I don't want to jump into the discussion whether or not an Opal/CoreGraphics based solution is the one and only way forward for GNUstep. Nor whether GNUstep should support the xlib backend indefinitely.
I rather would like to see a more pragmatic way forward. Why not add another backend "cg" to GNUstep back. This will be a very thin layer on top of Opal and will allow everybody a smooth transition from whatever backend they are using at the moment. As soon as this backend has matured enough we are ready for the above discussion. And at that point I will be a strong proponent of the position that Opal is the way to go. Of course until then some of the capabilities that the direct usage of Opal would give to gui will be lost. This should give all of us a good enough reason to try to keep this intermediate period as short as possible. The way ahead for me looks like this: - Sort out any outstanding licence/Copyright issues - Move Opal code into the GNustep SVN repository - decide on the make and dependency framework to use - Clean up to code and make files - Add what ever is needed for a full backend functionality (*) - write the new back component "cg" based on Opal - test, test, test - start discussion about making Opal the default drawing mechanism (*) I already found a few points for this item, but we should discuss them, when we are at this point. _______________________________________________ Discuss-gnustep mailing list [email protected] http://lists.gnu.org/mailman/listinfo/discuss-gnustep
