From: Brian Calder 
Sent: Wednesday, December 10, 2003 9:20 PM
To: Various Recipients
Subject: RFC: Workshop to develop Open NavigationSurface

To paraphrase a truly unique & marvelous statement of political intent:

"We the People, in order to form a more perfect union, establish 
justice, insure domestic tranquility, provide for the common defense, 
promote the general welfare, and secure the blessings of liberty to 
ourselves and our posterity, do ordain and establish ... an Open 
NavigationSurface"

Folks, it's been a topic of conversations for a while, but we as a 
community now need to make sure that we don't end up with n (n >= 3) 
different implementations of the Navigation Surface that are mutually 
incompatible.  Of course, an entirely uniform implementation is unlikely

to be possible since each commercial vendor will want to add features 
and improvements in order to distinguish their implementation.  But it 
should be possible for us to build a baseline implementation that will 
allow the code to talk some useful minimal mandatory dialect, and make 
provision for the vendors to extend it for their own nefarious (:-) 
purposes.

With this in mind, I'd like to extend an invitation for each of your 
respective entities (I count five: CARIS, IVS, NAVO, NOAA, and SAIC) to 
nominate one (or at most two) technical point(s) of contact to come to 
CCOM for a two day workshop with the goal of building such a baseline 
implementation.  And by 'build' I mean write code for a couple of days, 
so that at the end of this time, we'll have source code that we can all 
access, which will allow all of us to write at least minimally compliant

Open NavigationSurface objects on at least Linux and Win32 platforms. 
This is an important distinction: it's complex to talk about an abstract
entity without an implementation, so the goal of the workshop is to make
a concrete implementation of the ideas, rather than just come up with a 
wishlist or get trammelled in a maze of twisty-turny standards 
committees, all alike.

The goals of the workshop are fairly clear.  By the end of the workshop,
we want:

1.  A minimal 'baseline' Open NavigationSurface object definition that 
contains sufficient information to reconstruct the depth data required 
to make navigational entities, including a concrete format in a platform
independent, vendor independent format which allows extensions on a 
per-vendor basis.

2.  Source code to build libraries to read and write same on at least 
Linux and Win32, checked into a CVS repository which is openly available
to all of the people at the workshop (and anyone else who's interested, 
but in a controlled manner).

3.  Commitments for documentation of the format and the libraries, 
divided by section between the participants, to be held in the CVS 
repository.

Having conferred with He Who Must Be Obeyed, I can volunteer CCOM to act
as host, and trusted [hopefully] independent third party.  We will 
provide an e-mail list to get discussions going at the technical level 
(Development List) and for general discussion 
(General List), and we will also provide our external 
CVS server to host the source tree in perpetuity with password protected
access for all interested parties.  Can't say fairer than that, eh? (:-)

To get things rolling, I'd like to suggest the week of 19th January 2004
for the workshop, with exact dates to be determined according to 
availability of the appropriate technical contacts.  If you and your 
entity are interested in taking part, could you, please, determine:

1.  Who you want to come to the frozen north (or steaming south, 
depending on your point of view) to hack code as well as ice.
2.  Dates of availability of said person(s) within timeframe of 19th Jan
(not earlier, please: I know already that this would be a problem).
3.  E-mail addresses that you would want added to the two lists 
described above.

... and send the information to me so that I can get things set up.

Also, if I've forgotten to send this to anybody that I should have, 
please let me know so that I can do a bit of groveling and get them 
involved.

Thanks, and I look forward to hearing from y'all,
Brian.

--
------------------------------------------------------------------------
Brian Calder (Address) Tel Number (Fax Number)
Center for Coastal and Ocean Mapping & Joint Hydrographic Center Chase
Ocean Engineering Lab, University of New Hampshire, Durham NH 03824