[GRASS5] opendwg as _last_ resort

[Bernhard Reiter]

This is the mail about the reasons 
why to look hard for alternatives to opendwg.

On Fri, May 16, 2003 at 10:29:13AM +0200, Radim Blazek wrote:
> On Thursday 15 May 2003 04:23 pm, Bernhard Reiter wrote:

> I tried everything. Last year, I wrote also import
> program based on DIME. I was going exactly the way you suggest.
> When everything other failed, I took Open DWG as last option.

Radim,
I'm convinced that you checked all the alternatives, 
before you decided to use opendwg as last resort.
This is the right way to work.

I'm not opposed to use opendwg as last resort.
Part of my writing tried to explain how to do this in a good way.
I might make this subject of another email thread.

While I agree with you that opendwg as a last resort
which solves problems now is hard to avoid,

I maintain the notition that we need to start making plans 
to get rid of the problem that the this proprietory license
still poses for us. Even if we cannot finish an dxf library
we at least have to start to work on it. This is very important.

In the history of the GNU project (that brought us the now
well known GNU/Linux system everybody uses) similiar situations
have often occurred: The majority basically said to a technical problem:
"Hey you can never make it, that's useless efforts."
The GNU project tried anyway and the first attempts 
sometimes were lacking badly technically, but came with full freedom.
Next line of the critics were: "Nobody is going to use this
crippled stuff; nobody is going to help you."
Again they were wrong, if might have taken a lot of time,
but in the end the result and gained was revolutionary.

> > As an development direction, maybe we should focus our work to
> > integrate it into GDAL, though the win for Free Software would even
> > be higher.
> 
> Frank Warmerdam (Mon, 25 Nov 2002 08:58:04 -0500):
> "It is my general intention to someday add DXF/DWG support to OGR based based
> on the OpenDWG libraries. ....."
> (http://remotesensing.org/pipermail/gdal-dev/2002-November/000008.html)
> 
> Frank Warmerdam (Sun Feb 23rd, 2003 at 02:40:47 AM):
> ".... While I would prefer a true open source license, the licensing terms 
> of the OpenDWG alliance is quite reasonable. ..."
> (http://www.digitalearth.org/comments/2003/2/1/16238/10919/1/post)
> 
> (Frank Warmardam is GDAL/OGR developer)

I was aware of Frank's statements before.
In some ways he also looks at opendwg as a "last" resort
and would prefer Free Software. 
The strategical situation of GDAL is a bit different
than the one of GRASS. GDAL is meant to be an interlink piece.
In some situation it makes sense from the strategical Free Software
point of view to connect to non-free software in order to 
progress the adoption and development of Free Software.
In short when this leads to more freedom in the end.

GRASS is much more than an interlinking piece. It is a full suite.
The incentive to keep it together is higher as it provides
a full set of functionalities in itself. 
This is not entirely easy to explain. 
GRASS is an alternative in some ways to some proprietory CAD
software (for fun called SPCS),
we want more people to use GRASS and not SPCS.
If we make it very easy to exchange data with SPCS
in its proprietory format, there is a higher motivation
to continue to use SPCS. Some people even might use SPCS
for the tasks that GRASS cannot do.
In short there is less potential win GRASS users,
but even a bit more to loose them.

> > What do you think Radim?
> 
> We have to do our projects now. 
> We cannot wait 10 years until somebody 
> develop GPL DWG library. 

As written above, this is one argument, 
but not mid term strategy.

> You cannot tell people who sent you data:
> "We are using GRASS it is free SW, so we cannot use DWG you sent me. 
> Could you convert those files to DXF v.10? It is easy, open each
> of those 100 files in ACAD and save it as DXF v.10 and make a new CD. 
> It cannot take you more than half day."
> 
> You can of course, but then you are not in next project.

But we have to do it as often as we can.
And we have to make sure that it was _their_ mistake 
to save in a proprietory dataformat which now holds their data hostage.
The situation is a bit similiar to the word format.
Also most word users could just export data as rtf or HTML.
Pointing this out to people is important:
http://www.gnu.org/philosophy/no-word-attachments.html
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
Url : http://lists.osgeo.org/pipermail/grass-dev/attachments/20030516/56ce6347/attachment.bin