Sorry syx people for the double post, I wanted to include help-smalltalk.
Vincent Geddes <[hidden email]> wrote: > Why don't the three of us all support each other in creating a GNOME/ > GTK+ IDE and community class library? I think it would be a great > folly for each of us to re-invent the wheel. > > I think its quite clear (to me at least) that merging VM code is not > going to happen. I am very attached to Panda, and Luca to Syx. So lets > at least rectify the situation above the VM level. I think this is > very doable, in the current technical and legal climate. > > I suggest creating a "GNOME Smalltalk" commons project which will > consist of bindings for the entire GNOME stack, as well as a fancy > development environment. We could put all sorts of other goodies in > this project as well. The bindings FFI/API should more or less follow > the GST's implementation (as it is the most advanced). The code would > be kept in a neutral repository, under a license suitable to all three > parties. I hope GST's copyright assignment still allows for retaining > copyright for use in non-GST contexts. The code could possibly be > licensed under the LGPL to allow crossbreeding with the strong GST > libraries. > > There is this new tool in GNOME called gobject-introspection. Its > primary aim is help create language bindings. For a long time in > GNOME, language bindings were third-class citizens, as it was very > hard and tedious to wrap GObjects. However, gobject-introspection > makes things easier by allowing runtime introspection of GObjects, > amongst other things. For instance, Colin Walters did a cool hack in > which he generated GTK+ Java bindings at *run-time* using gobject- > introspection (http://cgwalters.livejournal.com/19537.html). > > For GNOME Smalltalk to work, we will need to agree on the subset of > our class libraries that need to be compatible with each other. I > don't think this should be too hard. > > As I mentioned to Paolo, I am very interested in Vassili Bykov's > Hopscotch IDE (http://gbracha.blogspot.com/). From what I have read of > it, their rationale and usability principles seem to be very sound. > With that in mind, I am certainly not keen on doing a straight port of > the Squeak browsers to GTK+. I get irritated by the modality of the > Squeak class browser. For instance, If I am editing a method, and want > to quickly view another method, I have to cancel editing. Only being > able to view one method at a time is also restrictive. Someone very > aptly called it a "pinhole" style of browsing. > > PS: I called this hypothetical project "GNOME Smalltalk", but we can > improve on the name if needed. Paolo _______________________________________________ help-smalltalk mailing list [hidden email] http://lists.gnu.org/mailman/listinfo/help-smalltalk |
JFYI
I am very interested but don't have time to work on it right now. Please let us know if a separate resource would be created for GObject/GTK/Gnome. Best regards, Ildar -- Ildar Mulyukov, free SW designer/programmer ================================================ email: [hidden email] home: http://tuganger.narod.ru/ ALT Linux Sisyphus ================================================ _______________________________________________ help-smalltalk mailing list [hidden email] http://lists.gnu.org/mailman/listinfo/help-smalltalk |
In reply to this post by Paolo Bonzini-2
> Still, if there is standardisation at the Kernel and Class library
> level, then there really shouldn't be a problem having multiple VMs. Fully agreed. > > 1) Syx and Panda have no clue (not yet) of what a pollDictionary, a namespace, > > security policies, syntax, packages and concurrency are GNU Smalltalk supports the fileout syntax, we can use it. Packages and namespaces are not necessary features for this project (namespaces are visible in a GUI, but building a good GUI framework like OmniBrowser but for Hopscotch-like GUIs is what this project could be about). Security policies are not functional right now, they are there only because they were part of my master thesis. > > 2) Since the VM is different, some of the choices in the smalltalk code could bring problem for > > scaling and performances This has not been a problem for Seaside, which has much heavier requirements on scaling. > > Then for GNOME Smalltalk, in the far future (at least for Syx): > > 1) I'd make a GSIDE (GTK+ Smalltalk IDE) then add features for GNOME when you're on > > GNOME, as you all know I'm very careful about portability. Yeah, GNOME is just a desktop environment. I'm sure that 99% of the features would only need GTK+. Paolo _______________________________________________ help-smalltalk mailing list [hidden email] http://lists.gnu.org/mailman/listinfo/help-smalltalk |
Free forum by Nabble | Edit this page |