TWL: Free ENC Charts
I have been experimenting with these ENC charts. Here are a few thoughts
about them.
The ENC format is a vector rather than a rasterized image. Lines and points
versus a digitized, dot rendition.
NOAA has been experimenting with raster charts that have less detail but
more emphasis on the detail that a mariner is really interested in. Who
cares about railroad lines and highways on a marine chart?
The vector chart is another method of implementation of this less is better
strategy. In a sense with raster charting you define what detail you don't
want and remove it. With vector charting you start with NOTHING and add
detail. I think you can see where this could be leading. All the big ships
are using vector charts, not raster, at least for primary charting work. If
the small boat field can be migrated to vector charts then raster charts
will be a thing of the past.
There are marketing opportunities galore in this sort of scenario. Don't
get me wrong, I am not trying to paint the ultimate paranoid scenario. But
the possibilities are there and someone is bound to try to take advantage
of it. On the other hand, vector charting could be a way to least cost
surveying and charting which could pave the way for low cost charts for the
small boating community.
My examination of the presently available ENC charts finds that only about
10% of all US charts are presently done. There is only one chart for the
entire Columbia River system and only about 15 for the entire US coast in
Oregon, Washington and California. The chart files are not named with the
NOAA chart numbers, but are using a obscure numbering system which makes
their use very awkward. Finding any particular chart or the next chart you
need is time consuming. This last item is a deal killer.
As far as chart detail is concerned. I examined several charts for areas
that I know very well or have excellent charts for. For instance, Cape
Flattery. The detail is ok, but not nearly as complete as the detail of
rocks awash and just below the surface as in the raster version. In other
places like south of San Francisco at some of the anchorages the detail is
complete enough for anchoring. In other words, the level detail varies and
I have found no documents on their site that describes how this is going to
be fixed. This variation in detail next to the beach is a serious problem
for the small boating community.
Since these are vector charts, it should only take a complaint or report to
get a wreck or submerged rock to be included in the next update patch. It
should be possible to patch the files with your own additions and
corrections and then be able to pass them directly to NOAA for evaluation
and possible inclusion. If NOAA can find some suitable way to reward
contributions, then much of the inshore obstructions might eventually be
documented. Whether any of this idealistic scenario can every come to pass
is not a safe bet.
Mike
Capt. Mike Maurice
Wilsonville, Oregon (Portland).
I have been experimenting with these ENC charts. Here are a
few thoughts about them...
Mike brought up some excellent points about the raster vs vector debate.
I've been spending a fair amount of time messing around with these ENC/S-57
charts too. As a developer of navigation products (and a cruiser!) I have a
couple of comments to add to the mix from my, rather different perspective.
I've always preferred raster charts for a few reasons:
-
They "look" better and seem to convey more information. This is a weak
subjective reason but I think I understand a little more about why this
is...see below. -
I always use paper charts alongside my electronics. I like having the
exact same display (screen and paper) for quick "calibration" between
mediums. My wife and I have a continual dialog while cruising along the
lines of "we're at the 'a' in Bay in the middle - see it?". We keep little
sticky "mice" and Post-It Notes on the paper to mark our position.
Along with my raster displays on board, I also have a full Raymarine C-Map
HSB2 setup with multiple heads. I believe that C-Map (NT+) has come a long
way and Raymarine's change in firmware (July 2003+) makes incredible
improvements in the quality of the image presented. It isn't bad - I would
feel completely comfortable if it were my only navigation display.
Vector data has the advantages of being smaller in size as well as being
capable of higher compressibility through lossless techniques. This smaller
size means that it is easier to move the data around via wireless/cell/WiFi
techniques, etc. If there is anything that I'd bet on, it is that data
delivery by those techniques is going to be everywhere within a few years.
...even chart data. Heck, I'm working on it right now.
My recent experiments with the ENC/S-57 data gave me an "aha" moment about
why raster charts look better. It's really pretty obvious but as I was
decoding the lower level data that makes up a chart, I realized that I, as a
programmer, had the ability to make many decisions about how the data is
actually drawn on the screen. This is especially true with respect to text
placement, soundings display, contours, etc., not to mention color, line
thickness and many other parts of the output. Contrast this to a raster
chart which was produced by a series of professional cartographers who have
spent a lot of their life studying the history and techniques of conveying
information. To be honest, I've never even sat in on a cartography class
let alone know any of the issues that they face. Cartography has been
around for thousands of years. Their ability to solve the issues of clarity
and quality are never going to be duplicated programmatically by a software
developer.
Don't get me wrong...S-57 is pretty good. To go further, I think that it
needs a group of cartographers and programmers to get together and extend
the standards so that "hints" of object placement can take some of the
graphic decisions out of the programmer's hands. As I see it now, there
isn't much room left for cartography in this future vector world and I fear
that a lot of techniques and skill will slowly die away.
For me, I'm going to keep playing with the data and eventually have a
version that supports it.
================
Jeffrey Siegel
M/V aCappella
DeFever 53PH
W1ACA/WDB4350
Castine, Maine
Jeffery,
Are you in a position to compare the detail in a Superwide C-Map cartridge
with the exact same location on a Wide + cartridge? When I did this on a
C-Map cartridge programming PC at a dealer's, I found the obscure locations
up creeks on the Severn River appeared on the Wide +, but not the Superwide
cartridge. I'm looking for empirical evidence here, not C-Map's explanation.
No rush. Thank you.
Ron Rogers
Hi all!
When I was working on the design of cartography systems in the late
60's, we did not have raster,
too memory intensive. The systems were vector based. Most original
data is from a vector database and
enhanced for raster display as a value add. When you see high res.
computer generated special effects,
they are vector, rasterized for display by raster systems (crt's).
Vector systems do not imply poor visual quality.
I do not know but I would imagine that rasterized charts are copies of
paper charts that were generated
from vector databases, therefore making them at least third generation
reproductions.
I would like to see a company do the value added work on the original
ENC vector databases. This value
added work is independent of the information critical to navigation.
Updates to the databases would be easily applied
using standard database manipulation techniques. I think many people
would pay for enhanced ENC charts. Since
this work is less labor intensive than working from the original paper
chart, it 'should' be less costly.
I personally do not think that ENC charts are a government conspiracy to
trap us all into committing to that
form of chart then charging outlandish prices to reduce the national
debt. I think it is placing in the public domain
what the public has paid for.
Regards,
Bob Davis
three enc nav packages:
http://www.rosepointnav.com/default.htm
http://www.globenav.com/
http://www.fugawi.com/indexcom.html
ENC site:
http://chartmaker.ncd.noaa.gov/mcd/enc/download.htm
about 50 charts added per month, all identified by there correct chart
number
perhaps 350 charts now available
IENC site:
http://crunch.tec.army.mil/enc/echarts/IENCFileRequest.cfm
not active since Oct 2003
Jeffrey Siegel wrote:
I have been experimenting with these ENC charts. Here are a
few thoughts about them...
Mike Maurice wrote:
NOAA has been experimenting with raster charts that have less detail but
more emphasis on the detail that a mariner is really interested in. Who
cares about railroad lines and highways on a marine chart?
The vector chart is another method of implementation of this less is better
strategy. >>> snip<<<< All the big ships
are using vector charts, not raster, at least for primary charting work. If
the small boat field can be migrated to vector charts then raster charts
will be a thing of the past.
snip<<<<
Since these are vector charts, it should only take a complaint or report to
get a wreck or submerged rock to be included in the next update patch. It
should be possible to patch the files with your own additions and
corrections and then be able to pass them directly to NOAA for evaluation
and possible inclusion. If NOAA can find some suitable way to reward
contributions, then much of the inshore obstructions might eventually be
documented.
Jeffery Siegel wrote:
As a developer of navigation products I have a couple of comments
snip<<<<
My recent experiments with the ENC/S-57 data gave me an "aha" moment about
why raster charts look better. >> snip <<<
raster chart are produced by a series of professional cartographers who have
spent a lot of their life studying the history and techniques of conveying
information. Their ability to solve the issues of clarity
and quality are never going to be duplicated programmatically by a software
developer.
Comment
With respect Mike, it depends on what you are doing as to whether or not things
like railroads and streets are shown on a chart.
As a delivery skipper you do not need this information.
BUT! as a cruiser it might be very convenient to know which railroad serves
which town and which street is visible from the water.
Many long term cruisers have to plan in advance where and when friends join
them for a short section of the cruise.
Picking up "tourist" type brochures dealing with points of interest in
various towns also man you read about some place that would be of interest if
only it was reasonably close to the waterway and some place you can dock.
CHS is giving serious consideration to inclusion of such non navigational
information as the location of grocery stores, Joe's bait and tackle shop, the
beer and liquor store and Laundromats.
In addition emergency response crews often need vital information concerning
which roads provide access to waterfront and beach locations.
At one hydrographic conference I attended they showed a composite map/chart of
the Vancouver area which merged street maps with hydrographic chart data. This
initiative was driven by requests from ambulance crews wanting better
information as to where they could meet up with a coast guard boat delivering
boat accident victims to shore.
Not having to drive a boat to a marina in order to transfer a patient to a road
ambulance can drastically reduce the time it takes to get someone to a
hospital. You cannot always count on having helicopters available for an
airlift.
A second point about roads. They can often provide a very accurate range when
seen from the water. Provided you can identify which street you are looking at.
I think it is in New York that pilots routinely use streets as alignment
markers for determining when the vessel is correctly positioned at the
designated anchoring place as directed by the harbor traffic control center.
This technique is also used in the Great Lakes where canals literally bisect
urban area.
Concerning the extent and nature of information conveyed by the drawn symbols on
a chart.
Steep slopes and cliffs have in the past been shown as striated lines clearly
indicating this is a prominent cliff.
When the cartographers shifted to contour lines, this became less obvious.
contour lines are a natural development in vector charts since the software can
connect isolated point of same altitude or depth.
Drawing in the lines to indicate a steep cliff is much more time consuming and
does not lend itself to vector style formats.
CHS has always been focused on commercial mariners and the needs of large
ships.
In reality the total number of registered ships sailing in Canadian waters and
which are required by law to have up to date charts on board is very small
compared to the number of recreational boats.
The now recognize that if they expect to sell more charts in future years they
will have to offer data and information which is considered relevant to
recreational boaters.
I'm sure NOAA is coming to the same realization for much the same reasons.
NDI and Maptech is attempting to provide the value added services to the
digital product that recreational boaters want.
However the up front cost of creating the necessary database of information to
provide such services is considerable.
In order to make this a cost effective proposition, both CHS and NOAA must
become involved and begin the process of including data not considered
traditional navigational information.
Regards
Arild
Outgoing mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.555 / Virus Database: 347 - Release Date: 12/23/2003
As a person who has been mucking about with boats for nearly five decades
(who happens to be a computer programmer in the electronic navigation
business), I could not agree with Jeff more.
I have a lot more experience than Jeff in terms of reading and using
nautical charts, but I am NO more competant to make cartographic decisions
within my programming code than Jeff.
One reason vector chart data appears more compact is because you're only
getting a pint instead of a gallon worth of information. When vector data
is as complete as a raster chart or paper chart in terms of detail, the size
of the data becomes huge, even in terms of modern day computer hard drives.
The military's "DNC" vector format contains as much data as a paper chart.
But DNC requires up to three times as much storage space ..... in part
because of sloppy data standards in DNC, but also in part because of the
complexity of the data.
Bob makes an important point when he discusses the potential for merging
other commercially available data with the chart data provided by NOAA in
vector form.
The challenge the industry will face is how to present such a wealth of
information on a small display aboard a boat that is pitching and rolling.
And doing so in such a way that the skipper and crew are not overwhelmed by
the amount of information being displayed.
Dennis Mills
While not being in the electronic navigation business I have worked for
many years on the question of image storage and restoration. Actually
contributed to a proposal about 15 years ago to do Point of Sale
Printing of NOAA charts.
I would simply point out that raster representation of vector data is in
general a long term loser. The format problem is of course real and is
often made significantly worse by the inability of programmer to allow
specialists such as cartographers to drive the display process. Some
simple things tend to demonstrate the point...the proper raster display
for instance is dependent on the display parameters and the zoom level.
You really shouldn't use the same raster under different conditions.
Personal preferences are also better served by vector displays.
I believe that stochastic calculations will always lead to vector
formats being more efficient for storage. Rasters may appear more
efficient in some cases as the raster actually loses a lot of the
information content.
Jim
-----Original Message-----
From: trawler-world-list-bounces@lists.samurai.com
[mailto:trawler-world-list-bounces@lists.samurai.com] On
Behalf Of Dennis
Sent: Sunday, February 15, 2004 8:08 AM
To: jeffrey.siegel@activecenter.com; 'Mike Maurice';
trawler-world-list@lists.samurai.com
Subject: TWL: Re: RE: Free ENC Charts
One reason vector chart data appears more compact is because
you're only getting a pint instead of a gallon worth of
information. When vector data is as complete as a raster
chart or paper chart in terms of detail, the size of the data
becomes huge, even in terms of modern day computer hard drives.
The challenge the industry will face is how to present such a
wealth of information on a small display aboard a boat that
is pitching and rolling. And doing so in such a way that the
skipper and crew are not overwhelmed by the amount of
information being displayed.
Dennis Mills
http://lists.samurai.com/mailman/listinfo/trawler-world-list
To Unsubscribe send email to
trawler-world-list-request@lists.samurai.com
Include the word "Unsubscribe" (and nothing else) in the
subject or body of the message.
Regarding:
As a person who has been mucking about with boats for nearly five decades
(who happens to be a computer programmer in the electronic navigation
business), I could not agree with Jeff more.
I have a lot more experience than Jeff in terms of reading and using
nautical charts, but I am NO more competant to make cartographic decisions
within my programming code than Jeff.
What's that one branch of Computer Science called? Knowledgebased
Engineering.
Cartographic Science may be large and complex, but it's still finite. How
about studying the science, encoding its decision making process, and use it
to drive the presentation part of chartplotting programs?
I can say things like that because I retired from IT 2 years ago and don't
have build that kind of stuff anymore. This is probably what we'll see as
human intelligence moves into our hardware in about 10 to 20 years.
-- Jim
Jim & Rita Ague
M/V Derreen, Monk 36