<html> <body> Hi Folks -- The guts of the Croquet scheme are quite independent of Squeak, and were thought of as "a higher layer of TCP/IP". Also, the "Islands" were deliberately set up to be quite independent of what is inside the Island (which is a kind of abstraction for a virtual-machine-address-space that has objects). The idea here is that good things can happen if you really do "real messaging" and if you include time (pseudotime) specifically in your computing model. As some of the folks on this list have realized, somewhat careful organization of a publish/subscribe facility can be the lingua franca for the Croquet coordination ideas. As you can see from the website, Croquet was done to explore what I thought was a terrific PhD thesis (of Dave Reed ca. 1978 at MIT) but that never got implemented. Dave Reed is across the street at the Media Lab, is friendly and a fun guy to talk to, and is one of the best systems thinkers I've met over the years. Cheers to all, Alan At 12:27 PM 11/27/2006, Ian Bicking wrote: <blockquote type=cite class=cite cite="">Ken Ritchie wrote: <blockquote type=cite class=cite cite=""> <a href="http://www.opencroquet.org/" eudora="autourl"> www.opencroquet.org</a> (especially the TeaTime components)</blockquote> I just read the "Croquet System Overview" section of this: <a href="http://www.opencroquet.org/Site%20PDFs/Croquet%20Programming%201.0B.pdf" eudora="autourl"> http://www.opencroquet.org/Site%20PDFs/Croquet%20Programming%201.0B.pdf</a> It was a very nice overview of the architecture of the system.  I like the approach a lot; the description of the architecture makes me want to bang out code *right now*, which is always a good sign.  I'll try to resist actually doing so. Whether this should be reimplemented in Python or implemented in a language-neutral way, I'm not sure.  I can kind of imagine the Router being a service accessible over dbus, but I'm not really sure what that would accomplish.  The dbus message format is also possibly something to use (since Croquet messages, I assume, are tied to Smalltalk).  But I don't know if that even matters -- there's nothing here that really facilitates inter-language communication, as it assumes that all Islands (aka objects) use exactly the same code. I also wonder if there's room for more sloppy communication.  E.g., situations where out-of-order message execution is preferable to blocking.  If it damages the integrity of a simulation or 3D world, it might be preferable to just block.  OTOH, I think there are other kinds of collaboration where responsiveness may be more important than complete integrity.  It perhaps depends in part on how good the network connections are.  What will collaboration feel like over several hops on a mesh?  What about over a satellite internet connection?  I have no idea how this will effect the experience.  And perhaps good message design can help with this anyway.  For instance, if you are editing text you don't necessarily want to send a message for every keystroke; the UI can batch things up and resolve conflicts, even if the underlying objects are less forgiving.  So maybe I'm imagining the problem. -- Ian Bicking | ianb@colorstudy.com | <a href="http://blog.ianbicking.org/" eudora="autourl"> http://blog.ianbicking.org</a> _______________________________________________ Sugar mailing list Sugar@laptop.org <a href="http://mailman.laptop.org/mailman/listinfo/sugar" eudora="autourl"> http://mailman.laptop.org/mailman/listinfo/sugar</a> </blockquote></body> </html>