[Qgis-psc] Proposal that QGIS.ORG will become a member of the OpenDesign Alliance

[Andreas Neumann]

Hi,
> I think the main point in Vincet mail is about being perfectly
> legit to distribute binary QGIS versions linked against  LibreDWG
> as it'd make a GPL3 boundle.
>
> What's missing from LibreDWG that makes it a non-interesting solution
> at this moment ? How much would it cost to enhance it to add those
> missing parts ?
Currently there is no one actively developing, funding or using LibreDWG 
(or the fork LibDWG).

I don't know what it would cost to bring LibDWG into a usable level 
(probably at least a mid-to high level 6-digit Euro figure (just a wild 
guess)). I also don't know, if there would be developers available to 
work on it. For obvious reasons it is not very popular amongst Open 
Source developers to work on reverse-engineering a closed-source format.

The projects LibreCAD, FreeCAD, Blender, Inkscape would like to use 
LibreDWG or LibDWG. They are all GPLv2. They are not GPLv2 or higher 
like QGIS. So QGIS is in a slightly better situation compared to the 
other libre graphics projects.

If QGIS.ORG (or other QGIS users) would invest into LibreDWG/LibDWG, 
they have to understand that they are the only funding source of the 
library - which is a very high-risk I would say. They would also be the 
only professional user of the library - which is also very risky.

In summary, I am not against using LibreDWG - but I think the risk to 
invest into it and use it - as the only project is really high. It is 
also a (financial) burden for the future as we would have to make sure 
that it still works with future DWG file format versions. Again, if I 
had 500k Euros at hand, it would be a different discussion - but it 
would still be risky.

If we want to seriously push a Teigha alternative, I would team up with 
LibreGraphics projects like the above mentioned projects and make sure 
the license is GPL v2+. This would allow us to spread risk and financial 
burden. But it would also mean that we won't have CAD support in QGIS in 
2016 - as I assume it will take much longer to come up with an 
alternative to Teigha, which was actively and continuously developed for 
many years, by several developers.

Andreas