There was always supposed to be an additional low-level documentation of CLIM. It never happened. The first versions of CLIM (1.0) also were never as extensible and never had some of these layers (Silica). Much of the extensibility and the UI adaption have been added for CLIM 2.0. Xerox had worked on ist own version of the CLIM implementation and developed the Silica layer. Silica then was added for the joint CLIM 2.0 implementation. UI adaption also was added with CLIM 2.0. If you look at the comparable implementation of DW (Dynamic Windows), the father of CLIM, it was better documented, had a better look&feel, and had also more features in several areas. DW's documentation is about twice as large. CLIM was supposed to be a simplification, implemented with CLOS (instead of New Flavors) and it should have had better performance. Though my feeling is that DW was faster and had less bugs. What does that mean for McCLIM? There is only little documentation, tutorials etc.. So it needs to be written over time. Idea: integrate some wiki mode into Climacs and write the documentation there.
Von: Paolo Amoroso <amoroso@mclink.it> Organisation: Paolo Amoroso - Milano, ITALY Datum: Thu, 30 Jun 2005 11:30:21 +0200 An: McCLIM <mcclim-devel@common-lisp.net> Betreff: Re: [mcclim-devel] incremental redisplay and with-first-quadrant-coordinates
Robert Strandh <strandh@labri.fr> writes:
I see you point, but the layered design of CLIM makes it hard to distinguish between an advanced application programmer and a CLIM implementor. That is probably also why CLIM user manuals tend to turn into the spec after just a few chapters.
Or maybe because Lisp vendor resources for producing additional CLIM documentation or instructional material have always been limited.
Paolo -- Lisp Propulsion Laboratory log - http://www.paoloamoroso.it/log _______________________________________________ mcclim-devel mailing list mcclim-devel@common-lisp.net http://common-lisp.net/cgi-bin/mailman/listinfo/mcclim-devel