[prev in list] [next in list] [prev in thread] [next in thread] 

List:       gpsd-dev
Subject:    [Gpsd-dev] Update on the gpsd project, and an invitation to join
From:       esr () snark ! thyrsus ! com (Eric S !  Raymond)
Date:       2005-06-29 13:46:17
Message-ID: 200506291346.j5TDkHDD019156 () snark ! thyrsus ! com
[Download RAW message or body]

I know from your "DGPS corrections over the Internet" page that you
are aware of the gpsd project.  You may not be aware that it has been 
much more active in the last year and has a new website with a lot
of resources and GPS-related technical information on it: 
<http://gpsd.berlios.de/>.

gpsd itself is now being regularly maintained and has acquired
important new capabilities, including

* autobauding and self-configuration -- there is no more GPS type option
* the ability to speak several important vendor binary protocols
* support for multiple GPSes (useful for attitude sensing)
* client-side libraries for application developers in C and Python
* GPS latency profiling and scattergram generation
* actual documentation for everything.

We are also pushing hard for highest possible reliability and 
code quality.  There is a comprehensive regression-test suite,
and the code is now statically audited with splint and dynamically
audited with valgrind before every release.

I am trying to make the gpsd project the one-stop shop for developers
working with open-source GPS infrastructure.  We already have a
diagnostic monitor for SiRF chipsets, and one of our developers plans
to do a Linux-based firmware loader for them.  We have a program that
simulates one or more GPSes, so location-aware applications can be run
with pre-specified track data.  And we have code for auditing GPS
performance.

The project has informal but effective cooperation from the
GPS-chipset vendor with 80% market share; we are cooperating with
the authors of kismet and gpsdrive; and we have attracted at least
three of the open-source world's most accomplished experts on
GPS-related issues onto an active development mailing list.  

I am writing to invite you to join that group of experts. and
to consider the possibility of merging DGPSIP with our project
at some point in the future.  

I'm suggesting this because, among other reasons, a unified
gpsd/dgpsip distribution could avoid the need for users to maintain
the list of DGPS servers or launch dgpsip themelves; instead, hotplug
scripts would handle everything and users would simply see
highest-quality location information with zero administration.

Whether or not the codebases eventually merge, I would very much like
to have someone on the project list who thoroughly understands DGPS 
correction and can advise us on how best to fit it into the new
multisession gpsd architecture.

Our mailing list <https://lists.berlios.de/mailman/listinfo/gpsd-dev>.
I think you'll find it's a group of people who are both capable and
interesting.
-- 
		<a href="http://www.catb.org/~esr/">Eric S. Raymond</a>

Freedom begins between the ears.	-- Edward Abbey

[prev in list] [next in list] [prev in thread] [next in thread]