To: dist-obj@cs
> "The Infospheres projects does research on the composition of
> distributed active mobile objects that communicate using messages."
We have now added this to the title of our Web page,
http://www.infospheres.caltech.edu/
> note that, explicitly, this statement says nothing about the whole
> object == documents hypothesis of some group members.
That is correct. I believe the only one in our group who maintains that
(objects == documents) is me, and I only do this because I'm such a big
fan of the Web. I'm willing to concede that these words are just words,
and that semantically we are talking about the same thing...
"I don't know what you mean by `glory,'" Alice said.
Humpty Dumpty smiled contemptuously. "Of course you don't --
till I tell you. I meant `there's a nice knock-down argument
for you!'"
"But glory doesn't mean `a nice knock-down argument,'" Alice
objected.
"When I use a word," Humpty Dumpty said, in a rather scornful
tone, "it means just what I choose it to mean -- neither more
nor less."
"The question is," said Alice, "whether you can make words mean
so many different things."
"The question is," said Humpty Dumpty, "which is to be master--
that's all."
-- Lewis Carrol, "Through the Looking Glass"
So I say documents, you say objects, we mean the same thing, let's move on.
Let's also put to bed two more myths:
1. Myth One: Infospheres == Munchkins.
Munchkins are a Rohit-ism dating back to 1994, evolving out
of some ideas that he had about a gedankenexperiment using
active messages in small ubiquitous two-way communication
devices. You can read about what munchkins are at
http://www.cs.caltech.edu/~adam/LOCAL/faq-fork.html#munchkins
If you require more explanation than that, tell me and I
will elaborate on any of the points therein.
On the other hand, you can read about Infospheres at
http://www.infospheres.caltech.edu/papers/framework/
Infospheres are a Caltech project dating back to 1995,
working toward the idea of persistent message-passing objects.
To that end, Infospheres has developed an experimental
system in Java which just went 1.0 beta 3 release a week ago:
http://www.infospheres.caltech.edu/releases/
I still have to respond to Ron's post about Infosphere sessions,
which are groups of these persistent message-passing
objects who collectively work toward some specific task.
I'll do that in a separate post.
2. Myth Two: Infospheres == *TP.
Like Munchkins, *TP is a Rohit-ism dating back to 1994, when
he realized that different application layer transport protocols
(SMTP, NNTP, HTTP, etc) all have the same basic functionality
at an abstract layer. Then, each mechanism adds features
specific to the particular data being sent (flood fill for
news, on demand for web, etc).
By building the most basic underlying application-level
transport layer and then layering each of the current
mechanisms on top of that layer, *TP unifies data exchange.
Furthermore, PEP can be used to extend *TP with new plug-in
features such as security, etc. A system that implements
*TP then could have a little checkbox for things like caching,
so an application could customize its transportation of data.
This allows users to have more control of the specific tasks
they have in mind (e.g., "Tape Seinfeld for me every day."),
as opposed to today where we have to specify every particular
detail of that task (e.g., look up the channel and time that
Seinfeld comes on every day and then program the VCR accordingly).
If you require more explanation than that, tell me and I
will elaborate on any of the points therein.
Note that such a system could be built atop the Infospheres
layer, but such a task is not really the focus of anyone in the
Infospheres group. I conjecture that Rohit will want to
build such a system when he matriculates at Irvine next year.
So, to reiterate, Infospheres is a completely separate project
from Munchkins and *TP. Sorry to anyone who might have been confused
about this.
----
adam@cs.caltech.edu
the issue i have with this is that, for the most part, most products of
engineering in our field are written by the 97% and thus, no matter how
open and beautiful and wonderful the framework/bus/system/whatever is,
they'll screw it up. these wonderful plug-and-play, self-aware gizmos
will be equivalent to little blind, deaf, and dumb mechanical
cockroaches that run up to each other, touch antennae, and vomit.
-- Joe Kiniry