[Dxspider-support] An observation about adding Skimmer spots

[Michael Carper, Ph.D.]

Lee,

Not to divulge my source... but one of the other members of this forum has
Skimmer spots on one of his nodes and I obtained them from there.

Overall... what I think you and I would easily agree about... is that the
balance of cluster spots have moved from user-generated spots to Skimmer
generated spots... just as spotting (in-general) have moved from RF packet
cluster nodes to Internet-based telnet sessions.

The point is... for better or worse... Skimmer spots are the future of DX
spotting for all modes (including voice... for which I'm sure someone is
working on that now).  Users don't create as many spots as they used to (in
my opinion).  This is largely because Skimmer sources either beat them to
it or have "dumbed-down" the user community to the point where they no
longer know how to create a spot (even if it's done by their software).
Heck, most of them have no idea where to find an announcement or WWV
posting.

So... the future of DX spotting is automated through something like a
Skimmer.

What I found quite surprising is that most of the users who saw these spots
on my cluster were completely confused by them.  They thought they were
"fake"... "deliberate QRM"... a bug in the software... a defect in the
cluster code... and for that reason, they began to complain.

It makes me think that - even though Skimmer spots and RBN (for CW, at
least) have been around for almost a decade, most folks are unaware that it
exists.

Puzzling.

Anyway... my cluster node was loafing with these Skimmer spots coming in.
 (Is there a RTTY contest this weekend?)

Mike, WA9PIE

On Sat, Aug 20, 2016 at 5:10 PM, Lee Sawkins <ve7cc at shaw.ca> wrote:

> Mike
>
> I am curious where you are getting your Skimmer spots from.
>
> At my node today I see about 30K skimmer spots per hour and about 600
> human generated spots per hour.  During really busy contests, such as the
> CQ WW CW, I have seen over 220,000 skimmer spots per hour.  Most logging or
> contest programs have some difficulty handling over 200,000 spots per hour,
> even if the node can.  Individual users try to handle this massive flood by
> setting filters so they only receive a small percentage of these spots.
> This still leaves the user with a huge number of duplicate spots.  It is
> very helpful if the node can remove these duplicates.  To do this the node
> needs to keep track of which spots each user has already received over the
> last few minutes.  Since several percent of spots are either busted or
> unique, the node needs to find these and drop them as well.  Most unique
> spots are actually busted, either in frequency or call, or are non cqers.
> Just a couple of things to look out for if you are planning on modifying
> Spider to handle skimmer spots.
>
> Lee VE7CC
>
> ------------------------------
> *From: *"Michael Carper, Ph.D." <mike at wa9pie.net>
> *To: *"The DXSpider Support list" <dxspider-support at dxcluster.org>
> *Sent: *Saturday, August 20, 2016 7:04:05 PM
> *Subject: *[Dxspider-support] An observation about adding Skimmer spots
>
>
> This is more of an editorial, based on a recent experiment.
>
> I was able to add Skimmer spots to WA9PIE-2.  Here's what I observed:
>
> - the CPU utilization went from an average of about 8% to about 12% (not
> too bad)
> - the rate of disk consumption went up quite a bit (expected)
>
> Those things were fine.  But the users didn't like them.  I had one guy
> email me and he thought that the Skimmer spots were some sort of "DQRM"
> (deliberate QRM).  He said something like, "I've never seen spots with
> mode, signal strength, and WPM in them before."
>
> So as an experiment... it's fine.  My server had no problem with the
> volume of Skimmer spots... even with over 100 users connected.  (By the
> way, I also "isolated" the source of the Skimmer spots so they would not
> propagate outside WA9PIE-2.)
>
> But I've turned it off.  I don't see the point in trying to do it until
> users can "opt-in" for them (similar to what CC Cluster does).
>
> Mike, WA9PIE
>
> _______________________________________________
> Dxspider-support mailing list
> Dxspider-support at dxcluster.org
> http://mailman.tobit.co.uk/mailman/listinfo/dxspider-support
>
>
> _______________________________________________
> Dxspider-support mailing list
> Dxspider-support at dxcluster.org
> http://mailman.tobit.co.uk/mailman/listinfo/dxspider-support
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.tobit.co.uk/pipermail/dxspider-support/attachments/20160820/3c3260d7/attachment-0001.html>