On Tue, 25 Jan 2005 18:35:45 -0500, Jeff Morris
wrote:

Jack Dale wrote:
On 25 Jan 2005 15:55:04 -0600, Dave wrote:


On Tue, 25 Jan 2005 16:16:50 -0500, Jeff Morris
said:



Yes. On thinking about it it was the early 80s. Remember the Timex Sinclair
and the Tandy TRS-80?


My first was an Osborne Model 1. I thought it was great! The 5 inch
floppies (really floppy) cost $10 (cdn) and held 92K. The screen was
just a bit small.

Jack
I remember how everyone was so excited about the Osborne because we
finaly had a portable computer!

Since it weighed as much as a sewing machine, a more appropriate
phrase is "transportable." I took it to work to calculate marks. I
remember using z scores to scale marks.

I bought it after a student submitted a word processed essay (TRS 80).
And I realized that I could not mark the spelling on his essay.
Should I mark spelling on the others' essays? I began to realize
that these things might have an impact on learning and teaching.

Jack

On 25 Jan 2005 15:55:04 -0600, Dave wrote:


Don't remember what it was. 1620 sounds vaguely familiar. It was a
"scientific" computer, and we programmed it to do mortality tables for an
insurance company. I do remember that is was in a carefully temperature
controlled space, and the way we programmed it was to hand sheets with the
instructions to a punchcard operator who punched the cards and had them fed
into the machine overnight so we got a printout of the results in the
morning. Time was far too expensive to let the programmers anywhere near the
machine itself.

You describe the 1620 perfectly. It was even shaped like a desk with a
selectric on the desk for operator input. The punch-card
reader/puncher was another box about the same size.



Rodney Myrvaagnes NYC J36 Gjo/a


"Accordions don't play 'Lady of Spain.' People play 'Lady of Spain."

Sound like a 650 to me and you probably used fortran.
JR

Dave wrote:
On Tue, 25 Jan 2005 16:16:50 -0500, Jeff Morris
said:


A Z-80? Wow, 15 years ago that was already obsolete, given that Windows
3 and OS/2 were already running on 486's by then. What are you, a
technophobe??


Yes. On thinking about it it was the early 80s. Remember the Timex Sinclair
and the Tandy TRS-80?


Let's see, 40 years ago - the IBM 1620 was over the hill, the 1401 was
commonly used but hardly "state of the art." The 7090 was bigger than a
desk, as was the CDC 6600. Maybe the 360, though you would have had to
be special to see one in 1965. I'll guess one of the early Digital's,
like the PDP 7 or 8.


Don't remember what it was. 1620 sounds vaguely familiar. It was a
"scientific" computer, and we programmed it to do mortality tables for an
insurance company. I do remember that is was in a carefully temperature
controlled space, and the way we programmed it was to hand sheets with the
instructions to a punchcard operator who punched the cards and had them fed
into the machine overnight so we got a printout of the results in the
morning. Time was far too expensive to let the programmers anywhere near the
machine itself.

Dave wrote:

Yes. On thinking about it it was the early 80s. Remember the Timex Sinclair
and the Tandy TRS-80?

I think I had a Sinclair, but I don't remember it much. I kind of
poo-poo'd the early hobby machines because I had a Digital 11/23 at
home, and my partner had a DG Nova 1200. Our first "small" machine was
a DEC "Robin," a VT100 with a Z80 CP/M machine tucked in.

....
Time was far too expensive to let the programmers anywhere near the
machine itself.

After several years of cards decks and 300 baud phone lines, my first
real programming job was with a DG mini-computer where we "owned" the
machine 100%. It was a real trip to see the lights flash as I ran my
program. I never went back to the "mainframe world" after that.

Jr Gilbreath wrote:
Sound like a 650 to me and you probably used fortran.
JR
Was that the one with the drum memory, so programming it efficiently
meant understanding how much the drum went around during each
instruction? Fortunately that was a bit before my time.

My first programming was on a 1620 in 1963, but it wasn't state of the
art then- it was more like a glorified card reader with Fortran
programed in with a punch board. I independently derived the "Bubble
Sort" on that machine!

Right, It was a drum machine. Actually it was before my time also. We
had a IBM 1410 40k for normal processing and the actuaries used the 650.
We wrote in autocoder for the 1410. It was amazing how much program you
could put in 40k.

Jeff Morris wrote:

Jr Gilbreath wrote:

Sound like a 650 to me and you probably used fortran.
JR

Was that the one with the drum memory, so programming it efficiently
meant understanding how much the drum went around during each
instruction? Fortunately that was a bit before my time.

My first programming was on a 1620 in 1963, but it wasn't state of the
art then- it was more like a glorified card reader with Fortran
programed in with a punch board. I independently derived the "Bubble
Sort" on that machine!

Jim Donohue wrote:



Sure they were. by the time they were 200 miles out they had a multiple
mile error.

The chief officier and navigator both stated that a cross check of the LORAN
was done. The second officier stated the LORAN was used only as a backup to
the GPS.

The NTSB found that the LORAN was never checked and should have been.

The 2M IS the navigator. At any rate, in the past, occasional multi mile
variances between GPS and Loran, though not common, did occur ....
frequently when working the limits of Loran ranges. So, a variance was
not necessarily a "drop dead". What it was, was cause to check systems
and do concurrent plots to see if it corrected .... they did not.



They had failure indicators staring them in the face the whole time. They
just never looked. Makes little difference what the failure is if the
indicator is not noted.

Complacency ..... it'll bite you in the ass at the worst possible moment.



If you are going to be navigating, you want back-ups to your back-ups.
Sure, You can carry 20 hand helds and 4 cases of batteries, but what
happens when the Chit, negatively hits the fan and all you know is GPS?


No they were simply not very competent. They were plotting the positions
hourly...but taking the positon from the busted system. Plotting the
positon from one of your 20 handhelds would have blown the whistle before
they were an hour into the failure.

I can't remember if they were transferring plots from the plotter to
paper charts. No matter, doesn't matter if it was a hand held or Loran,
a plot from either would/should start mental alarms ringing.


What happens when the system fails and you are not aware of it and you go
merrily on your way, believing it's working?
No, they were GPS Centric. They believed everything was fine, when a
system check would have said otherwise. They didn't use back-ups/double
checks, they believed the GPS would always work so consequently the
integrated nav system must be correct and never needed checking on, for
whatever reason.

Basically what I'm saying is go back to the first basic error. The GPS
wasn't working, no one checked to see if it was... they were "centric" in
that they assumed it would be, when in truth, for any number of reasons it
may not be and the prudent navigator realizes this and checks, especially
in areas such as the Nantucket-Boston Safety Fairway.

otn


And a major finding was that the system design was deficient in a number of
human factor ways that provided the environment for the grounding. For
instance the GPS had a suitable external alarm...which was not hooked up.
The report strongly suggest that redundant receivers should also have been
part of the system. The system should have cross checked the LORAN and the
GPS. All kind of simple stuff.

Yes to all, however, the officers should also have been doing this since
otn's law applies (anything designed by man, built by man, operated by
man, and maintained by man, is subject to catastrophic failure, for any
of the above reasons.) and THEY were part of the alarm system, since
alarm systems are known to fail.


otn

As I have pointed out before Jeff you simply read the conclusion you have
already reached into the data. But I will try once more though it will
likely do little good.

"Jeff Morris" wrote in message
...
Jim Donohue wrote:
Calling it a "system" was a euphemism. It was a GPS attached to an
autopilot. They ignored the depth sounder, the radar, and visual cues.
Actually, the same thing could have happened to most anyone with an
Autohelm and a Garmin, except the the Autohelm (now Raymarine) gives a
better indication of faulty input.


Neither the page of causals nor the 3 pages of recommendations has a
single mention of the term GPS...not one. There were a number of
failures but not one that indicates GPS was a problem. Incompetent
seamanship is the proximate cause with poorly designed and poorly
operated equipment creating the opportunity for the incompetent seaman to
ground the boat.



You've made this claim several times and its total Bull****! You're
claiming, in essence, that because the satellites broadcast a good signal,
there is no problem with over-reliance on GPS. This is a textbook case in
what can go wrong with such over-reliance.

The bizarre thing here is that on one hand you've been arguing that GPS
should be learned first because it is nearly perfect, but then here you're
claiming that the problem was the "incompetent seaman" who relied too much
on GPS.


To the contrary...two GPSs integrated into the system would have fully
solved the problem. Or integrating the LORAN would have worked as well.
There are places where the LORAN would not work so for a broad solution it
is not as useful as the second GPS. But two GPS or three GPS and the LORAN
would have been fine.

We agree they were over-reliant on the automatic bridge system. We agree
that they should not have been dependent on a single navigational device.
But this single device reliance is a fault of the bridge system design not
the GPS. If either the LORAN or, better, a second GPS the failure would not
have occurred. We would also suggest an automatic cross check or at least a
manual one was a minimum requirement for the operation of the bridge. The
automatic bridge system was defective and the staff incompetent...what more
needs to be said?
In the "Conclusions" section of the report the NTSB describes what
happened and what went wrong. The majority of the 22 conclusions talk
about the failure of the GPS (mentioning it specifically) system or the
failure to double check by other means, i.e. the over-reliance on one
system.

Here's some examples:

5. Had the fathometer alarm been set to 3 meters,
as was the stated practice, or had the
second officer chosen to display the
fathometer data on the control console, he
would have been alerted that the Royal Majesty
was in far shallower water than expected
and, thus, was off course. He would
have been alerted perhaps as long as 40
minutes before the grounding, and the
situation could have been corrected.

Nothing GPS...simply incompetent bridge operation.

6. The watch officers’ monitoring of the status
of the vessel’s global positioning system
was deficient throughout the voyage from
St. George’s.

Does mention GPS but is actually directed at the monitoring of the system.
Suggests the monitoring was deficient not the GPS.

7. Deliberate cross checking between the
global positioning system and the Loran-C
to verify the Royal Majesty’s position was
not being performed and should have been
on the voyage from St. George’s.

Such a procedure supposedly existed according to the Navigator and Chief
Officier. So failure to follow established procedures indicts the GPS?
Nahh

8. Even though it is likely that the watch officers
were not aware of the limitation inherent
in using the position-fix alarm to monitor
the accuracy of GPS position data, it was
inappropriate for them to rely solely on the
alarm to warn them of any problems with
the GPS data.

Are boys did not know how the system worked. Clearly not a GPS problem.
The appropriate handling of alarms and errors is a bridge sytem problem.

9. The sighting of lights not normally observed
in the traffic lanes, the second officer’s inability
to confirm the presence of the BB
buoy, and the sighting of blue and white
water should have taken precedence over
the automation display on the central console
and compelled the second officer to
promptly use all available means to verify
his position.

The incompetence included denying the obvious. That is not a GPS problem.

10. The chief officer and the second officer did
not observe good watchkeeping practices or
act with heightened awareness of the precautions
that are needed when a vessel approaches
the Boston traffic lanes and landfall.

Not a GPS problem.

11. The master’s methods for monitoring the
progress of the voyage did not account for
the technical capabilities and limitations of
the automated equipment.

That was really dumb...using a system as the input to a check on its own
accuracy. Dumb.

12. The watch officers on the Royal Majesty
may have believed that because the global
positioning system had demonstrated sufficient
reliability over 3 1/2 years, the traditional
practice of using at least two independent
sources of position information was
not necessary.

That is a breakdown in the capabilities of the officiers not the GPS.

13. All the watchstanding officers were overly
reliant on the automated position display of
the navigation and command system 25 and
were, for all intents and purposes, sailing
the map display instead of using navigation
aids or lookout information.

Yes they were not competent sailors.

The report continues with other items in the same vein, though focused
more on the problems with the integrated system and the training, such as:

16. Had the navigation and command system 25
autopilot been configured to compare position
data from multiple independent position
receivers and had a corresponding alarm
been installed that activated when discrepancies
were detected, the grounding of the
Royal Majesty may have been avoided.

These independent position receivers would likely be multiple GPS with
(perhaps) a LORAN. Remember LORAN is not an all areas system. They would
have almost certainly prevented this problem whether or not a LORAN was
included.

Although the "Probable Cause" section which follows does not mention GPS
specifically, it is quite short (two small paragraphs) and mentions simply
"overreliance on the automated features of the integrated bridge system,"
the lack of training, and the failure to take "corrective action after
several cues indicated the vessel was off course." In other words, they
relied too much on one source of position (the gps) and ignored others.

While the "Recommendations" section does not mention GPS specifically, it
clearly recommends against over reliance on one system. We've never
claimed there was anything "wrong" with GPS, only that other forms of
navigation are just as important. The report includes comments like:

Actually I think it says one should not rely on a single input device.
Multiple GPS are much more likely the solution then GPS and something else.

Review the bridge watchstanding practices
on all its vessels, and revise, as
necessary, 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.
...

As I have said before failure to look out the window and at the RADAR and
using what you see to verify that all is working correctly is incompetence
upon the part of a bridge officier.

As part of the foreign flag passenger
ship control verification examination
program, verify that the watchstanding
procedures of ships’ officers include the
use of multiple independent means of
position verification.

It is true that the bulk of the recommendations have to do with better
standards for automated systems, but even then it deals largely with the
need to use more than one form of input:

comparing position-receiver data for
significant discrepancies between
position receivers, and subsequent
positive annunciation to the crew;

It says nothing as far as I can determine about different "forms" of input.
It suggests multiple receivers. I would think the mostly likely of this
would be multiple GPS.

Sorry Jim, its clear that you've been disingenuous with us. The NTSB
study is quite specific in finding fault with relying completely on GPS.
Though they don't fault the GPS system itself (i.e. the signal leaving the
satellite) they make it quite clear the overreliance on one electronic
navigation system was the cause of the grounding.

And again you read what you believe not what is there. The NTSB believes
the input should have been redundant both in the automatic system and on the
bridge. So do I. You believe some other "form" should have been
integrated. I believe it was integrated in bridge procedures but was not
performed. Another "form" was not needed though it was there. What was
needed was a level of redundancy either through the system or bridge
operation or, even better, through both. Again the system was deficient as
was crew performance.


Jim

What did I miss? You were not teaching basic navigation without the use of
electronics? Have I misunderstood your intent? Does your daughter use her
GPS with as much skill as she does the boat's compass? She will navigate as
comfortably in low visibility as she does in CAVU?

Jim

"Dave" wrote in message
...
On Tue, 25 Jan 2005 00:25:10 -0800, "Jim Donohue"

said:

So again we disagree. The instructor wanted to teach without the use of
the
electronic navigation systems...I consider this nonsense.

I think you need to read a bit more carefully. Look at what I wrote, not
what you imagined might furnish an excuse for a tirade.

Dave

Remco Moedt wrote:

GPS might be too precise, though.... I know a few people using a bouy
as a waypoint using a GPS coupled to an autosteer......well, they sure
did find out when that waypoint was reached.....g



I've noticed the same phenomenon around "convenient" round number lat
& long positions that make useful waypoints. Seems there are 3 or 4
boats all converging on the same spot on the water that made such a
convenient waypoint to enter.

Evan Gatehouse