Re: Subscription to NASD CD-ROM (was: embedded ...)

[Brian Lloyd <brian@lloyd.com>]

On Fri, 11 Feb 2000, Bob_Deep wrote:

> Andy Ross wrote:
> 
> I thought that the Jeppeson approach plates, though different looking
> than the NOS ones, where functionally equivalent.  

They are.  I use both interchangeably.

> If memory serves,
> they added a bit more information to their approach plates, but nothing
> you could not figure out from the NOS charts you should have in the
> plane anyway.

That is correct.

> I also believe that it's the FAA that approves ALL instrument
> approaches, going so far as to actually fly them in a specially equipped
> aircraft to verify that everything works together right.  Does Jeppeson
> verify approaches?  Do they go out and spend the money to fly them?  No
> they don't.

Verification of the approach is independent of generating the charts.  
You seem to be arguing that the NOS plates/charts are better than
Jeppesen.  I don't think you can support that statement.

> They got their start by making approach plates that went into binders so
> they where easier to open.  

No, Jeppesen got their start by Mr. Jeppesen actually producing the first
charts for making low-visibility arrivals into the airports to which he
flew.  There were no charts in those days.  He then began sharing them
with his fellow pilots.  All this happened before the CAA ever got
involved.

> The paperback NOS plates got difficult to
> manage in-flight and if you ever dropped your NOS book, it was going to
> take some time to look up the approach plate again.  

You have the horse before the cart.  Jeppesen charts were not produced in
response to the NOS charts.  It was actually the other way around.

> It went from there to this "electronic" format.  This type
> of data did not become popular until we started seeing LORAN units, GPS
> and Internal Nav stuff, but Jeppeson was there when the air carriers
> started looking for this data in a consistent format, so their format
> got adopted by the "big boys" long before the average Joe could afford a
> Loran or GPS unit for their very own, we just where along for the ride.

Until very recently (if it has ever changed and I don't think it has ...
yet) the charts were hand-drawn.  They were not generated electronically
by automatically compiling the data about locations, waypoints, courses,
etc.  The early "electronic" approach plates were just scanned images of
the plates since that was the *only* acceptable source of approach data.  
Sure it makes sense to shift off to approach plates synthesized from the
data (the amount of data required would be a lot smaller too).  One of the
main reasons for this was that the electronic data maintained by the FAA
was "dirty" and there was no defined process for cleaning up that data.

Jeppesen is the only source of which I am aware that has "clean" data.  
Yes, they use the FAA data as input but they do have a process by which
that data is verified and cross referenced.

> It's the government data that is the "legal" data.  You must fly your
> Instrument approach using the government verified data on the current
> approach plate with any NOTAMs taken into account, or it's against the
> rules.  If Jeppeson takes liberties with the data, and "fixes" something
> in the government's certified data, then you cannot legally use their
> approach plate.  Now if they ADD information to the government's data,
> that's OK.. They just cannot change the certified approaches in any way.

We may be agreeing violently here.  Certainly the Jeppesen data must be
verified as correct.  I think that the difference here is that you are
claiming that, since the FAA uses their data and since they are the
ultimate authority, that the data they release to the public in electronic
form MUST be good.  But where we differ is on WHAT data is the proven
data.  

The information I had when I was working on this for a commercial project
was that the only verified data was the approch plate itself, not the raw
data that was initially used to create the plate.  So if an error is
discovered in the plate, the correction is applied to the picture of the
plate and not to the original raw data.  This means that the FAA raw data
was never fully verified nor guaranteed to be correct even tho' the
approach plate IS correct.  The plate evolves as the changes are applied
and there didn't used to be tracabilty back to the raw data itself.

An analogy is to the technique used back in the good (bad?) old days of
programming where programmers would patch the object code in core and then
save the core image.  The image would be correct but the original source
code may or may not have ever been updated.  Sure this is stupid and no
one would even consider doing that today but it is a fact that this
approach to debugging was used on systems in the past.  It is possible
that the FAA is still committing the same sin with regard to the approch
data.

Now it is possible that, in the last couple of years since I last looked
into this, the FAA and/or NOS has put a process in place to "clean" the
raw database thus producing guaranteed good data from which they
can synthesize the approach plates.  That may have happened and I may have
missed it.

But I am only human and have been known to make mistakes in the past.  
The only correct answer here is to go and verify the correctness of the
data either by examining the data itself or by verifying the process by
which the FAA verifies the data.  We can argue until we are blue in the
face but that doesn't change which of us is correct and neither will the
eloquence of our respective arguments. Your arguments, no matter how
strong, will not make it so if it isn't.

> After thinking about it, I believe I know what Jeppeson basically does. 

And Aristotle, after thinking about it, deduced that heavy things fall
faster than light things.

> They provide a consistent input format.   A Jeppeson database is a
> standard well known format, that will not change.  The government may
> change their data format and different countries may have vastly
> different data formats, ranging from paper maps to databases.

Jeppesen uses an extension on the ARINC standard for data format.  They
probably are members of ARINC and had a fair bit of input and involvement
in the specification of those formats.  I doubt they worked in a vacuum.  
Also, as more types of data have been created, Jeppesen has responded by
adding to the format (their own extensions) and ARINC hasn't yet changed
their "standard".  I don't know where the FAA sits in all this so I won't
comment there other than to say that I have never been impressed with the
speed with which the FAA has embraced changes.  My *guess* is that the FAA
lags in this process but I don't *know* for sure.

> Jeppeson also crosses other government sources of information and adds
> this to their data.  Thus the "additions" to the standard approach
> plates.  I seem to remember where they added the local center's comm
> frequency so if you miss the approach you will have it written some
> where and some other "nice but not necessary" information.

This is a matter of presentation and doesn't tell us exactly what is in
the raw data.  It appears that you are equating what is on the approach
plate with what is in the database and I am not sure that your supposition
is correct or supportable.

> This is all for "in the states" data.  Outside of the states, Jeppeson
> does have some good data, and again in the same format.  Jeppeson is
> selling data in a consistent format from various government sources.  I
> seriously doubt they are making any modifications to the data that they
> receive from the various governments around the world.

Maybe and maybe not.  Have you asked them?  If you think about it, most of
the critical data doesn't change.  Airports and mountains don't move very
much from one rev of the data to another.  Critical navaids don't change
very often either.  I beleive that it is very possible that Jeppesen DOES
verify the data.  Remember, there are other ways to verify data than to
fly your own airplane there and fly the approach.

> What does this mean for us?
> 
> Well, I'm basicly saying that if we can deliver a means for pilots to
> print or view approach plates let's say.  Then if we have the certified
> data from the fed, you should be able to legaly use this data to fly by
> in IFR conditions, assuming we can faithfuly render the data.  

And here is where we disagree.  You have made the assumption that the
approach plates are mechanically derived from the raw data and I hold that
this *may* be an erronious assumption.  I *know* that this asumption was
incorrect a few years back but I don't know if or when this changed so I
am not making any assumptions at this point.

> This will eliminate the need to keep a bookcase of NOS charts in the
> right seat, and you can simply print all possible charts over your
> intended route, or for the area you normally fly.  If you have a
> computer on board, then you can load it with the data for the region
> you could possibly ever fly, and have a ready referance of flyable
> charts just a few keystrokes away, should you get in a spot with the
> weather closing in.

I know that both Jepp and the FAA are moving in this direction but I am
not sure that the FAA data is ready to do this yet.

	WE NEED VERIFICATION!

> I suspect that the FAA would be VERY picky about software that displayed
> and printed such information, and would want to certify such software
> before allowing it to be used for real IFR operations.  But, I also
> suspect that the NOS folks would be VERY interasted in the software, and
> perhaps they would sponcer a FAA Review and publish "approved" versions
> of the open source.

No arguement there.  I know of a couple of projects to do this that are
quite far along.  I still think that only images of the existing hard-copy
approach plates from either Jepp or NOS are accpeted as being correct.  
Again, I may be wrong and things may have progressed to the point where
synthesized approach plates are acceptable but ...

But I not about to make a claim one way or the other without verification.  
I strongly urge you to either verify the quality of the FAA data or to
continue to hold the FAA data with suspicion as I do.  Don't go off
assuming the data on the FAA CD-ROM is clean, correct data unless you know
it to be so.

Brian Lloyd
brian@lloyd.com
+1.530.676.6513

-
Archives of linux-aviation: http://mail.nl.linux.org/lists/linux-aviation/
To unsubscribe: send the command "unsubscribe linux-aviation" in the body
of a mail message to <Majordomo@mail.nl.linux.org>.