| |
ZL3AI > APRDIG 07.02.03 11:35l 227 Lines 10153 Bytes #999 (0) @ WW
BID : 838-ZL3AI
Read: GUEST
Subj: TAPR Digest, Jan 31, 3/5
Path: DB0FHN<DB0RGB<OK0PPL<RZ6HXA<SP7MGD<VK7AX<IK6PYS<VK3TE<ZL2BAU<ZL3VML
Sent: 030204/0914Z @:ZL3VML.#80.NZL.OC #:58785 [Chch-NZ] FBB7.00i $:838-ZL3AI
From: ZL3AI@ZL3VML.#80.NZL.OC
To : APRDIG@WW
Subject: Re: New spec - working notes
From: Miller Scott Contr 30SCS/FTI <Scott.Miller@vandenberg.af.mil>
Date: Fri, 31 Jan 2003 20:31:24 -0000
X-Message-Number: 19
I think we can live with degrees. Everyone still uses degrees for compass
bearings, don't they?
Yeah, I'm aware of the problems with converting things like NMEA format - I
recently wrote a GPS data logger that stored coordinates in a similar
binary format, and it was a bit more complicated than I'd like. But
allowing everyone to use their source data is what got us into the mess
we're in now, with things like GPGGA sentences going out unmodified over
the air, and confusion between units. For a new spec, I think consistency
is more important than convenience. Plus, devices like the antitracker
will benefit a lot from a more consistent spec. If you want to pull out
just positions, you only have to know one message format - you just watch
the stream and skip over anything you don't want. Same with weather data.
If I want to build a portable LCD display to show wind conditions at the
surrounding paraglider flying areas, I can do it only knowing that one
message format. I don't have to do more than one unit conversion, and I
don't have to understand things like Ultimeter 2000 messages.
Scott / N1VG
----------------------------------------------------------------------
Subject: Re: New spec - working notes
From: "Ray McKnight - WB3ABN" <shortsheep@worldnet.att.net>
Date: Fri, 31 Jan 2003 12:33:18 -0800
X-Message-Number: 20
Why are we not considering the worldwide standard already in use, by
WMO?
If we're talking weather, and surface observations, if we don't follow
recognized standards then all you're doing is introducing confusion. All
major world met organizations generally follow WMO, including the FAA (in
the U.S.) for METARS (another worldwide standard), as well as NOAA/NWS,
which has adopted WMO almost verbatim. There will always be regional
differences such as the kts/mph debate, but overall we should adopt WMO, or
forget porting some kludged standard outside of our little APRS world. The
original APRS protocol loosely followed WMO by way of NWS standards, but it
seems the keepers of the "spec" can't even keep THAT up to date to follow
NWS standards. The APRS weather spec appears to be driven by what
equipment manufacturers cram down our throats, not what is the proper way
to do things. Kelvin? Name one surface observation reporting standard that
uses that? (okay, useful when we put a digi on MARS...)
The vast majority of APRS and home weather equipment is manufactured in the
U.S., or at least FOR the U.S. market. I don't see how it would be
feasible to include multiple reporting protocols, or as a minimum, custom
configurable to your region (like professional weather translators provide
like RM Young and TWI). So, that leaves us with settling on one standard.
It would be nice to have the ability to at least keep it current, if not
customize it completely (if customization IS incorporated, then you
introduce the ability for each user to gump with it at will, also not a
good thing if done arbitrarily, like I feel the current discussion is
leading to).
Take a look at FCM-S4-1994 and tell me why we should do anything
differently?
Ray - WB3ABN
Kingston, WA
----------------------------------------------------------------------
Subject: Re: New spec - working notes
From: Brad <BradLists@HeritageQuest.com>
Date: Fri, 31 Jan 2003 14:00:33 -0700
X-Message-Number: 21
While we are on the subject of standard formats.... Does your spec take
into account Map Datums? This has always been a grip with various
countries.
WGS84 vs. NAD27 vs. whatever the UK uses (I don't remember right now)
vs. whatever the next datum is that comes out.
I think Degrees are still the best. I have only heard of Radians and
Gradients being used in engineering fields, not for navigation.
The only benefit I see of Kelvins over Celsius is Kelvins are unsigned. Is
this enough of a benefit?
Just me two cents...
-Brad
----------------------------------------------------------------------
Subject: Re: New spec - working notes
From: Miller Scott Contr 30SCS/FTI <Scott.Miller@vandenberg.af.mil>
Date: Fri, 31 Jan 2003 21:05:12 -0000
X-Message-Number: 22
Well, this is why I'm asking for input in a public forum. =] I did look
around for existing standards, but didn't find much.
FCM-S4-1994 is "Federal Standards for Siting Meteorological Sensors at
Airports." Good info, but I don't think it specifies a data standard.
What looks more promising is the WMO BUFR surface observation format. It's
got quite a bit of stuff that would duplicate the position and timestamp
messages, but we could at least adopt some of the elements. Here's what
looks useful to me:
Averaging time period - seconds, 13 bits
Wind speed - m/s^-1, 12 bits
Wind direction - degrees true, 9 bits
Pressure - Pa, 14 bits
Sea level pressure - Pa, 14 bits
Temperature - K, 12 bits
Relative Humidity - %, 7 bits
Rainfall rate - Kg/m^-2/s^-1, 12 bits
So to answer your question, the WMO Surface Observation BUFR message format
uses Kelvin for surface temperatures. =] And we may not have any digis on
Mars now, but how cold does PCSat get?
Note that BUFR uses odd field widths. I've been trying to avoid more bit
packing than necessary, but it might be warranted here. At least this gives
me a good reference on what units and scales are appropriate.
I think we'd still need at least one cumulative rainfall measurement, since
it looks like BUFR just gives rate over the specified averaging period.
Gust speed would be good, too.
http://www.profiler.noaa.gov/bufr_format_docs/surfaceBufr.pdf
Thanks for bringing this up. If you've got a better existing format to look
at, let me know.
Scott / N1VG
----------------------------------------------------------------------
Subject: Re: New spec - working notes
From: Miller Scott Contr 30SCS/FTI <Scott.Miller@vandenberg.af.mil>
Date: Fri, 31 Jan 2003 21:13:41 -0000
X-Message-Number: 23
Yes, the spec specifically says that all coordinates are to be in WGS84.
And converting Kelvins to Celsius just requires one integer subtraction
operation. Yes, I think it's worthwhile to have it unsigned, and as I
pointed out in a message I sent just a minute ago, the WMO BUFR surface
observation format uses K as well.
Several people have brought up international concerns with the new format.
I started out this spec with the intention of making nothing US-centric,
and I think so far it's doing a good job of it. I'm eager to hear from
non-US amateurs what they'd like changed or included.
Scott / N1VG
----------------------------------------------------------------------
Subject: Re: New spec - working notes
From: "Ray McKnight - WB3ABN" <shortsheep@worldnet.att.net>
Date: Fri, 31 Jan 2003 14:42:49 -0800
X-Message-Number: 24
I failed to include the U.S. adopted reporting standards outlined in
FMH-1 (Federal Meteorology Handbook, vol 1).
http://205.156.54.206/oso/oso1/oso12/fmh1.htm
This is where it gets down to the observation and reporting standards
based on WMO. For instance, there's a whole slew of info dealing with
"algorithms", how the raw data is manipulated prior to being reported.
This part is where we define the sampling rate (most automated stations
use 1 sample/sec, for instance, for wind) and the observation period
(surface observations have several standards depending on whether you
are observing a gust or general winds - note: a gust is not just an
instantaneous peak!). There's actually another entire document dealing
with algorithms, which is quite hard to find.
We have several issues that may not be clear to most folks. There are
two separate issues regarding APRS weather. The first is what I
consider to be the most important - how is data collected and reported.
The second only affects the end user, and I feel is irrelevant to the
first, how is the data displayed within an APRS client. They are
certainly closely dependant on one another, but #1 should always take
priority when designing the protocol. #2 doesn't matter to anyone
outside of APRS. We should strive to disseminate the most accurate
observations, following recognized standards. Flexibility in displaying
data is an issue with each client software. For this reason, it never
satisfies me, I will always use a better method than APRS to view my
data and others if possible. With this in mind, there should be a means
for your APRS client to collect wx data (from others, over the air)
BEFORE it gets mangled by the display protocol.
Lastly, we do have built-in limitations to overcome, or at least address
and deal with. Most APRS folks are using off-the-shelf weather
hardware, most common are Peet, Davis, Radio Shack/OSI, and Dallas.
Each of these units have built-in algorithms for data manipulation (most
importantly, wind speed and gust, as well as rain averages, etc). These
are almost always hard coded. That is, if Peet has determined that wind
speed be averaged over 1 minute, then there's no way to change that,
even though the CURRENT standard is 2 minutes. So you basically can't
access the RAW data in most cases, and must live with the massaged data
that the translator gives you. Professional units generally put the raw
data out on the serial port, and only DISPLAY the massaged data on the
translator readout. Or at a minimum they can be programmed to provide
the correct values based on user defined algorithms.
What's the solution for this? Is it significant enough to be concerned
about? I don't have that answer. All I can say is I don't report data
from my station unless it conforms to a standard I can live with, such
as NWS/FMH-1. The current equipment I have won't support that, so I
don't report. I'm working on upgrading the equipment, but it's costly
and this is just a hobby...
Just my personal thoughts and do not reflect the policies or opinions of
the movie studio...
73's
Ray - WB3ABN
Kingston, WA
----------------------------------------------------------------------
Read previous mail | Read next mail
| |