Tom Emerson wrote:

>The BSA Arizona examples note AutoDesk software.  AutoDesk / AutoCAD is a 
>tough software to replace with a GNU/Open Source equivalent.
>
>In the case of a commercial package for AutoCAD, I don't see how it could 
>be replaced by a GNU'd equivalent.  (for specific engineering and design 
>disiplines, there are very sophisticated software packages that run on top 
>of AutoCAD).
>
>Does anybody on the list have experience with CAD/CAM/CAE GNU/Open Source 
>software?
>
>Doing a quick google, I found these, some that claim proper use of the 
>DXF format (AutoDesk file formats):
>
>Electric, GNU.
>http://www.gnu.org/software/electric/electric.html
>
>LUnIx by Tech-EDV, collection of CAD links for Linux
>http://www.computer.privateweb.at/tech-edv/default.htm
>
>LUnIx by Tech-EDV, another copy
>http://web.utanet.at/tech-edv/
>
>GNU-CAD-CAM stalled project?
>http://gnu-cad-cam.sourceforge.net/
>
>GNU-CAD, stalled project?
>http://sourceforge.net/projects/gnu-cad/
>
>Some of the systems mentioned at by 'Tech-EDV' look for-real.  It's been 
>several years since I had a need for AutoCAD, but thinking back, I would 
>be hesitant to replace it with anything else.  It's a religious thing, not 
>so different from the choice of vi vs. Emacs as a text editor.  (ah, that 
>might not be such a good comparison ... let me say that the AutoCAD I am 
>familiar with is a very powerful tool, but you do need to know how to 
>use the tool!)
>
> - tom e.
>
And you've brought up our Achilles Heel, or at least the road that open 
source needs to travel to become "mainstream" on the desktop.  Yes, 
"it's the applications, stupid...".  I was basing my rant ;-) on the 
Microsoft software that these companies were busted on, but AutoCAD is 
in fact a more salient example of the dilemma most of these companies 
face.  Are they going to go out and buy a copy of Kylix or code with 
Qt/KDevelop or GTK+ and write their own version of AutoCAD, then give it 
away?  Probably not in our lifetime unless the individual/team is 
incredibly dedicated/motivated.  

Specialty software is just that: it caters to a specific set of skills 
and requirements of a specific user group, and long-term support is a 
large component of its longevity and the loyalty of its customer base. 
 I would venture to say that, in cases like this, the OS is immaterial, 
and the application is everything.  If a commercial version is the only 
(read "industry standard") version, then these companies are between the 
rock and the hard place.  

A case could be made, though, if a software vendor's customers scream 
loud enough, that the application in question might be ported to a 
different (read "open source") platform, the savings compensating the 
need to pay for all needed copies of the application's licenses coming 
out of the lower cost of the open source  OS.  This only legitimizes the 
open source platform even more, as ultimately, again, its the 
applications that define a platform's success.  In a very real sense, 
Microsoft, by it's own licensing schemes and its greed to "incorporate" 
ever more peripheral functions into the operating system (or directly 
compete...witness the newest plan to market CRM software), may end up 
shooting itself in the foot....trying to garner an ever-growing 
percentage of the licensing fee "pie", they may push application vendors 
to competing platforms where their former customers have fled for 
relief.  Utimately, when a "critical mass" of applications have migrated 
to open source platforms, we should see a return to genuine competition 
on all fronts......