Live Distributed Objets - Why not in Smalltalk ?

Previous Topic
classic Classic list List threaded Threaded
2 messages Options
Reply | Threaded
Open this post in threaded view

Live Distributed Objets - Why not in Smalltalk ?

Hi people  (crosspost pharo and research)

While googling for live objects, ie. an API to query live objects
coming from the outside world (trade stuffs, information, or whatever
information coming from a sensor etc...), I think I just found what I
was lokking for. At least something exists. I think this is different
from Croquet or Spoon (I think live objets don't need to be in another
image, they are just independant entities).

On this page [1], you'll find video demonstrating live objects (2nd
part of the first one is really cool, especially when they drop the
coordinate of a plane in the window...or when they inspect live
objets). 2nd video id about integration in office tools. On this same
page, there are several publications at then end. [2] and [3] are
wikipedia references.

Were you aware of that ?
Could we build a platform in Smalltalk (I don't really need it, but I
find this would be really cool).

I don't need such a platform, this was just a midnight googling
session... but hey it's coool :)

Quote from wikipedia (as you can see Smalltalk inspired them ;) :
"Originally, the term was used to refer to the types of dynamic,
interactive Web content that is not hosted on servers in data centers,
but rather stored on the end-user's client computers, and internally
powered by instances of reliable multicast protocols. The word live
expressed the fact that the displayed information is dynamic,
interactive, and represents current, fresh, live content that reflects
recent updates made by the users (as opposed to static, read-only, and
archival content that has been pre-assembled). The word distributed
expressed the fact that the information is not hosted, stored at a
server in a data center, but rather, it is replicated among the
end-user computers, and updated in a peer-to-peer fashion through a
stream of multicast messages that may be produced directly by the
end-users consuming the content; a more comprehensive discussion of
the live object concept in the context of Web development can be found
in Krzysztof Ostrowski's Ph.D. dissertation[3].

The more general definition presented above has been first proposed in
2008, in a paper published at the ECOOP conference[8]. The extension
of the term has been motivated by the need to model live objects as
compositions of other objects; in this sense, the concept has been
inspired by ***Smalltalk***, which pioneered the uniform perspective
that everything is an object, and Jini, which pioneered the idea that
services are objects. When applied to live distributed objects, the
perspective dictates that their constituent parts, which includes
instances of distributed multi-party protocols used internally to
replicate state, should also be modeled as live distributed objects.
The need for uniformity implies that the definition of a live
distributed object must unify concepts such as live Web content,
message streams, and instances of distributed multi-party protocols."


Reply | Threaded
Open this post in threaded view

Re: Live Distributed Objets - Why not in Smalltalk ?

Ralph Johnson
Very cool.  Ken Berman is very smart, and has been doing distributed
programming for a long time.

When I say "live object", I usually mean the fact that Smalltalk lets
us change the code of an object while it is running.  it is obviously
a different meaning.  Ken Berman is thinking about large scale
systems, about how objects can be combined and can interact over the
internet.  Smalltalk focuses on a single image, where all the objects
run on a single computer.  In Distributed Live Objects, the code the
runs an object doesn't have to be object-oriented.  Ken is not trying
to build a system that is objects all the way down; he is trying to
build one that is objects all the way up.

-Ralph Johnson