Thanks Marcus. I’ve read the paper and now I’m REALLY excited! YAY for Pharo :)
On 03.07.2014, at 11:01, Marcus Denker <[email protected]> wrote: > > On 03 Jul 2014, at 10:52, Max Leske <[email protected]> wrote: > >> >> On 03.07.2014, at 10:21, Marcus Denker <[email protected]> wrote: >> >>> Now classes with special Slots are shown with a special class definition >>> (this is just a stand-in, not >>> final… e.g. we need to add first class globals and think about what to do >>> with #category… so we >>> will see. For now please treat this as just a stand-in for playing with >>> things. >>> >>> >>> <slots.png> >>> >>> The idea is that we need to actually use it and then see what to change in >>> the next iteration. >>> >>> And it should stay always compatible: if you do not use special slots, it >>> is compatible, if you do, >>> you get more power at the price of not being portable to other systems. But >>> even in this case, >>> the slots do give value: e.g. we will be able to add very very easily >>> breakpoints or reflectively >>> change behaviour for a slot (without chaining its definition). This will be >>> very powerful. >>> >>> Marcus >> >> >> Sounds exciting! But I don’t quite get what ‘special slots’ are and what >> they can do. Is there any resource I can take a look at to get a better >> understanding? >> > > The paper has some examples: > http://rmod.lille.inria.fr/archives/papers/Verw11a-OOSPLA11-FlexibleObjectLayouts.pdf > > Another direction that is interesting is Magritte: instead of meta-describing > instance variables with some convention of class > side methods, you could just define Slots that have all the Magritte style > meta data. > > Marcus
