[Xastir-dev] Topic: Functional Requirements, Xastir-NG

Tom Russo russo at bogodyn.org
Sun Jun 15 10:24:19 EDT 2008


On Sun, Jun 15, 2008 at 01:18:30PM +0200, we recorded a bogon-computron collision of the <magne at samfundet.no> flavor, containing:
> I've been reading the recent xastir-ng threads with some interest, and
> while the programming language, programming paradigms and supported
> platforms are interesting questions, I guess the functional
> (and non-functional) requirements should be determined first !?

Yes.

Proper OO design begins with study of use cases and requirements capture.

> What are we trying to accomplish,  why isn't Xastir good enough?

Xastir has become unmaintainable due to years of creeping featurism.  The
Motif library on which it is based is archaic, and the GUI features are too
deeply entangled with the functional core (due to they way it was written) to 
be teased apart enough to replace the GUI tool kit.  And yes, the portability
issue is a big one -- it is too tied to X11, making it difficult (almost
impossible) to port to non-X11 platforms.

> Are we looking out for handling more and different protocols
> (AIS, Mode-S ES, ...) ?   Do we want to support an instance without
> a GUI (for serving objects, alerts, message handling) ?  Do we
> want to support small devices (how small?  PDAs, phones, AVR, PIC) ?
> Do we want to have possibility for different GUIs, or is a single
> GUI like the current good enough?   What are the operations we would
> like to perform in the GUI (as users) ?

Good questions.

-- 
Tom Russo    KM5VY   SAR502   DM64ux          http://www.swcp.com/~russo/
Tijeras, NM  QRPL#1592 K2#398  SOC#236 AHTB#1 http://kevan.org/brain.cgi?DDTNM
 "It's so simple to be wise: just think of something stupid to say and
  then don't say it."  --- Sam Levinson




More information about the Xastir-dev mailing list