Jim Donohue wrote:
So we actually agree? The system was seriously flawed but the Crew should
have caught it anyway as there were multiple and pretty obvious indications
that competent sailors should have caught.

Jim

G To a point. I don't think the system was "seriously" flawed, since
SOP and good training should have caught it, with the system as is/was.
I tend to use "competent" in a stricter sense .... they may well have
been competent mariners, but they sure screwed up.
If I have a beef, it would be (as expected) with simply doubling up on
GPS, especially on approaching Nantucket. Although a 2nd GPS would be
nice (and generally standard), I would still prefer that double checks
would involve different systems..... Radar, Loran, Fathometer,
Visual..... when available.

otn

Jim Donohue wrote:
As I projected responding will do little good. The way they are fixing
this Jeff...is to plug in multi GPS. Don't you think it odd Jeff that the
fix for over-reliance on GPS by making the system even more dependent on
GPS.

Where does it say that?


And you do read strangely...multiple independent position receiver
inputs...turns automatically in your mind to multiple independent
sources...I wonder why this is? Like your use of "form" could it be you
can't find the right language so you substitute your own?

Two identical GPS's are not "independent". They may help in some
failure modes, but certainly not all.

However, the recommendations also talk about the need for for techniques
other a large pile of GPS units. In particular, the cruise lines were
advised to:

"ensure that all watch officers
adhere to sound watchstanding
practices and procedures, including using
landmarks, soundings, and navigational
aids to verify a vessel’s position,
relying on more than one source for position
information, and reporting to the
master any failure to detect important
navigational aids."

You seem to keep forgetting about this part. This was the specific
advice to the cruise lines, and by implication, their ships' officers.
All of the other advice had to do with the design of the hardware,
which, while somewhat interesting, is not directly related to the
actions of the deck officers.

So what lesson can the typical daysailor/coastal cruiser take from this?
Is it that they should always run two GPS's and continuously compare
the answer? I don't think so, though there may be time when its prudent
to have a second one handy. The advice is clearly to not be so over
reliant on a single source, and to use traditional piloting techniques:
landmarks, soundings, navaids.



One of the reviews of this grounding pointed out that weather was in fact a
factor. It was so good that the crew was complacent. Given your line of
reasoning we could simple claim the good weather made the crew complacent
causing all of the other items that you ascribe to over-reliance on GPS.
Kind of silly but then so is your interpretation. Why not presume the NTSB
people meant what they say?

Why do you ignore what they say? Their specific advice to the seamen is
to remember to use piloting and not to be so reliant on a a single
source for position.

There actually was a pretty big error...the technique used by Raytheon to
flag bad data was different then the one the bridge system was using to
respond to bad data. In my view as an electronic systems guy that was a very
big error...particularly in a system that could lead to a people/property
hazard.

I agree though that it should not have been so significant because there
were multiple other ways to catch it in this case. But I could postulate
cases where it could be quite serious quite quick.

In the heavy boat business I would have thought two or three GPS receivers
would be standard aside from any additions like LORAN. Three is a little
magic because it almost always allows you to isolate the mis-behaving unit.

I don't think you can integrate radar into such a system. You can show the
picture or the picture enhanced but the software to match a view to a map
would be very difficult. If however you know where you are (or think you
do) it is possible that you might match the patterns. Cross check
perhaps...but quite sophisticated software.

Of course your people can cross check radar and fathometer. If you can keep
them focused to do so. A known hard problem.

Jim

"otnmbrd" wrote in message
ink.net...
Jim Donohue wrote:
So we actually agree? The system was seriously flawed but the Crew
should have caught it anyway as there were multiple and pretty obvious
indications that competent sailors should have caught.

Jim

G To a point. I don't think the system was "seriously" flawed, since SOP
and good training should have caught it, with the system as is/was.
I tend to use "competent" in a stricter sense .... they may well have been
competent mariners, but they sure screwed up.
If I have a beef, it would be (as expected) with simply doubling up on
GPS, especially on approaching Nantucket. Although a 2nd GPS would be nice
(and generally standard), I would still prefer that double checks would
involve different systems..... Radar, Loran, Fathometer, Visual..... when
available.

otn

Jim Donohue wrote:

In the heavy boat business I would have thought two or three GPS receivers
would be standard aside from any additions like LORAN. Three is a little
magic because it almost always allows you to isolate the mis-behaving unit.

As stated, two are standard ...... three is overkill. After all, with
multiple radar, Loran, fathometer, celestial capabilities, etc., they
have any number of proven methods to navigate to their destination with
the needed accuracy, and it's not like a GPS unit is all that frequently
breaking down.

I don't think you can integrate radar into such a system. You can show the
picture or the picture enhanced but the software to match a view to a map
would be very difficult. If however you know where you are (or think you
do) it is possible that you might match the patterns. Cross check
perhaps...but quite sophisticated software.

All ready being done. I was looking at a chart overlay on a Radar,
yesterday. Noted that the plotted position and radar position for the
Seabuoy had a significant (to me) difference ( we can all understand the
possible problems this could create). I didn't have time to determine if
the difference was a plotting, alignment, or off-station problem.
Later on I was looking at a chart plotter (Based on Admiralty charts)
and noted that a particular Range line showed slightly right of channel
centerline. In actuality, it's slightly left.
EG In both cases, since I was employing other methods as checks,
neither problem (admittedly both were extremely minor and should not
have created a problem even for someone unfamiliar with the area) caused
me any grief and were filed away as "possible discrepancies to look for".


Of course your people can cross check radar and fathometer. If you can keep
them focused to do so. A known hard problem.

G Not really. If you consider the number of ships at sea on any given
day, plus mileage traveled and the overall accident rate (which no doubt
has plenty of room for improvement) then the percentage of failures is
fairly small, coupled with the fact that for all the cases such as the
Royal Majesty that we hear of, are the many cases that the problem was
observed and corrected, so consequently, never heard of.

otn