Den 23.02.2010 13:00, skrev Göran Krampe: > Hi! > > Torsten Bergmann wrote: > >>> So you are educating them.... >>> Frankly I stopped trying long time ago. >>> >> No, I just tell them what I think is necessary to move forward. >> Other may be more educated here, it is just sharing MHO so others >> can to form an own opinion. >> > Right, discussion is good. And believe me - the squeak-dev community is > NOT oblivious to the needs of a package system. I mean, come on, > Monticello, SqueakMap, Universes etc etc - it all was around before > Pharo was born. > > What I have seen so far in Metacello is indeed nothing new - the > senarios and discussions about dependencies etc go loooong way back. > > BUT... that *doesn't* mean I am passing any judgment on Metacello - IMHO > all efforts are good per definition! :) I haven't formed an opinion myself. > > No system is going to perfectly fulfill everyones perception of what would be "The Best Way"(tm), but as long as the need for functionality provided by a system is real, actually deciding on one, IS important. >From what I've seen in archive discussion, that seems to have been the biggest showstopper for noticeable progress in the Squeak community. Paradoxically, existing systems which clearly contain lousy code, are left in (or replacements objected to) on the basis of backwards-compatability... IMHO, this is the biggest difference between Squeak and Pharo I've seen. Sure, you can now unload Etoys from Squeak, your image will still contain code like ImageSegment>>findRogueRootsAllMorphs: though. Sure, you can new add Preferences with pragmas now, you still have a class with a Dictionary accessed and stored into by the entire system, and 1223 accessors for things local to specific packages though.
Metacello is nothing new for sure, but contrary to other systems, it seems to have reached wide acceptance in Pharo. Understandably, many here feel it'd be nice if the "classic" Squeak-community decided to use it as well. Personally, I doubt it will happen (for any PMS really, as evidenced by Sake/Packages, as well as all the other previous attempts), yet I appreciate Torsten's and others effort to at least correct the plain-out misconceptions some people seem to have about deficiencies of Metacello. Pretty sure noone would argue Monticello is a perfect system either, achieving critical mass was though. >> Meanwhile the Pharo community has grown with new members and many people >> dont care about old discussions. >> >> Let's use our energy to move forward and (if possible) share as much >> knowledge and code as long as parts of our communities overlap. >> > Yes, exactly. > +1. > Anyway, "common exchange mechanism" - I am working on Deltastreams which > I even presented at Brest, though unfortunately colliding with the > Seaside tutorial. Deltas and Deltastreams are indeed focused on EXACTLY > this problem (fork interchange), and it was born in 2007: > > http://lists.squeakfoundation.org/pipermail/squeak-dev/2007-August/119471.html > > I and Igor are the ones working on it - and after Brest we have had very > little time to move it forward. I still think it is a very important > piece of the puzzle though. > > regards, Göran I've been meaning to make system-change notifications for Pharo so the recording-part could work there too, but I never seem to get around to it :( Part of the problem is of course my own personal "The Best Way"(tm), which f.ex. would include Announcements. Since I'd like the ability to have Weak subscriptions, I'd first have to deoptimize the Announcement-package Levente made a bit to make it "prettier", make it a strict superset of what if found in Core already, fix WeakRegistry so each resource registered for an object actually gets finalized, which again means... etc... Oh, and hope it'd end up good enough that it'd warrant inclusion:) Cheers, Henry _______________________________________________ Pharo-project mailing list [email protected] http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project
