On Feb 16, 7:45*am, nimbusgb wrote:
On 15 Feb, 21:08, Papa3 wrote:

On Feb 15, 9:09 am, nimbusgb

[Jim wrote: * and can provide just as secure a record for badge

flights]

[Ian wrote: * Not even nearly]

Okay Ian - prove it. * Prove that a COTS unit governed under the
manual OO procedures I've laid out in the referenced link doesn't
provide an equivalent level of security. *Come on, I dare ya *:-)

http://home.netcom.com/~pappa3/files...rity_draft.pdf

A cots GPS is still not as secure as an IGC logger.

Your presentation appears to have merit and I'd support using it for
badges.

Ian

"Not as secure..." Why? Specifics. Specifics. I have yet to
hear anyone make a specific argument. Generalities, yes, but
specifics, no.

I'm not trying to stand up on a soapbox here. However, I do expect
folks to do a little more than just spout the "conventional wisdom"
without a little more thought or work going into it. I'll make it
easier on folks. What specific security threats are there which fall
in the realm of reasonable probability:

- Most likely: Someone downloads a log file from a COTS logger and
alters the file using a text editor to fix a height penalty, tweak a
few fixes by a couple of feet to get into an OZ, etc. It turns out
that there a number of low tech and medium tech solutions to this:
* Lowest tech: Seal off any data ports and only allow downloading
and analysis of the file under the supervision of the OO. Very low
tech. Very high security.
* Medium tech: Rely on the encyrption provided by third party tools
such as G72Win or SeeYou. Could someone with access to a Cray
supercomputer for a few hours break their encryption methodology?
Sure. Does it matter as long as the OO takes control of the file
immediately upon download? No.

- Less likely: Someone manufactures a log on their PC using Condor
and SeeYou and loads it up to the logger before a flight. Solutions:
* Very Low Tech: Inspection of the logger to ensure memory is empty
prior to flight. Very easy to do since COTS loggers inherently
provide the ability to manage log files through their UI.
* Medium Tech: Rely on the inherent feature of (some) loggers which
zero out the timestamps of any files loaded into memory which are not
generated by the GPS itself.

- Unlikely: Pilot manipulates the datum in flight (have heard this
one spouted a lot). Has anyone actually looked at what would
happen? Depending on which datums are swapped, the impact is
somewhere between a few hundred meters and ... well a lot of
kilometers. Post flight analysis will pick this up in a heartbeat,
as the glider will have an instantaneous velocity approaching that of
the Starship Enterprise. This is much different than a single
spurios fix or group of dropped fixes where the smoothed average is
highly believable. I can show a trace of this if anyone wants to see
it.

Point being, it's easy to talk about this in vague terms, but when you
put your mind to it, it turns out that there's not that much which
would be required to make a very secure log file for the purpose of
supervised badge flights.

All good stuff to be pondering as we in the Northern Hemisphere await
the return of soaring season.

P3