Re: Re: About Dr. Geo Developement

Idealy, I would like DrGeo to not use decimal number
by algebraic (is the correct word in English) ones.

With such numbers we will get property checking as the one
described in point 1 of Andrea list for free in an *accurate*
way.

Beside that I don't know much about projective geometry, 
so I cannot tell how hard it could be to integrate it in
Dr. Geo. Do you have any good references I can read about it?

Will it help for 3D geometry?


Any way I really want we work on the TODO list, please start
to fell the TODO wiki list at http://www.ofset.org/wiki?DrGeoTodo

Then I will fill the bug tracker with priority order.


Hilaire


Quoting Marcel Birthelmer <marcel-aE3gVXIT7zoqDJ6do+/[email protected]>:

> Looking at these issues, it seems like a lot of them
> could be generally 
> addressed, along with one of my hopes (the integration
> of "conic" 
> objects), by using a projective system rather than an
> affine one. that 
> would get rid of a lot of case checking, and it would
> make constructions 
> that rely on parallel points "meeting at infinity"
> possible. I don't 
> know the code structure of Dr. Geo well enough to know
> if this is 
> possible, but I think it would be a definite advantage.
> It also seems 
> like a lot of things would be easier in terms of the
> math involved, by 
> simply using matrix multiplications for
> transformations.
> Marcel
> 
> 
> Hilaire Fernandes wrote:
> > Hello Andrea, 
> > 
> > Thanks for your feedbacks. You feature wish list, make
> me think about a
> > 1.0.0 release of DrGeo before we move to implementing
> new stuff which
> > could move DrGeo in unstable state.
> > 
> > Having a TODO list is indeed good, and the DrGeo one
> is truly non sense
> > and it should be clean up, just look at
> http://www.ofset.org/drgeo/todo.html. 
> > 
> > However the TODO file in the source distribution is in
> a better shape.
> > 
> > Therefore I would like to take the opportunity of your
> message to work
> > on that TODO list first and see how the stuff could be
> implemented (for
> > example your point 1. I see three differents way of
> doing that, some
> > ugly, some other much better but harder)
> > 
> > Of course writing a TODO list will not make the stuff
> to appear like
> > magic, it will depends on a how much help we can have
> in that. So the
> > TODO list should be shapped in priority order.
> > 
> > I suggest we take on the TODO list I published several
> week ago at
> > http://www.ofset.org/wiki?DrGeoTodo
> > 
> > I will clean up this TODO list, then I will let you
> add stuff into it.
> > 
> > Let's start!
> > 
> > Hilaire
> > 
> > 
> > On Mon, 12 Apr 2004 17:59:08 +0000
> > Andrea Centomo <[email protected]> wrote:
> > 
> >   > Hello all
> >   > 
> >   >              I'd like to expose some reflections
> to help the future 
> >   > developement of Dr. Geo. These reflections arose
> also from 
> >   > the use of the cvs version of Kig. As a general
> consideration
> >   > I think that Kig is more suitable than Dr. Geo for
> the study of
> >   > cartesian geometry, especially conics and cubics.
> >   > At the same time I think that for geometry without
> coordinates
> >   > and for programming Dr. Geo is the best.
> >   > 
> >   > 1. like in Kig it could be useful to have in Dr.
> Geo the possibility
> >   >     of checking if three points are collinear, if
> two lines are perpendicular
> >   >     or parallel, if a point belongs to an object.
> More generally 
> >   >     it could be good to improve a "proof
> apparatus";
> >   > 
> >   > 2. moreover, as just pointed out, it could be
> useful to have
> >   >     a fill up function to fill with colour a
> geometrical region 
> >   >     of a figure;
> >   > 
> >   > 3. also leaving to Kig analitic geometry it could
> be useful 
> >   >     to improve the structure of the coordinate
> grid adding 
> >   >     numbers and orientation;
> >   > 
> >   > 4. I have no precise ideas but it could be good to
> think
> >   >     of stuffs for 3D geometry: is it a nonsense?
> >   > 
> >   > 5. it could be good to have the possibility of
> printing directly a figure.
> >   > 
> >   > Amitiés
> >   > Andrea
> >   > 
> >   > 
> >   >
> ----------------------------------------------------------
> >   > 
> >   > Andrea Centomo
> >   > email: [email protected]
> >   > 
> >   > 
> 


-------------------------------------------------------
This SF.Net email is sponsored by: IBM Linux Tutorials
Free Linux tutorial presented by Daniel Robbins, President and CEO of
GenToo technologies. Learn everything from fundamentals to system
administration.http://ads.osdn.com/?ad_id=1470&alloc_id=3638&op=click