[cad-linux] Re: benefits of open source (was: database application scenarios)
- From: "Jeffrey McGrew" <JMcGrew@xxxxxxxxxxxxxx>
- To: <cad-linux@xxxxxxxxxxxxx>
- Date: Wed, 14 May 2003 16:37:18 -0700
> I guess its a matter of what you take as given. For an=20
> open-source system, I=20
> take the longevity (ability to keep the code), honesty=20
> (ability to see the=20
> code), and momentum (ability to change the code, but=20
> requirement to disclose=20
> it) as the givens. The value is not a given until the=20
> security, stability,=20
> and speed are there to enable it to compete with the=20
> commercial alternative. =20
> These things tend to grow faster in open-source because of=20
> the one-up-manship=20
> of the contributing programmers. But the security is=20
> definitely not a given,=20
> since the code is available and a cracker can find the=20
> weaknesses to exploit.
This is a very good way to look at it.=20
> So, what does this project need? =20
> marketing and awareness (marketing to users and developers)
Finally an area where I can help out. :)
> Feel free to add to the list or contest a point.
I would only add two things:
1. A UI that adopts consistent guidelines. The terrible
UI of a lot of CAD systems has really lead to a division
and alienation within a lot of Architecture offices between
the 'old guard' and us younger generation. We don't learn
enough, due to being tied to a bad CAD system, and their
superior knowledge doesn't get added to the project 'cept
for redlines. If the UI followed some standard; whether
it be the usability standards put forth by Apple, Gnome,
or a custom document, I think that the UI will be a=20
*critical* issue for it's adoption. I read about Linux
stuff for years, but until Mandrake & Redhat started making
it accessible, it wasn't really available to me. Now,
I've had a place to start, and I'm learning how things
really work; but having something to hold my hand at
first is a real boon. The other thing is that a better
UI really helps in training and productivity. I used
to teach AutoCAD, and in the time it took me to teach
someone how to draw something as simple as a bathtub
in Modelspace I can now show someone how to draw a
whole house to the end of schematics, in 3D, in Revit.
This is mostly due to a more accessible UI and a
much better working methodology that Revit has
over AutoCAD. This better UI and working methodology=20
has doubled my productivity alone.
2. A very simple way to extend the system for
automated content. If there is a way that someone
who is CAD savvy; but not a programmer, to very easily
make custom content/objects it will greatly aid
in the adoption of the software. Again, contrasting
Revit and Architectural Desktop; our firm never
bought ADT due to the amount of renovating and
interiors work we do. With ADT setting up custom
objects is complex, and requires specialized skills.
Every job we do would have required a great deal
of custom objects. With ADT all the time it would
have took to make all the custom objects required
by a job would have been significantly more than
the amount of time it would have saved with it's
automated features. Heck it might have been even
longer than it would have been to simply draw the
thing in plan-jane AutoCAD. Maybe if we had been
doing the same type of building over and over
it would have paid off; but otherwise it looked
like ADT wasn't a good choice for us, no matter
how many automated features it had. But because
it's trivial to make custom content within Revit,
it makes it viable to generate project specific
custom content, and so those automated features
available to a parametric system can save us time.
Sorry to blabber on so, and to talk about Revit
so much. It's just that Revit is the first program
to come along that really fits well with a small
office; however it's limited. It's got the right
'idea' as it were, but has other problems that
could be better addressed with a open source system.=20
Jeffrey
Other related posts:
