> internet, who/what else is likely to do a better job?
>
> I am confused.


Hi Dirk,

What I'm trying to suggest is:

1.  Enumerated bad prefixes/calls added to "country 666" are fine by me.
    Some of them already are:

    1B,X5,X8,ZC6,10G,50V,Q,0,T0

2.  Simply rejecting a call out-of-hand because the prefix can't be determined
    from the country file/table (i.e. those pesky TO/TX calls) is NOT
OK, as these
    prefixes/calls may be perfectly valid, but the country file is not
up-to-date.

- Jim

-- 
Jim Reisert AD1C, <jjreisert at alum.mit.edu>, http://www.ad1c.us