Hello,

  The GPX 1.0 schema is now official!  Congratulations, everyone!  It
  took a while, but our discussions and debates made for a better
  solution in the end.

  Now let's get some GPX-enabled applications and websites out there!

--
Dan Foster
TopoGrafix - GPS Software, Waypoints, and Maps
http://www.topografix.com - mailto:egroups@...

I guess this is more of a general XML question, but are there any tools
that, given a GPX file and the gpx.xsd file, will validate the GPX file?

--
Brian Lalor
blalor@...

Hello Brian,

Friday, March 01, 2002, 2:52:45 PM, you wrote:

BL> I guess this is more of a general XML question, but are there any tools
BL> that, given a GPX file and the gpx.xsd file, will validate the GPX file?

This is described at http://www.topografix.com/gpx.asp

Use the Xerces SAXCount program to validate your GPX file against the
schema.
http://www.topografix.com/gpx_validation.asp

--
Dan Foster
TopoGrafix - GPS Software, Waypoints, and Maps
http://www.topografix.com - mailto:egroups@...

I took a look at my website from the office, where I use Internet
Explorer 5.5 instead of Internet Explorer 6.0. The xslt stylesheet
transformations do not work correctly with Internet Explorer 5.0 and
5.5 because those browsers were developed before the release of xslt
and xpath as W3C Recommendations.

The trail contents are complete gobbledegook in Internet Explorer
5.5. So stylesheet transformations of gpx documents to display
contents may not be such a good idea after all, or at least not
universally applicable! If you are going to publish trails to a wide
audience using a stylesheet such as

http://www.cableone.net/cdwissenbach/GPXContents.xsl

you should be mindful of this incompatibility. I looked at one other
stylesheet transformation of GPX both at work (ID 5.5) and later at
home (IE 6.0) and noticed a huge difference in appearance and had
assumed that the stylesheet had been reposted. What I now realize is
that what had changed was the browser.

Hello,
I'm Doug Adomatis.  I guess you'd call me a more of a publisher than
a developer.  GPX has caught my interest because its proliferation
will have an impact on what I do.
More about me can be found at my web.
Doug Adomatis
www.travelbygps.com

There's a current thread at sci.geo.sat-nav that I thought someone
from this group would reply to.  I want to know more about how this
issue is handeled in the GPS format.
, Doug

Oops,  I ment to write "how is (waypoint naming) handled in GPX / XML
format.  More specifically, what are the guidelines for waypoint
names (number of characters) and what is the waypoint symbol set used
in the GPX/XML format?  Using these guidlines, how will these
waypoint names and symbols look after being converted to a form that
can eventually be intrupreted by specific brand/models of GPS
receivers?
Clearly, I'm a newbie at understanding these development issues, but
please humor me with a reply.
Doug





--- In gpsxml@y..., "gps_maps" <gps_maps@y...> wrote:
> There's a current thread at sci.geo.sat-nav that I thought someone
> from this group would reply to.  I want to know more about how this
> issue is handeled in the GPS format.
> , Doug

Hello,

Sunday, March 03, 2002, 7:35:06 AM, Doug wrote:

g> Oops,  I ment to write "how is (waypoint naming) handled in GPX / XML
g> format.  More specifically, what are the guidelines for waypoint
g> names (number of characters) and what is the waypoint symbol set used
g> in the GPX/XML format?  Using these guidlines, how will these
g> waypoint names and symbols look after being converted to a form that
g> can eventually be intrupreted by specific brand/models of GPS
g> receivers?
g> Clearly, I'm a newbie at understanding these development issues, but
g> please humor me with a reply.
g> Doug

GPX is a data exchange format, so there are no limitations placed on the
waypoint names or symbols.  It's up to the application that transfers
the GPX data to the GPS to impose any restrictions.

If we made GPX device-dependant, we'd be forced to adopt the least
common denominator out of all the GPS units.  My Brunton GPS is
restricted to 6 character names, and no waypoint symbols!

Symbols are specified by name in GPX.  <sym>Parking Area</sym>, for
example.

--
Dan Foster
TopoGrafix - GPS Software, Waypoints, and Maps
http://www.topografix.com - mailto:egroups@...

--- In gpsxml@y..., Dan Foster <egroups@t...> wrote:
> Hello,
>
> GPX is a data exchange format, so there are no limitations placed
on the
> waypoint names or symbols.  It's up to the application that
transfers
> the GPX data to the GPS to impose any restrictions.
>
> If we made GPX device-dependant, we'd be forced to adopt the least
> common denominator out of all the GPS units.  My Brunton GPS is
> restricted to 6 character names, and no waypoint symbols!
>
> Symbols are specified by name in GPX.  <sym>Parking Area</sym>, for
> example.
>

I'm letting a little bit of device-dependence creep into my sample
trails. What I've been doing is to use the name field as a short
name that will fit on my eTrex Summit display. 6 characters for a
waypoint, or 13 characters for a trail. Then I'm using the cmt field
as a longer version of the name. And, I think that I'll also repeat
the trail name as the first sentence of the description. However, I
have the luxury of supporting only the eTrex in my application. The
application itself doesn't enforce this restriction, except by using
extremely short display windows in the trail and waypoint name
dialog boxes, and by calling the <cmt> field the long name.

<name>NYC</name>
<cmt>New York City</cmt>
<desc>New York City. First settled by the Dutch, this city as become
the capital of the world.</desc>

But the format doesn't enforce this particular usage, because, as
Dan points out, we shouldn't go to the least common denominator in
naming trails and waypoints. I think that we might eventually be
able to converge on a common usage, though.

As far as waypoint names, I was never able to find any Garmin
documentation for the topografix names. Instead, I just matched the
topografix output and documented this on my program website. The
main reason that I repeat the long name as the first sentence of the
description for trails is so that I can see the long trail name in
ExpertGPS, in an effort to get the most utility for the greatest
amount of people.

I think that ultimately, the publishers of trail websites and users
of the GPX application will get to set the usage pattern. This might
be a better topic for the map_authors group. (That's probably the
first place to announce this format.)

Hi,

I was looking at delveloping a Java program that supports GPX
format.  I'm still learning about XML in general, but from what I've
read, JAXB looked like a good API for this.  Until, though, I noticed
that you guys were not using a DTD for this (JAXB v1.0 only supports
DTD right now).  Can anyone suggest an alternative Java API, or has
anyone written a DTD for GPX?

Thanks!
Thor

Thor,

Current versions of Xerces (http://xml.apache.org) support XML schema
validation.

Chris

ithork wrote:

>Hi,
>
>I was looking at delveloping a Java program that supports GPX
>format.  I'm still learning about XML in general, but from what I've
>read, JAXB looked like a good API for this.  Until, though, I noticed
>that you guys were not using a DTD for this (JAXB v1.0 only supports
>DTD right now).  Can anyone suggest an alternative Java API, or has
>anyone written a DTD for GPX?
>
>Thanks!
>Thor
>
>
>
>To unsubscribe from this group, send an email to:
>gpsxml-unsubscribe@yahoogroups.com
>
>
>
>Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
>
>
>

Thanks Dan, Dave.  I think I beginning to get a handel on the this.
Let me see if I have it straight.  If I publish trail maps in gpx
format, my readers will need an appilication that can translate gpx
into a format that can be uploaded to a GPSR or imported into a
mapping program.  Right?  Kinda like I'm using EasyGPS now - thanks
to TopoGrafix.  I can see where it is a good idea to have a common
format that all readers can use.  On one hand, that would make my
life easier - not having to publish in multiple formats to reach a
larger audience.  On the other hand(s), there are other issues that
come to mind.  Two of my first questions would be: Where will my
patrons find these applications?  Will they have to pay for them?
Clearly, I've gone beyond the subject of this thread.  Feel free to
reply with a new subject, say "GPX for Publishers"
, Doug

--- In gpsxml@y..., "davewissenbach" <davewissenbach@y...> wrote:
> --- In gpsxml@y..., Dan Foster <egroups@t...> wrote:
> > Hello,
> >
> > GPX is a data exchange format, so there are no limitations placed
> on the
> > waypoint names or symbols.  It's up to the application that
> transfers
> > the GPX data to the GPS to impose any restrictions.
> >
> > If we made GPX device-dependant, we'd be forced to adopt the least
> > common denominator out of all the GPS units.  My Brunton GPS is
> > restricted to 6 character names, and no waypoint symbols!
> >
> > Symbols are specified by name in GPX.  <sym>Parking Area</sym>,
for
> > example.
> >
>
> I'm letting a little bit of device-dependence creep into my sample
> trails. What I've been doing is to use the name field as a short
> name that will fit on my eTrex Summit display. 6 characters for a
> waypoint, or 13 characters for a trail. Then I'm using the cmt
field
> as a longer version of the name. And, I think that I'll also repeat
> the trail name as the first sentence of the description. However, I
> have the luxury of supporting only the eTrex in my application. The
> application itself doesn't enforce this restriction, except by
using
> extremely short display windows in the trail and waypoint name
> dialog boxes, and by calling the <cmt> field the long name.
>
> <name>NYC</name>
> <cmt>New York City</cmt>
> <desc>New York City. First settled by the Dutch, this city as
become
> the capital of the world.</desc>
>
> But the format doesn't enforce this particular usage, because, as
> Dan points out, we shouldn't go to the least common denominator in
> naming trails and waypoints. I think that we might eventually be
> able to converge on a common usage, though.
>
> As far as waypoint names, I was never able to find any Garmin
> documentation for the topografix names. Instead, I just matched the
> topografix output and documented this on my program website. The
> main reason that I repeat the long name as the first sentence of
the
> description for trails is so that I can see the long trail name in
> ExpertGPS, in an effort to get the most utility for the greatest
> amount of people.
>
> I think that ultimately, the publishers of trail websites and users
> of the GPX application will get to set the usage pattern. This
might
> be a better topic for the map_authors group. (That's probably the
> first place to announce this format.)

Hello,

Tuesday, March 05, 2002, 7:11:17 PM, Doug wrote:

g> If I publish trail maps in gpx
g> format, my readers will need an appilication that can translate gpx
g> into a format that can be uploaded to a GPSR or imported into a
g> mapping program.  Right?  Kinda like I'm using EasyGPS now - thanks
g> to TopoGrafix.  I can see where it is a good idea to have a common
g> format that all readers can use.  On one hand, that would make my
g> life easier - not having to publish in multiple formats to reach a
g> larger audience.  On the other hand(s), there are other issues that
g> come to mind.  Two of my first questions would be: Where will my
g> patrons find these applications?  Will they have to pay for them?

From http://www.topografix.com/gpx.asp :

What are the benefits of GPX?
Here are some of the benefits that GPX provides:
GPX allows you to exchange data with a growing list of programs for Windows,
MacOS, Linux, Palm, and PocketPC.
GPX can be transformed into other file formats using a simple webpage or
converter program.
GPX is based on the XML standard, so many of the new programs you use (Microsoft
Excel, for example) can read GPX files.
GPX makes it easy for anyone on the web to develop new features which will
instantly work with your favorite programs.

What programs already support GPX?
http://www.topografix.com/gpx_resources.asp

Publishing GPS data in GPX gives you flexibility in reaching a wider
audience of users.  Let's consider three types of users:

1. Uses a GPX-enabled program already.
    GPX files can be opened directly in the user's favorite program.
    You could keep a list of GPX-enabled programs on your website's
    resources page, or even create your own XSL stylesheet that
    mentions them.  (see #3)

2. Uses a non-GPX-enabled program.
    Quite possibly, someone has written (or will soon write) an XSL
    stylesheet to transform GPX files into that program's format.  For
    example, Kevin Read wrote an XML to OziExplorer converter.  Once GPX
    is introduced to the public, expect the XML-savvy users to start
    writing converters.  http://www.kevinread.com/HTMLTemplates/locform.html
    Once GPX starts to gain acceptance, expect more GPS program authors
    to support it.

3. Doesn't use any GPS software.
    Remember, a GPX file on the web doesn't have to look like XML.  If
    you've provided a link to your custom XSLT stylesheet in your GPX
    files, you can make the GPX file look as pretty as you like.
    Dave's trails are good examples of how you could create something
    that reads like a trail guide, but contains all the info your GPS
    needs, too!
    Here are two examples:
    Dave's Idaho trail page:  http://www.cableone.net/cdwissenbach/
    Dan's sample:
http://www.topografix.com/gpx/samples/topografix/gpx_sample_mbta.gpx


You should realize that GPX 1.0 is still only a week old, and that it
will take a bit of time before GPX-enabled apps show up.   Try
creating a GPX file with one of the apps on the GPX Resources page,
and see how it looks in some of the other applications.  You might
find it worthwhile to look at the sample files above, and begin
learning about using XSLT to create your own "house format" stylesheet
for GPX pages you publish.

--
Dan Foster
TopoGrafix - GPS Software, Waypoints, and Maps
http://www.topografix.com - mailto:egroups@...

This might have some interest to the members of this list, so I have posted it
here:

The purpose of the Mobile GPS Demonstration Platform (MGDP) project is to
demonstrate a proof of concept for the integration of a number of emerging
technologies including: Embedded/Realtime Java, Mobile, position-based
applications, GPS (Global Position System) technologies, Wireless/Mobile
connectivity, Web Services (XML, SOAP, WSDL, JAX-RPC, etc.), and server-
side J2EE (including EJB and DBMS persistence). The MGDP has been
constructed using commonly available, inexpensive hardware and software
(including open source software) to illustrate that applications of this nature
are
very feasible using existing tools and technologies. It is hoped that the MGDP
project inspires others to push the threshold of what can be done with position-
based computing and the Java platform.

This project will be demonstrated at JavaOne 2002 in San Francisco
(March 25-29, 2002). Drop by the Systronix booth in the JavaOne Vendor
Pavillion and see the MGDP in action!

A detailed white paper describing the MGDP project, architecture and
technologies is available (in .pdf format) at:

http://www.chaeron.com

Andrzej Jan Taramina
Chaeron Corporation: Enterprise System Solutions
http://www.chaeron.com

Dan,
Thank you for the very detailed information.  You've
given me a lot to think about here. I had a look the
examples, viewed the source, and saw the 'hidden' GPS
data.  I now understand that the stylesheet defines
what data in the gpx file is displayed and how it is
to be displayed.
Now it seems that I need to study up on what I can do
with stylesheets.
, Doug


--- Dan Foster <egroups@...> wrote:
> Hello,
>
> Tuesday, March 05, 2002, 7:11:17 PM, Doug wrote:
>
> g> If I publish trail maps in gpx
> g> format, my readers will need an appilication that
> can translate gpx
> g> into a format that can be uploaded to a GPSR or
> imported into a
> g> mapping program.  Right?  Kinda like I'm using
> EasyGPS now - thanks
> g> to TopoGrafix.  I can see where it is a good idea
> to have a common
> g> format that all readers can use.  On one hand,
> that would make my
> g> life easier - not having to publish in multiple
> formats to reach a
> g> larger audience.  On the other hand(s), there are
> other issues that
> g> come to mind.  Two of my first questions would
> be: Where will my
> g> patrons find these applications?  Will they have
> to pay for them?
>
> From http://www.topografix.com/gpx.asp :
>
> What are the benefits of GPX?
> Here are some of the benefits that GPX provides:
> GPX allows you to exchange data with a growing list
> of programs for Windows, MacOS, Linux, Palm, and
> PocketPC.
> GPX can be transformed into other file formats using
> a simple webpage or converter program.
> GPX is based on the XML standard, so many of the new
> programs you use (Microsoft Excel, for example) can
> read GPX files.
> GPX makes it easy for anyone on the web to develop
> new features which will instantly work with your
> favorite programs.
>
> What programs already support GPX?
> http://www.topografix.com/gpx_resources.asp
>
> Publishing GPS data in GPX gives you flexibility in
> reaching a wider
> audience of users.  Let's consider three types of
> users:
>
> 1. Uses a GPX-enabled program already.
>    GPX files can be opened directly in the user's
> favorite program.
>    You could keep a list of GPX-enabled programs on
> your website's
>    resources page, or even create your own XSL
> stylesheet that
>    mentions them.  (see #3)
>
> 2. Uses a non-GPX-enabled program.
>    Quite possibly, someone has written (or will soon
> write) an XSL
>    stylesheet to transform GPX files into that
> program's format.  For
>    example, Kevin Read wrote an XML to OziExplorer
> converter.  Once GPX
>    is introduced to the public, expect the XML-savvy
> users to start
>    writing converters.
> http://www.kevinread.com/HTMLTemplates/locform.html
>    Once GPX starts to gain acceptance, expect more
> GPS program authors
>    to support it.
>
> 3. Doesn't use any GPS software.
>    Remember, a GPX file on the web doesn't have to
> look like XML.  If
>    you've provided a link to your custom XSLT
> stylesheet in your GPX
>    files, you can make the GPX file look as pretty
> as you like.
>    Dave's trails are good examples of how you could
> create something
>    that reads like a trail guide, but contains all
> the info your GPS
>    needs, too!
>    Here are two examples:
>    Dave's Idaho trail page:
> http://www.cableone.net/cdwissenbach/
>    Dan's sample:
>
http://www.topografix.com/gpx/samples/topografix/gpx_sample_mbta.gpx
>
>
> You should realize that GPX 1.0 is still only a week
> old, and that it
> will take a bit of time before GPX-enabled apps show
> up.   Try
> creating a GPX file with one of the apps on the GPX
> Resources page,
> and see how it looks in some of the other
> applications.  You might
> find it worthwhile to look at the sample files
> above, and begin
> learning about using XSLT to create your own "house
> format" stylesheet
> for GPX pages you publish.
>
> --
> Dan Foster
> TopoGrafix - GPS Software, Waypoints, and Maps
> http://www.topografix.com -
> mailto:egroups@...
>
>
> ------------------------ Yahoo! Groups Sponsor
>
> To unsubscribe from this group, send an email to:
> gpsxml-unsubscribe@yahoogroups.com
>
>
>
> Your use of Yahoo! Groups is subject to
> http://docs.yahoo.com/info/terms/
>
>


__________________________________________________
Do You Yahoo!?
Try FREE Yahoo! Mail - the world's greatest free email!
http://mail.yahoo.com/

Hello,

I've updated the beta versions of all my software to add improved GPX
1.0 support.  I added several private elements (route color, leg name,
leg description, and leg url) to the GPX files produced by my programs
so that my users could save their data as GPX without losing any info.
I added a File Info dialog similar to Dave's, which also allows the
advanced user to specify an XSLT stylesheet for the GPX file.  I added
a "Publish to Web" feature which allows you to upload your GPX files
to your favorite FTP server.  An example of the output can be found at:
http://home.attbi.com/~topografix/mystic_basin_trail.gpx

For more info, see http://home.attbi.com/~topografix/

The new software is at:
http://www.easygps.com/beta.asp
http://www.topografix.com/beta.asp
http://www.expertgps.com/beta.asp

--
Dan Foster
TopoGrafix - GPS Software, Waypoints, and Maps
http://www.topografix.com - mailto:egroups@...

ping.

c.


_________________________________________________________
Do You Yahoo!?
Get your free @yahoo.com address at http://mail.yahoo.com

I have posted an announcement of Version 2.0 of Wissenbach Map on
the map_authors group

http://groups.yahoo.com/group/map_authors

in the hope that people will begin to use the format.

I am also now distributing a local-version CD of the Wissenbach Map
program in the Boise area to try to get a few other trails websites
out there, and I've added my web site to the search list on

http://www.travelbygps.com

I hope to see other announcements of programs with support for our
new format out there soon! (In fact, I'm counting on it, as I won't
be supporting my own program with any particular diligence!)

Dave

It would be a cruel world without cordless screwdrivers, ATM
machines, and wysiwyg HTML editors.  What do you use to edit your
XML?  Please dont' tell me "notebook"!
, Doug

--- In gpsxml@y..., "gps_maps" <gps_maps@y...> wrote:
> It would be a cruel world without cordless screwdrivers, ATM
> machines, and wysiwyg HTML editors.  What do you use to edit your
> XML?  Please dont' tell me "notebook"!
> , Doug

I rarely, if ever, edit .gpx files directly. But when I do, I use
wordpad to edit raw text. (I also ride a bicycle to work, every day,
rain or shine, cold or hot.)

You might visit

http://www.xslt.com

for a listing of stylesheet editors to transform xml documents. (I
even saw a stylesheet debugger there, visual XSTL). I tend to be
adverse to expensive new tools, although I do own a cordless
screwdriver!

Doub inquires:

> It would be a cruel world without cordless screwdrivers, ATM
> machines, and wysiwyg HTML editors.  What do you use to edit your
> XML?  Please dont' tell me "notebook"!

Well....I have been known to use notebook for quick on-the-fly changes to
small XML-encoded config files.  Usually use JPadPro or sometimes XMLSpy
otherwise.

Andrzej Jan Taramina
Chaeron Corporation: Enterprise System Solutions
http://www.chaeron.com

I am starting a group about GPS (Global Positioning System) in Europe.

Everybody is invited to join

  http://groups.yahoo.com/group/gps-europe/

to join, just send ANY email to:

gps-europe-subscribe@yahoogroups.com

Hello.

I've started an implementation of a program using GPX 1.0.  I intend to
make it freely available.  Though I'm not a an XML jock, I have a few
questions and I hope you can help me capture the intent of GPX.

1) <time> is specified as "conforming to ISO 8601" but that allows a
    plethora of potential encodings.  Is a conforming GPX reader expected
    to handle any of them?  Could I get a clarification on the precise
    allowable members of this field?   Bonus points for providing suitable
    strftime (as in ISO/IEC 9899:1999) specifiers to make it completely
    unambiguous.

2) By my reading, the example file is illegal.

	
http://groups.yahoo.com/group/gpsxml/files/Sample%20XML%20Files/Kevin%20Read%20U\
ntitled.gpx

   This file, among other deviations, contains "lat" and "lon" instead
   of "latitude" and "longitude".   Are abbreviations for the tag names
   really allowed?

3) Does the <url> tag include protocol?   ("http://www.mountwashington.org")

4) Are there guidelines for <sym>?  Tables for Magellan and Garmin would
    be helpful.

5) There is a reference a the elements in <gpx> that links to "Private
    elements" but there is no such tag on the page.   Is additional
    information on this available?


Additionally, this isn't strictly an GPX question, but I suspect this
group has experience on this subject.  Is there a table of "real
world" field sizes for the various makers of GPSRs available?  For
example, I've learned that Magellan supports 8 character waypoints
and 30 character descriptions in Magellan protocol but the NMEA
limits are 6 and 20.  (I've also learned that at least one well known
commercial program crashes when it gets waypoints from my 330 that have
descriptions longer than 20 bytes, so I suspect this is not exactly a
route well travelled.)  Similarly, are there published "best practices"
for adapting "portable" data for the various receivers?  "Truncate this
field from the right, preserving any trailing numeric data", etc.

Thanx for any help you can offer.

RJL

Hello,

Thursday, April 04, 2002, 10:21:48 PM, Robert wrote:

RL> I've started an implementation of a program using GPX 1.0.  I intend to
RL> make it freely available.  Though I'm not a an XML jock, I have a few
RL> questions and I hope you can help me capture the intent of GPX.

Keep us posted on your progress, Robert, and let us know if you run
into other areas of GPX that aren't sufficiently clear.

In general, GPX was designed with these goals in mind:
  - Easy to parse (only ONE way to represent things)
  - Easy for humans to read

RL> 1) <time> is specified as "conforming to ISO 8601" but that allows a
RL>    plethora of potential encodings.  Is a conforming GPX reader expected
RL>    to handle any of them?  Could I get a clarification on the precise
RL>    allowable members of this field?   Bonus points for providing suitable
RL>    strftime (as in ISO/IEC 9899:1999) specifiers to make it completely
RL>    unambiguous.

From http://www.topografix.com/gpx_manual.asp#time
<time>2002-02-10T21:01:29.250Z</time>

Creation/modification timestamp for element. Date and time in are in Univeral
Coordinated Time (UTC), not local time! Conforms to ISO 8601 specification for
date/time representation. Fractional seconds are allowed for millisecond timing
in tracklogs.

By "Conforms to ISO 8601...", we intended that the time format we
chose would be in a standard format, but we didn't intend that all
ISO 8601 representations would be valid in GPX.  GPX 1.0 is more strict,
and only allows two representations: ("Easy to Parse")

1. <time>2002-02-10T21:01:29.250Z</time>   (fractional seconds)

2. <time>2002-02-10T21:01:29Z</time>       (no fractional seconds)

RL> 2) By my reading, the example file is illegal.

RL>        
http://groups.yahoo.com/group/gpsxml/files/Sample%20XML%20Files/Kevin%20Read%20U\
ntitled.gpx

RL>   This file, among other deviations, contains "lat" and "lon" instead
RL>   of "latitude" and "longitude".   Are abbreviations for the tag names
RL>   really allowed?

lat and lon are the correct tag names.  "latitude" and "longitude"
aren't allowed.  The guiding rule was "choose a tag name that's long
enough to be understandable, but balance that against a desire for
small file sizes".

I've corrected the manual to remove references to "latitude" and
"longitude".  Please excuse my earlier error.

RL> 3) Does the <url> tag include protocol?   ("http://www.mountwashington.org")

Yes.

RL> 4) Are there guidelines for <sym>?  Tables for Magellan and Garmin would
RL>    be helpful.

http://www.topografix.com/gpx_manual.asp#sym
<sym>Scenic Area</sym>
Text of GPS symbol name. For interchange with other programs, use the exact
spelling of the symbol on the GPS, if known.

For any GPS that lists the symbol name in English, use the exact English
text.  Some GPS units (MLR, for example) don't give names, so it
becomes a guessing game.  Even Garmin has trouble keeping the names
straight - they've switched names on several GPS models, and their
MapSource software doesn't always use the same names.

EasyGPS includes the names for many brands of GPS.  You're welcome to
use those for reference.  As mentioned above, nobody does a perfect
job with the names, but in places where two programs share the same
name for a symbol, GPX will do the right thing.

You should never use the internal ID for the name.  <sym>1906</sym>
means nothing.  See "Easy to Read".

RL> 5) There is a reference a the elements in <gpx> that links to "Private
RL>    elements" but there is no such tag on the page.   Is additional
RL>    information on this available?

You can put any tags you want at the end of any element in GPX.  You
need to provide your own schema for these tags.  You can look at the
topografix or wissenbach private data for examples:

TopoGrafix "Active Point" element at top level:
<?xml version="1.0"?>
<gpx
  version="1.0"
  creator="ExpertGPS 1.1b1 - http://www.topografix.com"
  xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
  xmlns="http://www.topografix.com/GPX/1/0"
  xmlns:topografix="http://www.topografix.com/GPX/Private/TopoGrafix/0/1"
  xsi:schemaLocation="http://www.topografix.com/GPX/1/0
http://www.topografix.com/GPX/1/0/gpx.xsd
http://www.topografix.com/GPX/Private/TopoGrafix/0/1
http://www.topografix.com/GPX/Private/TopoGrafix/0/1/topografix.xsd">
<time>2002-02-26T21:50:02Z</time>
<bounds minlat="42.392077" minlon="-71.110768" maxlat="42.406111"
maxlon="-71.075170"/>
...
<topografix:active_point lat="42.402496" lon="-71.088814"/>
</gpx>

Wissenbach "Layer" element in Wpt:
<wpt lat="43.772503" lon="-116.092756"><name>INSPIR</name>
...
<sym>Scenic Area</sym>
<wissenbach:layer>plan_log</wissenbach:layer>
</wpt>

RL> Additionally, this isn't strictly an GPX question, but I suspect this
RL> group has experience on this subject.  Is there a table of "real
RL> world" field sizes for the various makers of GPSRs available?  For
RL> example, I've learned that Magellan supports 8 character waypoints
RL> and 30 character descriptions in Magellan protocol but the NMEA
RL> limits are 6 and 20.  (I've also learned that at least one well known
RL> commercial program crashes when it gets waypoints from my 330 that have
RL> descriptions longer than 20 bytes, so I suspect this is not exactly a
RL> route well travelled.)  Similarly, are there published "best practices"
RL> for adapting "portable" data for the various receivers?  "Truncate this
RL> field from the right, preserving any trailing numeric data", etc.

It's more complicated than just data length, since each GPS model has
it's own list of acceptable characters.  Converting "@MyCar" (Magellan
Meridian) to "MYCAR " (Garmin 38) is a difficult task.  My own method
is:
1. Remove all illegal characters.
2. Truncate to maximum length.

GPX doesn't impose any restrictions on data length or contents, so "My
very 1st waypoint with ASCII JunK!$#@$$$" is a valid waypoint name.
It's up to your program to impose any restrictions before sending to
a GPS.

Hope this helps!
--
Dan Foster
TopoGrafix - GPS Software, Waypoints, and Maps
http://www.topografix.com - mailto:egroups@...

Hello, Dan.

Thanx for the reply.


> Keep us posted on your progress, Robert, and let us know if you run

I'm largely successfuly parsing the inputs.  I haven't started on the
output side yet.   It was interesting that I started laying out my
data structures before learning about GPX and had to make very few
changes to accomodate it.   I'll probably use it as my primary storage
format.

> In general, GPX was designed with these goals in mind:
>  - Easy to parse (only ONE way to represent things)
>  - Easy for humans to read

I'm appreciate both goals.

To go to your first point, could you distinguish why some of the
example fields are encoded as CDATA while others have the tag data
directly in the bodies?

The falls creek example contains:
  <sym>Dot</sym>
  <type><![CDATA[Intersection]]></type>
yet the spec doesn't distinguish those two tags in any way that I see
that allows the different content styles.   Perhaps this is one of
those things that's obvious to XML jocks in which case, please just
point me politly to an authoritative paper.

>  GPX 1.0 is more strict, and only allows two representations: ("Easy
  to Parse")

Do I need to file some kind of formal erratum to get the official
spec clarified, or are message like this enough?   Is there any kind
of revision scheme in place since we've just made an incompatible
change to "1.0" or is slipstreaming it accepted practice here?   (I
don't mind - I'm just trying to understand the guidelines.)

> 1. <time>2002-02-10T21:01:29.250Z</time>   (fractional seconds)
>
> 2. <time>2002-02-10T21:01:29Z</time>       (no fractional seconds)

This, for example, should be spelled out in the spec, becuase I could
(and did!) fill that with an ISO 8601-compliant string that was
against the intent.   Spell it out very plainly.  Are leading zeros
required or optional?  Is 12-hour format with AM/PM ever acceptable,
etc.  (Don't answer the questions; fix the spec. :-)

Similary, clarifying the meaning of the "T" and "Z" in your example
would be helpful.   To this moment, I don't know what they are.



> lat and lon are the correct tag names.  "latitude" and "longitude"
> aren't allowed.  The guiding rule was "choose a tag name that's long

OK, good.

> I've corrected the manual to remove references to "latitude" and
> "longitude".  Please excuse my earlier error.

I've been involved in specifications more complex than this one.  I
understand. :-)


> RL> 4) Are there guidelines for <sym>?  Tables for Magellan and


> becomes a guessing game.  Even Garmin has trouble keeping the names
> straight - they've switched names on several GPS models, and their
> MapSource software doesn't always use the same names.

That's indeed the rub.   I was looking for tables from industry
leaders to help guide and normalize any output I may generate - after
all, avoiding yet MORE divergence in this area would be a Good Thing.

Magellan makes things even ickier by using different icon identifier
to token mappings in different models and even different f/w versions.

> EasyGPS includes the names for many brands of GPS.  You're welcome
to
> use those for reference.  As mentioned above, nobody does a perfect
> job with the names, but in places where two programs share the same
> name for a symbol, GPX will do the right thing.

I can't spot such a list at http://www.easygps.com/manual.asp.
Can you please provide a more precise reference?


> RL> 5) There is a reference a the elements in <gpx> that links to
"Private
> RL>    elements" but there is no such tag on the page.   Is
additional
> RL>    information on this available?
>
> You can put any tags you want at the end of any element in GPX.  You
> need to provide your own schema for these tags.  You can look at the
> topografix or wissenbach private data for examples:

Interesting.   This would be good text to put in the dangling internal
href link to:
http://www.topografix.com/gpx_manual.asp#gpx_private

> RL> route well travelled.)  Similarly, are there published "best
practices"
> RL> for adapting "portable" data for the various receivers?
"Truncate this

> It's more complicated than just data length, since each GPS model
> has it's own list of acceptable characters.

Oh, there's a form of joy I wasn't even expecting.   Thanx for the
tip.

> Converting "@MyCar" (Magellan
> Meridian) to "MYCAR " (Garmin 38) is a difficult task.  My own
method
> is:
> 1. Remove all illegal characters.
> 2. Truncate to maximum length.

So you don't try to get clever with making "@mycarisa78pinto" from
"@mycarisa71mustang" into seperate names, just letting them collide
after you apply the heuristics above?

If it's a problem you haven't seen fit to solve, then I'll feel a lot
better about not worrying about it. :-)


> GPX doesn't impose any restrictions on data length or contents, so
"My
> very 1st waypoint with ASCII JunK!$#@$$$" is a valid waypoint name.
> It's up to your program to impose any restrictions before sending to
> a GPS.

I agree that "cleansing" the data at the edges is most appropriate.  I
was just looking for guidance on the mechanics of that cleansing since
I figured this crowd had dealt with those issues.

> Hope this helps!

Indeed it does.   Thanx very much!
robertlipe@...

--- In gpsxml@y..., Robert Lipe <robertlipe@y...> wrote:
> Hello.
>
> I've started an implementation of a program using GPX 1.0.  I
intend to
> make it freely available. >
> 2) By my reading, the example file is illegal.

...

>
>  http://groups.yahoo.com/group/gpsxml/files/Sample%20XML%
20Files/Kevin%20Read%20Untitled.gpx
>
>
The required version attribute tells the story here.

<gpx version="0.2" ...

The released version is 1.0. That sample probably may not have been
validated against the version 0.2 schema. There is a different
schema for each version of the format. The schemaLocation should
select a schema to match the version.

Protect your application against schema changes by ignoring unknown
elements. Protect other applications from yours by validating test
files from your application against the schema.

Regarding symbols.

Tough question. I've put the ones I use for the Garmin eTrex on my
web page

http://www.cableone.net/cdwissenbach/map.html

>
> RJL

--- In gpsxml@y..., "davewissenbach" <davewissenbach@y...> wrote:
> --- In gpsxml@y..., Robert Lipe <robertlipe@y...> wrote:

> > 2) By my reading, the example file is illegal.
>
> ...
>
> >  http://groups.yahoo.com/group/gpsxml/files/Sample%20XML%
> 20Files/Kevin%20Read%20Untitled.gpx
> >
> >
> The required version attribute tells the story here.
>
> <gpx version="0.2" ...
>
> The released version is 1.0. That sample probably may not have been

Aaaah.   I'd missed that detail.   Perhaps the samples in the 'files'
section of the yahoo group should be updated or at least more clearly
marked as being for a prototype version.

> validated against the version 0.2 schema. There is a different
> schema for each version of the format. The schemaLocation should
> select a schema to match the version.

Forgive my XML virginity here, but I don't know what a 'schema' is.
Just reading it, it appears somewhat like what we dinosaurs would call
a 'grammar'.  In 50 words are less, can anyone offer an explanation of
what a schema is and why we'd care? (References to books or URLs or
whatever are fine; I'm not expecting a free education here. :-)


> Tough question. I've put the ones I use for the Garmin eTrex on my
> web page
>
> http://www.cableone.net/cdwissenbach/map.html

Thanx for the reference.   There is a substantial overlap betweeen
this table and the five (grrrrr.) magellan tables.   There are some
gratituous differences, but since this would  ultimately fall into a
case insensitive table match anyway, I think I can handle this pretty
easily with just some slightly redundant table entries.   Besides, at
the end of the day if an icon gets an inexact translation, it's not a
traumatic event.

This disucssion will help me honor the adage of:

"Be liberal in what you accept and conservative in what you
generate..."

Thanx!
RJL

Thor,

I have recently started using JAXB for another (unrelated) project.
I've been very impressed so far.  There are a few little glitches,
but no show-stoppers.  I'd like to see the GPX schema have a
companion DTD for now...  there seem to still be a bunch of tools
that only support DTDs.

I'll take a closer look at the schema when I get a chance - maybe it
would be an easy conversion to a DTD.

-Helen


--- In gpsxml@y..., "ithork" <ithork@y...> wrote:
> Hi,
>
> I was looking at delveloping a Java program that supports GPX
> format.  I'm still learning about XML in general, but from what
I've
> read, JAXB looked like a good API for this.  Until, though, I
noticed
> that you guys were not using a DTD for this (JAXB v1.0 only
supports
> DTD right now).  Can anyone suggest an alternative Java API, or has
> anyone written a DTD for GPX?
>
> Thanks!
> Thor

--- In gpsxml@y..., "robertlipe" <robertlipe@y...> wrote:
> >
> > The released version is 1.0. That sample probably may not have
been
>
> Aaaah.   I'd missed that detail.   Perhaps the samples in
the 'files'
> section of the yahoo group should be updated or at least more
clearly
> marked as being for a prototype version.
>

I've posted newer samples for GPX1.0 in the Files section.

> Forgive my XML virginity here, but I don't know what a 'schema' is.
> Just reading it, it appears somewhat like what we dinosaurs would
call
> a 'grammar'.  In 50 words are less, can anyone offer an
explanation of
> what a schema is and why we'd care? (References to books or URLs or
> whatever are fine; I'm not expecting a free education here. :-)
>

An XML Schema is an XML document which defines the grammar for
an XML application such as GPS eXchange.

See the standard with supporting documentation at

http://www.w3.org/XML/Schema

I've bought two XML books so far:

XML In a Nutshell, O'REILLY, ISBN: 0-596-00058-8

and

Professional XML Schemas, WROX, ISBN: 1-861005-47-4

>
> Thanx!
> RJL

I am finish my app for doing some GPSR management and I
would like to use the GPX format for saving all files. But I would
like to include some application specific info (like window
positions, preferences, etc.) in the <gpx>.

I couldn't file a spcification for private data in the GPX dev man.
Where should it go? Or, is there no allowance for private data?

Karl