Contact Consortium: Oworld List Digest, early posts

The Contact Consortium
Dialogue on the list since the J Eric Mason proposal (includes introductions)

Mr. Damer;

I am writing to put a bug in your ear.
The Web3D consortium is moving their next-generation standard (X3D) to
target:
( from http://www.vrml.org/news/x3d/mrd09.html )
1. Advertising/Page Animation
2. E-commerce Product Visualization
3. Database Visualization
4. Augmented News/Documentary
The raises the possibility that multi-user functionality will continue
to be unaddressed in the near future by Web3D despite its Living Worlds
group, the DIS-Java-VRML and the new VRTP group. In fact, I have read
comments from the X3D development team that specifically distance X3D
from this market. As an active member of the Living Worlds discussion,
this does not give me a warm feeling about progress.
How do you feel about the CCON creating a technical entity to encourage
and support MU3D/VR browser interaction and data/wire format standards,
and possibly even content standards such as Living Worlds? How would
your members feel about it?
I admire what the Contact Consortium has accomplished for furthering
interest in multi-user 3D/VR communication. And even though I am
interested in getting my company, VR Telecom, on board as well, in this
communication I speak for myself.
Others I have spoken to, including Bernie Roehl (cc'd on this message)
also see a need for a champion of MU3D standards, and don't feel Web3D
is going that direction. Is CCON the right place?
Thanks for your kind consideration.
JEM

Bruce writes:
> You can find the signup page for the group at:
> http://www.ccon.org/lists/
Done.
> All that said, I think I can speak for the organization that I would be
> happy to have us get behind a merged effort to forward the cause of
> multi-user virtual worlds/standards.
Then I think we have a good starting point.
> For years I have been vocal (and largely unrecognized) in my views that
> VRML was being handled wrongly.. aiming at technical specs rather than what
> users wanted
That may well be true. However, we *need* technical specs, and my guess is
that Eric and I (and many others, I suspect) are interested in helping craft
them. Interoperability depends on them, and interoperability is a key goal.
> the fact that multi user was never treated as a priority was a big problem
> for VRML and potentially its fatal mistake.
I would tend to agree. The VRML approach was to start with static scene
descriptions, then add behavior, and finally multi-user. I now feel that
such an approach is fundamentally flawed -- you can't bolt on multi-user
as an afterthought, it has to be designed in from day one.
> I am so sorry all those millions were spent designing VRML for and by
> engineers working on fat pipes.
Well, at least it wasn't *my* money. :-)
Seriously, though, I feel we gained a lot from the entire VRML process.
Some of what we did is still quite useful, and for everything that didn't
work, we learned something new.
We now know a lot more about what works and what doesn't, both in terms
of technical specifications and in terms of process.
> Well the most sensible thing would be to air this to what remains of the
> VRML community and the Web3D board.
My feeling is that it would be like spitting into the wind.
I see no point in creating ill will where none exists. I do see a point
in creating an atmosphere where we can get to where we need to be, without
waiting around for anyone else.
It's clear that multi-user isn't even on the X3D radar. Their focus is
squarely on things like banner ads and e-commerce. That's fine -- somebody
has to do that work, and better them than me.
I feel we should let the X3D folks do their thing, wish them well,
use whatever they come up with if it proves to be useful to us, ignore
it if it doesn't, and in the meantime move on to our real goal: building
Cyberspace.
--
Bernie Roehl
University of Waterloo Dept of Electrical and Computer Engineering
Mail: broehl@ece.uwaterloo.ca Voice: (519) 888-4567 x 2607 [work]
URL: http://ece.uwaterloo.ca/~broehl
Aloha love, having a good time here in honolulu with sue and bruce (when i am not online). I thought you wuld be interested in the below email i just wrote in response to a possible major schism in the vrml community (all the multi user community focus for vrml is being lost it seems) make for interesting reading..
all set for terences..
bruce

JEM,
Boy talk about serendipity! In fact the Consortium already has a special interest group called "OWorld" which started at our Biota2 conference last summer. This group was specifically set up to brainstorm technical approaches, application domain targets and community building for an open source multi user virtual worlds platform (or platforms). We have about 20 members on the list now and are seeking to not only expand it but to find a champion who will stoke that list and the SIG. You can find the signup page for the group at:
http://www.ccon.org/lists/
As you may know, the Contact Consortium has four years of practical experience in inhabited virtual worlds. We are "platform neutral" seeking to forward the development of any visual shared space on the net. We use any worlds that work for specific purposes. We are also nonprofit and based on a strong volunteer community, with a focus on users and their applications, rather than technologies and companies. We have a wealth of knowledge about what works and doesn't and what people really like to do in these spaces.
All that said, I think I can speak for the organization that I would be happy to have us get behind a merged effort to forward the cause of multi-user virtual worlds/standards. I also have been following the x3d discussions and have a variety of personal and technical opinions on it, which if you allow me, I will elaborate on here.
For years I have been vocal (and largely unrecognized) in my views that VRML was being handled wrongly.. aiming at technical specs rather than what users wanted and what environments to craft that would make it easy for ordinary folks to make 3D spaces. All those 3D modeling tools, however wonderful, will never appeal to a large content community or be able to be used by the ordinary netizen. In addition, and sharing your frustration, the fact that multi user was never treated as a priority was a big problem for VRML and potentially its fatal mistake. Building 3D is fundamentally a social activity (in the real world) and VRML desperately needed a multi user virtual commons where people could come in and kick the tires of new objects or properties, carrying out their development at the level of a real usable visual space. The VRML mailing list approach was a disaster and brought the level of discussion down from the experience of 3D to "text only" code talk and politics. The few multi user vrml spaces made were either not open to the community for development or were efforts too small to drive the development of VRML.
I am so sorry all those millions were spent designing VRML for and by engineers working on fat pipes. If only a fraction had been allocated to create a real cyberspace version of the VRML community tuned for dial-up net use, we would have something substantial today, not just islands of innovation, a file interchange format, and a couple of huge overdesigned plug ins. On a personal level I have spent years talking about the Active Worlds project (of which i am not any formal part, just a user) which with a very small investment and little public visibility, built a virtual commons, a strong community and a whole practice of user-built extension and design of worlds people find valuable. In fact, their streaming object format which I have demoed repeatedly at VRML conferences, now seems to be "rediscovered" with X3D. I find all this sad and ridiculous (my opinion, not necessarily that of the Contact Consortium). Did you know that AW citizens have built nearly one billion polygons of 3D content, all of which is still visitable (even the spaces built 3 years ago)? There were no versions to make old content incompatible, instead they employed serial upgrading, in-world social development of the technology and few people know or care about the file format, they build directly inside the world with the Lego parts contributed by a relatively few expert modelers. And hey, all of this was designed and built by two programmers working from home on dialup connections.. a powerful argument against the committee process and big company involvement. AW is owned by a company and is considered "proprietary" but it works, the users extend it and stands alone today as a community supported significantly large multi user 3D environment success.
In my opinion, X3D will build to a vastly shrunken audience and aims at an application space which is not very interesting frankly (does it stir the creative or innovative side of what kept many of you in the game and believing?). One is banner ads, and frankly who cares? Is that cyberspace? No! Frankly I find all of this blind following of the dictums of e-commerce disturbing. I think Ecommerce is not cyberspace, as it lacks a caring and creative aspect about people and their interactions and is anathema to why we go online (for other than utilitarian purposes). The net innovated out of an open community mechanism and we still spend the most time online connecting with other people, not ordering products. E-commerce has the attraction of "credibility" and possibly investment but it is a fickle market with poorly defined uses for database visualization and the other apps that X3D has identified. I feel they are chasing some goals that either have no proven basis for real applications or may be done best by simpler technologies built to suit. And they are going far far far from the original vision of "building cyberspace" as enuciated all those years ago by Mark Pesce, Tony Parisi, Mitra, yourselves and many others. Hey, the Web3D folks are missing the killer app of Cyberspace: people and their virtual communities. IYou are right in feeling that it is a travesty and something ought to be done. But what?
Well the most sensible thing would be to air this to what remains of the VRML community and the Web3D board. This should be done only if you share some of the above feelings. Again, these are just my opinions. I would have no problem being part of some sort of "co-proclamation about community and the future of virtual world cyberspace". Hammer it to the doors of the Web3D Consortium and see what happens. If they determine that shared cyberspaces are not important to their future, then you might consider picking up your folks, technologies, and energies and making a very public exit, clearly stating why to the community and what your vision is. Schisms have produced new flowerings in cultures and innovations in the past (protestantism, Intel etc).
Next, ask yourselves: why do you need the X3D/Web3D Consortium or the corporate partners to develop your vision? Look at the Contact Consortium site at:
http://www.ccon.org
See what has been accomplished with NO funding (other than conference sponsorships) and little official corporate backing over the four years of our history. Its all volunteers folks. And hey, like the VRML Consortium and others, we held conferences, wrote articles, received a great deal of press coverage, created great experiments in multi user spaces and most importantly, did the whole lot by sticking close to the user.
So where would you go? Well of course you could form your own separate community. Alternately you could take on and kick start the Contact Consortium's nascent OWorld SIG, join our community and benefit from a rich set of alternative people and approaches that are demonstrably pushing the MU envelope successfully every year.
I know this is a lot and some of the above should be taken in the context of my own personal frustrations at seeing good work going to waste in the VRML effort when it was largely avoidable. You have hit at least one nail on the head. So let me know what you think about all this... and the others on the CC, am I crazy?
How about we all get signed up on OWorld and talk? Sign up at:
http://www.ccon.org/lists/oworld.html
I would prefer that you do keep this particular dialogue to ourselves (please dont forward it on to anyone just yet) until we determine if there would be some useful common action. Frankly I have no reservations about being part of a message to yours/the Web3D community (and having my words used if they are at all of interest), but it would have to come from a group of us in a concerted action, not from just me.
Best to you and thanks for making contact (that is our name),
and I do look forward to continuing the discussion!
Bruce
cc Board, interested parties.
At 11:19 PM 2/15/99 -0500, J. Eric Mason wrote:
>Mr. Damer;
>
>I am writing to put a bug in your ear.
>
>The Web3D consortium is moving their next-generation standard (X3D) to
>target:
>( from http://www.vrml.org/news/x3d/mrd09.html )
>
> 1. Advertising/Page Animation
> 2. E-commerce Product Visualization
> 3. Database Visualization
> 4. Augmented News/Documentary
>
>The raises the possibility that multi-user functionality will continue
>to be unaddressed in the near future by Web3D despite its Living Worlds
>group, the DIS-Java-VRML and the new VRTP group. In fact, I have read
>comments from the X3D development team that specifically distance X3D
>from this market. As an active member of the Living Worlds discussion,
>this does not give me a warm feeling about progress.
>
>How do you feel about the CCON creating a technical entity to encourage
>and support MU3D/VR browser interaction and data/wire format standards,
>and possibly even content standards such as Living Worlds? How would
>your members feel about it?
>
>I admire what the Contact Consortium has accomplished for furthering
>interest in multi-user 3D/VR communication. And even though I am
>interested in getting my company, VR Telecom, on board as well, in this
>communication I speak for myself.
>
>Others I have spoken to, including Bernie Roehl (cc'd on this message)
>also see a need for a champion of MU3D standards, and don't feel Web3D
>is going that direction. Is CCON the right place?
>
>Thanks for your kind consideration.
>
>JEM
>
Bruce, the CCON board and friends -
I have signed up for the OWorld list, and will probably start posting there
after VRML99.
I agree with Bernie that distancing CCON from Web3D is not a great idea.
They're doing enough to distance themselves from VRML and further from MU. In
fact, I seek cross-development possibilities. At least in the psychological
sense of the word, "development."
Concentrating on the technical details of VRML was all the creators could
reasonably address. The market for MU3D was young and the hope was that VRML
would provide the foundation for higher-level interactions to be determined
later. Of course, it is easier for two programmers to create an application
from scratch to accomplish a particular task, than for a hundred to create a
widely useful framework. That reasoning has led to the Web3DC creating a small
core group to spec out X3D. Unfortunately, they have done this without
securing a feeling of trust from the community, but that's not the first nor
the most heinous of VRML/Web3DC's public relations blunders. When they moved
from VRMLC to Web3DC it was couched in the idea that Web3DC would embrace and
standardize more than one segment of the online 3D community, and VRML, Java3D,
and ChromeFX (R.I.P.?) were the big contenders. The current monomaniacal focus
on X3D leaves VRML appearing more dead than ever and gives me cause to wonder
if the expansion of the charter to all-3D-on-the-web really has happened.
Despite the "we're friends with Sun and encourage MS to join" press at the time
of Web3DC's name change, we are still looking at a single-issue Consortium.
This is the view based on PR . The fact is, there are a lot of people trying
to get multi-user and streaming VRML to work. The LivingWorlds working group
and the VRTP/DIS-Java-VRML working groups of Web3DC prove that fact, even as
they hurdle incredible technical difficulties in the process. Perhaps a more
active Technical Review Board will be able to push issues inside Web3DC to
provide at the least some basic core services that support these working group
efforts in X3D, and not lose sight of some of the most popular applications of
3D on the Internet. That's one reason I'm running for a seat on the TRB in the
Web3DC. If elected, I do intend to push MU and link-level issues in the TRB
and attempt to get some of the issues the CCON and Bruce have raised addressed.
I wish there was an organization with Web3DC's technical expertise and CCON's
user perspective and public relations ability. I don't believe the two will
ever completely join, but I think crosswalks can be encouraged.
JEM

Hello OWorld, my name Bruce Damer, moderator of this list (actually a
pretty silent moderator). I think we have enough folks signed up (over 40
now) so it is time to make introductions. I think we can all stick to a
pretty simple formula:
1. Where do I come from?
2. Why did I join OWorld (ie: why do I believe in open community efforts to
create virtual worlds and platforms)
3. What am I working on/would like to work on
4. Controversial statement of the day

1. Where do I come from?
******************************
I came from a background of ten years building graphical OS environments
for Xerox and others, having worked on some of the original windows and
icons stuff based on the STAR system (which the Mac and Windows copied).
See http://www.damer.com/background/cv.htm for details. I got tired of the
2D page-based metaphors of interaction and in the early 1990s left a
software project I was helping direct in Prague to retun to live in
Northern California and plug into the then emergent visual Internet. I
learned in MUDs and read Snow Crash and dreamed of a visual community
metaphor that would punch through the documents of the web and make
cyberspace into a true place. So I co-founded the Contact Consortium in
1995 with anthropologist Jim Funaro. The Consortium is the rather organic
virtual organization that provides OWorld a home and seeks to catalyze the
development of shared visual virtual worlds in cyberspace. We have held 5
conferences and a gazillion other events around the Earth and built many
experimental worlds in four years. The Consortium membership is large and
very diverse and knows one heck of a lot about multi user worlds and what
keeps users coming back. See http://www.ccon.org for the Consortium. I also
run a small company that builds worlds for real customers, see
http://www.digitalspace.com.
2. Why did I join OWorld (ie: why do I believe in open community efforts to
create virtual worlds and platforms)
******************************
Well at the 2nd annual "Digital Biota" conference (Contact Consortium and
Cyberlife) in Cambridge UK in sept 1998, there seemed to be a great brain
trust of folks who were passionate about bringing biological metaphors into
cyberspace. The OWorld list and SIG were born at that conference. I would
like to see worlds where biological processes can express themselves, like
Tierra, Nerve Garden and Karl Sim's evolving creatures (see
http://www.biota.org). Heck I want to live to see a "cambrian explosion in
cyberspace". These worlds have to be inhabited by users to provide the
stimulus for the organisms.
3. What am I working on/would like to work on
******************************
Well I am not a programmer anymore, but can be a good ringleader.
4. Controversial statement of the day
******************************
Can open source platform efforts work, or will they produce bloated kludges
that are Xanadu-ed into long delayed nonexistance? Java or ActiveX.. is
Windows the only client side platform that matters now?
Hey, I also wrote a book on avatars and virtual worlds too! Here is a
shameless plug for it.. find ALL the virtual worlds you could ever want to
download at:
http://www.digitalspace.com/avatars
and YOU are up next..
Bruce Damer
(DigiGardener)

++ The Contact Consortium ++
A Forum for Contact, Culture and Community in Digital Space
Visit us at: http://www.ccon.org
or at the Consortium by email via our Webmaster
/- Living, Learning and Working in Virtual Worlds -/

I am assuming that you are inviting us to send in brief introductions?
Here's mine.
1. Where do I come from?
I left college at 18, clutching my A-levels in one hand, a place at King's
College London in the other (to study zoology) and my eyes and heart firmly
set on becoming a ballet dancer. So I went to London and joined a ballet
school! Two ballet schools and 5 years later (life had been a revelation
to me whilst training), I found myself in Munich with my eventual wife,
Katy, dancing with the ballet company of the Bavarian State Opera Company.
5 years later, I left King's College London (yes, I went back eventually,
having partially scratched the itch) as an ecologist and eased nicely into
a job with English Nature, the Nature Conservancy Council for England, as a
Conservation Officer.
2. Why did I join OWorld
To learn what the heck most of you are talking about, most of the time ...
worlds within worlds.
3. What am I working on/would like to work on
I am still with the same agency (plus two lovely children) but have also
co-founded Wild World Limited, a company with an idea! Put simply, our
aims (my co-Director is Graham Allum) are to develop a product that can
have a significant market presence, in order to generate a revenue stream
for biodiversity conservation. Our product will, to begin with, be
predecated on a games play scenario, relying upon A-life, QTVR and motion
capture to render a believable, cognitive environment based upon endangered
species. Available on the Net, users will assume the identity of these
creatures (seeing the world through their eyes), needing to survive and
thrive.
Like the rest of you I assume, the biggest problem is finding sufficient
time to devote to this work, it forming a part-time activity at present.
Still, we are getting places and a few A-lifers know of our existence; even
more of you now!
4. Controversial statements of the day
Should we get A-life or should we GET A LIFE?
The only way that A-life can hope to evolve as a credible science is for it
to be based upon industrial, applied uses. (Witness the arms race, space
race and pharmaceuticals race as examples.) Theoretical science may have a
place but the world happens to be falling apart during our lifetimes so
technology and/or withdrawing technology should address this first.
Yours sincerely
Paul Sinnadurai
Hello Oworld, Bruce...
Bruce Damer wrote:
> Hello OWorld, my name Bruce Damer, moderator of this list (actually a
> pretty silent moderator). I think we have enough folks signed up (over 40
> now) so it is time to make introductions. I think we can all stick to a
> pretty simple formula:
>
> 1. Where do I come from?
> 2. Why did I join OWorld (ie: why do I believe in open community efforts to
> create virtual worlds and platforms)
> 3. What am I working on/would like to work on
> 4. Controversial statement of the day
>
OK, So, let's try....
1. Where do I come from ?
My name is Remi Sussan, french journalist in computer magazines. I hope
you
will forgive me for my numerous english mistakes in my posts. I
specialize in
Internet, virtual reality, intelligent agents...But I must precise that
I don't
subscribe to this mailing list (or any other list) *as* a journalist. I
am
interested in things above all because I find them interesting or fun, I
only
discovered that writing articles was the best way for me to get money
from my
favorite topics...

2. Why did I join OWorld (ie: why do I believe in open community efforts
to
create virtual worlds and platforms)
I am fascinated in virtual worlds because I want them to help me to
become more
intelligent, by letting me visualize and manipulate things and objets of
knowledge that cannot be apprehended intuitively in the real world.
Bruce,
since you quoted Tierra, I remember a Ray's text about "Evolution as an
artist", where he said that some evolutionary processes appeared as true
works
of art to evolutionary biologist: but he was alone to perceive them as
such. By
the way, Ray's tierra is the best example of the problem. fascinating,
but
almost unusable, because devoid of any kind of intuitive interface.

I'm convinced that the process of learning could be dramatically
accelerated if
we knew how to how to represent the field of knowledge we want to teach
(or
learn). I dream of "encyclopedic virtual worlds", not so far from the
mental
constructions of the Renaissance's "ars memoria", but less static, full
if
artificial life forms or agents able to help you and guide you in your
travel.
I'm also fascinated by the "augmented reality" concept: a virtual
reality
system which does not isolate you from the external world, but gives you
insights about it. for instance, a surgeon who sees some of the patient
data on
his screen, surimposed to the picture of the body he is working on.
3. What am I working on/would like to work on
Ia m not a coder either, but I can program small parts, especially if we
use
Java (I'm in the process of learning Java 3D). but above all, I can
Beta-test
by writing content for the system. BTW, I'd like the content to be
easily
written with a script language like VRML and Povray. It's no interesting
to
create an open source system if you need an expansive authoring package
to
write contents. Unfortunately, the VRML community seems to prefer create
a new
version which will need authoring packages,and this is IMHO a mistake
(for
instance, some want to suppress the most useful nodes, Extrusion and
ElevationGrid)
Controversial statement of the day
I divide it in two parts. the first will be my controversial statement,
the
second my point of view about Bruce's one.
1) Since we are in the process of devising a virtual world system which
will
(hopefully) gain some universality, we have to ask about the (relative)
failure
of the father project, VRML. In other texts, Bruce pointed out the
difficulty
to create new objects, the lack of any "Lego game" strategy. He also
mentioned
the difficulty to implement multi-user systems in this language. I'd
like to
add an other possible cause.
If you do virtual reality, especially a "serious" one, you want to
represent
something. When you do virtual reality in the net, the obvious thing
which has
to be represented is the internet itself. the virtual world is not a
static
one, it's a dynamic representation of what happens in various corners of
the
Net. When Mark Pesce had the idea of what was to become VRML, he also
created a
system -the cyberspace protocol- which has the goal to transform the Web
as we
know it in a gibsonian cyberspace. Finally, the cyberspace protocol was
abandoned and VRML was born as a simple 3D file format. My question of
the
day will be: is it time to resurrect the cyberspace protocol ?
2)
<<Can open source platform efforts work, or will they produce bloated
kludges
that are Xanadu-ed into long delayed nonexistance? Java or ActiveX.. is
Windows the only client side platform that matters now?
>>
open sources platforms worked for Linux, for Pov-Ray...If open source
has been
now chosen as a model by companies like sun (Jini) or Netscape
(Mozilla), I
suppose it can work...I suppose that the best way to avoid the Xanadu
syndrome
is to create a system with a bottom-up approach, by writing little
programs and
utilities which can connect together.
Windows is probably the only client side platform which matter as long
as the
traditional micro-computer remains the only hardware structure. IMHO,
the best
machine able to do virtual reality (and of course augmented reality)
would be
the wearable. It's therefore important to create an universal, cross
platform
product (and Java is therefore the language of choice), so we can test
it on a
PC, but be ready for future machines...
Remi
Hi
OK, my turn - thanks to Bruce for kick-starting the activity on the
list again; it's good to see that people are active again.
1. Where do I come from?
Degree in chemistry, PhD in "chemistry" as well - actually
visualisation and calculations on complex biomolecules. In reality it
was a way for me to play around with computers while giving the
impression I was "doing science". I became tired of that charade and
took a job with BT doing research into ALife, 'cos it seemed
interesting at the time. It certainly has been!
2. Why did I join OWorld (ie: why do I believe in open community efforts to
create virtual worlds and platforms)
Virtual worlds are boring if there are no inhabitants. Who wants to go
and walk around a static, quiet world more than once? Not only do we
need a living, growing, expanding environment, we need lots and lots
of people and creatures to live in it. The only way we can get a
critical mass of content and users is by going Open Source.
I was also lucky enough to attend Digital Biota II in Cambridge last
year. While that conference was inspiring, there was little in the way
of *real work* that came out of it. I see OWorld as a way of
channeling some of that knowledge and enthusiasm into a real (well,
virtual!) application.
3. What am I working on/would like to work on
Currently working on lifelike characters to inhabit virtual worlds and
perform some task, e.g. guides, helpers, hosts.
My skills are in Java, plus VRML, Java 3D, general web stuff. I like
developing new ideas that people will look at and say "wow"!
4. Controversial statement of the day
Static virtual worlds are boring. Discuss!
--------------------
We should try to leverage as much open source stuff as possible, so
that we have to do as little development as possible ourselves.
We should also try not to tie ourselves into a particular graphics
technology, as these are notoriously fickle. Look at what's happened
to VRML over the past few months. I suggest we have the "guts" as a
separate, non-graphical back-end, probably in Java, and have viewers
that can connect to the system. e.g. the world itself is like a huge
behive, and we can write viewers that will "lift the lid" and peek
in. Of course, we'll need clean APIs etc for this, but that's a long
way off!
Cheers
Glenn.
Hi,
My name is Danny Ayers. Following Bruce's lead :
1. Where do I come from?
Started with an 8k PET about 20 yrs ago, though had a huge gap where I
was doing electronic music related stuff, until I ended up getting back
into IT with a 286. Academically I've done a lot of studying but have
little on paper to show for it.
2. Why did I join OWorld (ie: why do I believe in open community efforts
to create virtual worlds and platforms)
Vaguely anarchistic tendencies, I suppose...
3. What am I working on/would like to work on
In part because of the lack of degrees etc, I'm currently looking after
a network in a college in the north of England.
At home I'm working on a search engine in Java, the plan is to build in
loads of AI features.
I'd like to work on AI/Internet stuff professionally.
4. Controversial statement of the day
If every user on the Internet had a neural network of about 1000 cells
on their machine, the total number of cells would be about the same as
that in the human brain.
--
Intermittent Web site :
http://members.xoom.com/dayers
or
http://www.danja.free-online.co.uk

Alternate email :
Danny_Ayers@yahoo.com

Jay Fenton AKA Jamie Faye Fenton
http://www.fentonia.com/bio/
http://members.tgforum.com/jamie/
1. Where do I come from?
A video game designer since 1975, later invented MacroMedia Director.
I did several early MUVE experiments with the Vivarium project and later
at Electric Communities. Right now I am at SRI International doing a MUVE
project for educational purposes.
2. Why did I join OWorld (ie: why do I believe in open community efforts to
create virtual worlds and platforms)
I believe the only way VR will take off is as an open, multi-colaborator
standard.
3. What am I working on/would like to work on
I am developing a MUVE technology in Java called MEOW. MEOW stands for
Multiuser Educational Online Workspace. I am awaiting clearance from
the legal department to do an open-source release of this.
MEOW is intended to be a distributed MOO which can be text-based, 2D, or 3D.
The earliest use of MEOW is a chat space I am setting up for an Internet
site I operate called Transgender Forum. We hope to be open in a few
days.
MEOW emphasizes wide participation by not requiring any client software to
be installed beyond a "late model browser". I have found that requiring
a "custom client install" cuts the audience by 90%.

4. Controversial statement of the day
VR is being held back by the ability of existing Telecom companies to
manipulate the political process to keep costs 3000% higher than the
actually are. A T-1 line should cost $50/month.
-- Jamie
Hello oworld,
I am Margaret Corbit. I am interested in using 3D virtual worlds for
science education. More from the museum perspective (informal ed or
edutainment) than from that of a classroom teacher.
1. Where do I come from?
*****************************
Well, I have been the science writer at Cornell's supercomputing center
since 1992. It was quite a shock coming here from the botanical garden and
a liberal arts background, but I love it. (I have a degree in Theatre and
an MS in Plant Ecology--thesis paper coming out this spring.) We have a
CAVE-like environment, the Visual Insight Zone or VIZ, for scientific
visualization and a wonderful group of professionals and students
developing applications for 3D: molecular docking, fracture mechanicx,
asteroid sampling, etc. I've been a groupy for the VIZ group since I arrived.
As we moved onto the Web for publishing, I pushed to get real data into my
features in 3D-early VRML. (see some of my projects from my homepage listed
below.) So I have been out preaching the potential of 3D on the Web for
science outreach for a while now.
About a year and half ago, I saw Bruce give a talk/demo at WebNet and
became hooked on virtual worlds. I am not a technical person; I mostly rely
on talented undergraduates at Cornell to support the technical side of
developing our features. They are now working on an Active Worlds project,
SciCentr.
2. Why did I join oworld? For the same reason that I have the Active Worlds
browser set up to use Netscape on my NT desktop. I want to support
development of a communications medium that will be universal, even if that
isn't possible. At any rate, I see an interesting interplay between
opensource and commodity applications. The students want to work in
opensource, for philosophical reasons primarily and also because they want
to get their hands into the guts of the software. It's the nature of the
Internet that there will and should always be a free and open way of
viewing all the file formats. Maybe once a concept has matured, it should
be handed off as a product for maintenance and upgrading, afterall these
things aren't fun and therefore cost money.
3. What am I working on?
I have been doggedly trying to get an interactive, multiuser 3D virtual
science center going ever since the day I met Bruce. Progess is slow but
picking up. We are a world in the Active Worlds universe: SciCentr. Not
much in the world yet, but the web pages
(http://www.tc.cornell.edu/Exhibits/Worlds) will give you an idea of what
we hope to accomplish. We are having a great time setting up collaborations
with artists, builders, designers, scientists, etc.
I am interested in incorporating simulations and emulations of simulations
into the world and into making multiuser activities. This is one situation
where an open source would be great. We are focusing on Bots right now. And
rendering topography files on the fly. Don't know much about Java3D,
although it was all the rage at SC98. See my controversial statement below.
4. Controversial statement of the day
*********************************************
Unfulfilled promises for Java and Java3D will be the main stumbling block
for opensource development.
Remember, I am not a technical person :-), so this is the observation of
someone from the fringes.
Cheers,
Margaret
Margaret Corbit
Science Writer, Cornell Theory Center
533 Rhodes Hall, Cornell University
Ithaca, NY 14850-3801
corbitm@tc.cornell.edu
http://www.tc.cornell.edu/~corbitm
Voice 607 254 8716
FAX 607 254 8888
Hi Listmates,
I’m Sue Wilcox author of ‘The Web Developers Guide to Avatars’ the tools and
technology book about the virtual worlds arena.
1. Where do I come from?
I live in Hawaii but am an ex-pat Brit. I have degrees in Psychology and
Computer Science but I’ve probably spent most of my life reading science
fiction. I spent several years at the beginning of VRML being VRMLPro to the
developer community. I believed and still do that the 3D interface is what
computer users need and will one day want. But then disillusionment with
progress set in and I’ve moved on to the greater next big thing: Digital
Biota.
I have no doubt that an inhabited cyberspace is going to be the growth
phenomena of the 21st century. I’ve known Bruce Damer for several years and
worked with him on the earlier Biota conferences. Now I’m doing the organizing
for the next and third conference. Digital Biota are what we need as plants,
animals, ecosystem, amusements, inhabitants, friends, pets, and a challenge to
our human-centric view of the virtual worlds.
2. Why did I join OWorld (i.e.: why do I believe in open community efforts to
create virtual worlds and platforms)
I was at Digital Biota 2 when Bruce proposed the idea of a list to shelter
discussions on encouraging the production of inhabited virtual worlds and I
thought it was a good idea. Most discussions of virtual worlds ignore the
possibilities of creating lifeforms to live in these worlds. I think they’re
essential parts of having captivating places to visit in cyberspace. The
internet is a creatively chaotic place I believe in the dynamism of such
community efforts and the way things emerge unexpectedly from such attempts.
Digital Biota are a subset of the ALife universe and deserve special attention
because they have the potential to be new life not just computer programs we
can use.
3. What am I working on/would like to work on
I have a project underway to produce a book on the work currently in progress
in the realm of Digital Biota. Its working title is: "Digital Biota: ALife in
Cyberspace". Its evolution is closely hooked into the work going on in the DB
field so I’m not expecting it to be ready until some time late in 2000.
I’m also working on a SF book exploring the possibilities of symbiogenesis for
higher organisms. It’s good to try to actually work out how these theoretical
ideas might work out in practice and its another way to get the memes of ALife
into the mass mind.
I want to encourage work on creation of life projects by helping more people
see what is possible and by getting cross-discipline communication going among
all the divergent groups who could contribute to the development and
understanding of artificial life. See the latest mission statement (which I’ve
just reformulated) on the Biota.org site:
<http://www.biota.org/>www.biota.org
4. Controversial statement of the day
I think cross pollination is incredibly important to the development of
inhabitable virtual worlds. We need to hook up not just with the other Alifers
but with people in the games industry and entertainment. And lets face it the
military are into avatars as well as major world simulations now so they must
be using ALife technology. We should at least try to be aware of what they are
up to. The other big driving force in computer development is the porno
industry. Can we expect pornographic virtual worlds? There are a lot more way
of conceiving of virtual worlds and their inhabitants than we have concerned
ourselves with in the shallow pool of the virtual worlds community. Do we need
to be in the ocean in order to evolve? And do we need the big bucks to help
our
development efforts? More than open source we need open minds to see the
opportunities and possibilities before us.
I look forward to the day when we have to concern ourselves with the ethics of
how we deal with artificial lifeforms.

Hi All,
Here is my intro:

> 1. Where do I come from?
> ******************************
My name is Róbert Viðar Bjarnason, born on the 19. of July 1972 in Reykjavík,
Iceland.
In 1984 I did the design and implementation of an emergency public broadcast system
for the Icelandic government. This was designed for and used in the big Icelandic 1984
media strike. In this strike there was no radio, no TV and no newspapers for two weeks,
affecting all the Icelandic population. The people got all their news trough the
public broadcast system that was composed of a central computer, close circuit
video system and 100' of TV monitors placed in public location around the largest
towns. Since then I've been involved in coding and innovating. I started the first commercial
ISP in Iceland in 1994 and the one of two first commercial ISP in Denmark in 1995.
One day I found my self immersed in Alpha world and I've been hooked on VR since.
> 2. Why did I join OWorld (ie: why do I believe in open community efforts to
> create virtual worlds and platforms)
> ******************************
I joined oworld originally because I hosted the Icelandic arm of Avatars98 and got
hooked up trough that event.
> 3. What am I working on/would like to work on
> ******************************
I´m currently heading the development of an open-source multi-user
3D client called BANG. Its 100% Java with support for VRML97 and native J3D
for worlds and avatars. The next version of the client is due 15 of march and will include
simple authoring tool for placing and manipulating 3D objects in the shared sector based
persistence space. http://the.bang.is/
> 4. Controversial statement of the day
> ******************************
The Sun is shining brightly outside.

Róbert Viðar Bjarnason
robofly@bang.is
Bang Space Inc.
CA, USA.

Hello, everyone!
I'm Bernie Roehl, and if you want to know more about me just check out:
http://ece.uwaterloo.ca/~broehl/vrbio.html
> 1. Where do I come from?
I was born in Montreal, and I'm currently living in Waterloo, Ontario (about
an hour west of Toronto).
> 2. Why did I join OWorld (i.e.: why do I believe in open community efforts to
> create virtual worlds and platforms)
I'm convinced that the only way for multi-user virtual environment technology
to progress is for there to be publicly available source code.
> 3. What am I working on/would like to work on
My main interest at the moment is in designing an architecture for shared
virtual worlds, and creating or joining a team to produce free software for
that purpose.
> 4. Controversial statement of the day
Check back with me in a month or so. :-)
--
Bernie Roehl
University of Waterloo Dept of Electrical and Computer Engineering
Mail: broehl@ece.uwaterloo.ca Voice: (519) 888-4567 x 2607 [work]
URL: http://ece.uwaterloo.ca/~broehl
>X-Authentication-Warning: vrml.org: majordom set sender to
www-vrml-approval@web3d.org using -f
>Date: Sun, 28 Feb 1999 23:27:24 -0800
>From: Don Brutzman <brutzman@nps.navy.mil>
>Organization: Naval Postgraduate School
>X-Accept-Language: en,fr
>To: VRML mailing list <www-vrml@vrml.org>,
> java3d mailing list <java3d-interest@Sun.COM>
>Subject: new book on networked virtual environments
>Sender: www-vrml-approval@web3d.org
>Reply-To: Don Brutzman <brutzman@nps.navy.mil>
>
>"Networked Virtual Environments - Design and Implementation"
>By Sandeep Singhal and Michael Zyda
>
>Cool book. Table of contents available at
>http://www.npsnet.nps.navy.mil/zyda/NVEBook/Book.html
>
>and ordering info available at
>http://www.npsnet.nps.navy.mil/zyda/NVEBook/SinghalZyda2.pdf
>
>Site extracts follow. Highly recommended.
>Disclaimer: I get fired if I don't really really like this book. :)
>
>Networked Virtual Environments (Net-VEs or NVEs)
>allow multiple users to interact in real time, even though
>participants may be located in different places around the
>world. Providing realism for users by incorporating 3D graphics and
>stereo sound, NVEs create an immersive experience in shared virtual
>space via the Internet.
>
>Networked Virtual Environments, by Sandeep Singhal and Michael
>Zyda, teaches the design and implementation of NVEs to commercial
>NVE infrastructure and applications developers in the military,
>in the aerospace industry, in entertainment, and in other commercial
>areas such as engineering design, scientific research, and even
>retail business. Demonstrating how to build and operate interactive
>3D graphics sites online, the book includes information on ways to
>overcome the problems and limitations encountered in real-time
>interactive system development. Readers find out how to build tools
>for NVE technology, learn the underlying principles and practical
>techniques for making successful NVEs, and glimpse future direc-tions
>in this growing field.
>
>The authors plan an associated web site that will show how to build
>a usable networked virtual environment.
>
>“I love this book. It is technical, comprehensive, well-written, and
>deals with an area which is gaining popularity rapidly, both in
>academia and industry.”
>—Dr. Gurminder Singh, Director (Learning Lab), Kent Ridge Digital Labs,
>Singapore
>
>“[The] first comprehensive reference on the subject of Networked
>VR. Very readable, with a mix of the history and the science on
>the topic.”
>—Michael Macedonia, Chief Scientist and Technical Director, US Army
>Simulation, Training, and Instrumentation Command (STRICOM)
>
>“The book provides a good road map of the field, pointing to
>recent, relevant research in the field ... it identifies all the critical
>topic areas and provides pointers for in-depth study.”
>—Katherine L. Morse, Senior Computer Scientist, Science Applications
>International Corporation
>
>
>all the best, Don
>--
>Don Brutzman Naval Postgraduate School, Code UW/Br Root 200 work
831.656.2149
> Monterey California 93943-5000 USA fax
831.656.3679
>Virtual worlds/underwater robots/Internet
http://www.stl.nps.navy.mil/~brutzman
>-------------------------------------------------------------------------
>for list subscription instructions,
>send email to www-vrml-request@web3d.org with text "info"
>
>
>
My name is Mark Shirley.
1. Where do I come from?
I've loved computers since getting the boy scout merit badge
way back in, well, a long time ago. My first love was
computer graphics. In high school, my nerd friends and I read
the Newell and Sutherland graphics book and created a wire
mesh animation of an X-wing fighter rolling over the
deathstar.
My second love was Artificial Intelligence. I joined MIT's
AI lab to build Asimov's robots, and ended up building
algorithms for diagnosing faulty computers. My third love has
been computer-based communities. After MIT, I joined Xerox
PARC and initially worked on algorithms for diagnosing faulty
copiers. Eventually we discovered that using computers to
better connect the thousands of people who fix copiers with
each other was going to be more powerful in the near term, and
we developed and deployed groupware for proposing, validating
and distributing problem/solution cases. At PARC, I also fell
in love with online communities through a project called
Pueblo, a MOO hosted in an elementary school in Phoenix, AZ.
Recently, I moved from PARC to NASA's Ames Research Center
where I'm again working on AI software to diagnose machines
(service calls are more expensive in space than in offices on
earth). However, I retain my interest in online communities
(see below).
2. What am I working on/would like to work on
I work in a research group that creates new kinds of autonomous
systems. We focus mainly on spacecraft, rovers and subsystems
on manned flights like controllers for life-support systems,
but we also touch on sofware agents. One example is the Remote
Agent Experiment that will be uploaded and tested on Deep Space
1 (http://nmp.jpl.nasa.gov/ds1/). This is the first instance
of AI planning, execution and monitoring technology getting to
control a real spacecraft.
3. Why did I join OWorld (ie: why do I believe in open community efforts to
create virtual worlds and platforms)
I am deeply impressed with the ability of moving, 3D worlds to
engage the mind (e.g., the best Nintendo 64 or PC games). I
would like to use such worlds used for education and learning
play. But authorship in the game examples is too centralized.
I like the mud model, where lots of people create and share
things. ActiveWorlds is a great start, but I'm interested in
worlds with rich and interesting behaviors in them. I believe
the best way to have worlds with these characteristics is from
an open source foundation which can grow and change rapidly;
there's too much still to work out to be bound by any company's
api's. And naturally, I think that NASA has some of the
coolest examples of subjects and places that it would be
interesting to model in such worlds. So, in addition to what I
am working on, I would like to be working on an oworld too.
4. Controversial statement of the day
Bruce, you don't ask much. Ok, I'll ask it a question.
Is there an existing open source development project that might
serve as an appropriate model for work on an open source 3d
world server?
I mean this in a broad way. For instance, linux had an
existing OS, unix, as a target. This gave a lot of focus and
structure to the project because people understood the sort of
thing they were building in a lot of detail. Unix also limits
Linux. There are really cool ideas about how to do an OS that
are outside what Linux can do because they would involve too
radical a change from Unix. There must be open source projects
out there that are building something totally new (I don't have an
example on the tip of my tongue, though).
In the world of world servers, there are some good existing systems.
Perhaps an open source server similar to one of those would be
a good foundation and allow a reasonable amount of innovation
while anchoring the conversation and keeping it from drifting
in our differing and endlessly complex visions of a future
metaverse. On the other hand, all of the existing servers (in
my opinion) are missing critical things and would be too
limiting in the long run.
Perhaps there should be a few oworld projects?
[In looking back over this last, I realize want to ask the question
in two senses. (1) should our goal be to create an open source
version of something that basically already exists, or not, or both?
and (2)
does anybody know of an open source project they feel went
or is going especially well? If so, what are they doing right?]

Hi to everybody,
My name is Roberto Guijarro and I am from Madrid (Spain)
First of all I would like to apologize for my english (I promise to improve it)
1. Where do I come from?
I have a degree in fine arts and graphic design and I have dedicated twelve years
of my life doing traditional and computer animation. I have worked for TV programs,
spots and animation series for a lot companies from around the world. But it was
a long time ago. After this I worked in the development of several computer based training
programs designing interfaces, animations and even developing plots and writing scripts
for them. I consider myself a creative person and I like experimentation very much
so I have seen on virtual communities an new open door as a way to come
into a house of fun with a lot of surprises and discoverings to do... I feel like
I am a child again, looking around, discovering and learning a lot of thinks...
2. What am I working on/would like to work on
Actually I am working on a interesting project building a 3D model of a very well known spanish
monument and our end goal is to create a multiuser Internet based virtual space in order to let
people from around the world to visit this monument offering the chance to interact with other visitors.
The project is in a preproduction step and we hope to start by next May month.
3. Why did I join OWorld (ie: why do I believe in open community efforts to
create virtual worlds and platforms)
Because I believe in people. I allways have thought what really makes a
software projetc to be a success are the users, and if the users participate directly
not only using but also developing, any kind of software project will be a success.
And afterall, a virtual world is a piece of software.
4. Controversial statement of the day
Money. I mean, big companies investing big amounts of money in R&D are not
going to open any source code. I think, only Universities, Research centers and Consortiums
with no private interests can undertake such kind of open work. Someting like Oworld can do it.
In the other hand, in an open community project willing to create
virtual worlds there is an organizational problem because it is necessary people
with time, a lot of time and willing to work for free, just the satisfaction to be a part
of the growing of an open project and be able to say: "It is MY project too, there is a
little part of me on it".
Finally, it is necessary a coordinator and a leader, right know you Bruce. It is a hard work
but I hope you will be succeded.
********************
Roberto Guijarro
rgbelda@lander.es

Hi Bruce + OWorld,

My name is Hamish Meares.
1. Where do I come from?
****************************
Basically I trained as a medical doctor in Australia, moved to London a
couple of years ago, but I have always had an insatiable curiosity for
interesting non-medical things (Richard Dawkins "The Selfish Gene", William
Gibson etc). Last year at the Digital Biota II Conference I had the
pleasurable experience of a fair few of my favourite topics suddenly
appearing under the one umbrella, ie biology, chaos, complex systems, JAVA,
neural nets, alife, virtual worlds etc.
Unfortunately, my programming skills are somewhat rudimentary, but I'd love
to throw a few ideas into the pool.
2. Why did I join OWorld?
***************************
After seeing Bruce's talk on virtual worlds at Biota, I was fairly blown
away by the possibilities. I very much like the idea of an open platform
with lots of people doing there own thing because I like the idea of
freedom of information and interaction of people, and because it's
basically a chaotic system, hence unpredictable, and therefore likely to be
produce things that none of us could dream of.
3. What am I working on/would like to work on
**************************************************
At present, apart from medical stuff (Emergency Medicine, and hopefully a
bit of Anaesthetics), just playing with ideas that arose from the Biota
conference. Such as, the fundamental interconnectivity between all living
things ie the impossibility of understanding an individual or group without
considering how it functions with other agents in the system, and how
drawing neat boundaries to define an object or individual within a system
gets farily blurred when you look closely at things.
I like Bruce's idea of a cyber Cambrian explosion. Ultimately I'd love to
see a world with an exploding abundance of digital life forms all growing
and dying, eating, crawling, multiplying, evolving, with humans being an
integral part of the ecosystem (but hopefully not being eaten). Maybe even
cyberbiologists leading cyber-expeditions to catalogue and study these new
forms, and perhaps discovering an amazing class of butterfly creature, the
germinal code being written by some 12 year old girl many years before.
Who knows...
4. Controversial statement of the day
**************************************
I still use a Macintosh

1. Where do I come from?
*************************
I originally did a degree in mathematical economics and econometrics and then revolted
against the subject and went into computing with ICL (a UK computer firm). Got
interested in AI about 1982-3 while working at Sheffield University and since then have
done various AI things (see my web page at
http://www.salford.ac.uk/iti/rsa/aylett.html).
I eventually got into robotics and became interested in biological approaches via work
on cooperating behaviourally driven robots. Now I am in a Centre for Virtual
Environments I am carrying on my interests in virtual worlds.
2. Why did I join OWorld
*************************
(ie: why do I believe in open community efforts to
create virtual worlds and platforms)
Open communities are likely to be inclusive: I am very wary of corporate ambition and
look forward to the day when the technology can be used to cement local communities in
which it is not just the computer literate, the educated and the well-off who have
access to the technology. Open communities are also likely (I hope) to be willing to
make access equal for those who do not choose to use Wintel platforms.
3. What am I working on/would like to work on
**********************************************
I do a lot of AI planning - my programming background is mainly in Lisp but I am
learning Java and VRML (because I am running a course in them!). I am interested in
synthetic characters in VEs and have just been involved in a Virtual Teletubbies project
(http://www.salford.ac.uk/iti/rsa/teletub.html) which I'd like to extend. I would like
to work on making VEs more welcoming to ALIFE - I am very conscious of how sterile and
clean a VE is compared to the richness and messiness of the real world in which robots
try to make their way. I believe that complexity arises from interaction and that VEs
have to become ecologies to be interesting, rather than just infiltrating bots, agents
or ALIFE constructs into them. I want plants, growth, decay and weather. I want a VE
that is an interesting dynamic system. I am very interested in how the symbolic arises
from the non-symbolic and would like to create the type of VEs in which this might
happen. I am also interested in the kind of emergent narrativity you find in some
physical games (such as football - note I mean soccer).
4. Controversial statement of the day
**************************************
Oworlds should be really open - and not Wintel-specific. A dynamic world has to develop
a history, and who knows what platforms will be around in five years time?
Cocktail party OWorlds are essentially fun only for that subset of humanity who enjoys
cocktail parties (count me out). Where is the world for people who like the interaction
between control and spontaneity found in gardening?
Ruth Aylett email: R.S.Aylett@iti.salford.ac.uk phone: 44-161-295-2912
Home page: http://www.salford.ac.uk/iti/rsa/aylett.html
fax: 44-161-295-2925
Centre for Virtual Environments, Business House,
University of Salford, Salford, M5 4WT, UK
'Life is beautiful.'
Greetings, OWorld.
-------> [1].
Where did I come from?
I escaped from the void 28 years ago, just over one year into the
epoch, if you follow the UNIX date format. Along the way I have taken up
the cause of many a doomed vision, including the Amiga and VRML, so I'm
pretty used to watching corporations shoot themselves in the foot. But that
is another story. _My_ story ends up soundling like a confused pile of
disciplines: illustration, programming, 3D design, user interface,
marketing. . . Mostly, over the last 16 years I have attempted to discover
and enhance the _art_ in the state of the art.
-> Why <--------
did you join OWorld? [2]
I was curious to see if the Contact Consortium was interested in
producing technical specification documents to allow VEs to
intercommunicate. OWorld was recommended as the most technically-minded
list to join. I would like to talk about content representation, world
database, and wire protocol standardization tracks.
[3.0]? -> WHAT? am I working on. . ?
I work for VR Telecom these days; since we're too small to shoot
ourselves in the foot, we prefer hitting our thumbs with hammers. We have a
new multi-user VRML-based app called Holodesk (TM) Communicator, and it's
currently free in an open beta test at http://www.holodesk.com
----> . . . What would I *LIKE* to be working on?! <---- [3.1]
Seems I primarily spend my time finding ways to fuse technology and
media. Holodesk is a good playground to see what can be done with VRML in a
multi-user setting, though authoring worlds is still a steep learning curve.
Hopefully object libraries, runtime object loading and persistence can help
that down the road.
But, I really want to direct.
<> Controversial statement of the day [4.05]
Pointy brackets good. Curly brackets bad. Discuss amoungst
yourselves.
<> J. Eric Mason aka jem@vrtelecom.com aka duckboy
<> VR Systems Programmer [geekyArtistType] for VRTelecom, Inc.
hello OWorld,
i'm Gerald de Jong
>1. Where do I come from?
originally from Canada, where i studied Math (U of Waterloo), but i've
been living in The Netherlands for a good number of years now. here,
i've formed my own company to do consulting and training and explore
other interesting opportunities.
>2. Why did I join OWorld
it's becoming clear enough that Open Source Software will play a very
significant role in the future of computing, and there has never been a
movement in this domain that has represented "community" to the degree
that OSS has. the core motivation behind virtual worlds scenarios is
nearly identical: community and communication. it seems completely
natural to me that an open source platform for virtual worlds will
prove to be the fittest and will therefore survive all other efforts.
the simple fact is that a virtual world scenario is to a very high degree
an enabling technology, and as such there are few better candidates
for successful open source development.
>3. What am I working on/would like to work on
my passion for years now is the exploration of natural geometry. to
most people educated in our age "natural geometry" sounds almost
oxymoronic, but that couldn't be further from the truth. geometry
is taught as something you draw on a flat piece of paper, but flatland
represents such a limited and dry subset.
my exploration has led me to a definition of geometry that is minimal
in many key aspects, and as a result it is optimal for application in
a virtual world scenario. the geometry is all based on the notion of
a "spring" or "elastic interval", and the diversity of form and behavior
that emerges from this simple core is baffling.
(see http://www.beautifulcode.nl)
i have built and evolved a piece of software called "Struck" which
facilitates this exploration, and i have a mailing list of users as well
as a weekly IRC session associated with the program. Struck is
capable of exporting to Active Worlds RWX format, as well as
VRML, POV (including animation), and my first efforts at employing
Java3D as a rendering mechanism have been surprisingly successful.
at Digital Biota II, Bruce Damer and Stuart Gold were both generous
enough to grand me 5 minutes of "borrowed time" to show some of
the results and tell a brief story about Elastic Interval Geometry
to the distinguished crowd. the reaction was overwhelming, because
EIG clearly has the potential of providing the "bones and muscles"
for digital life and/or the animating forms that should populate
virtual worlds in order to make them compellingly natural-looking.
>4. Controversial statement of the day
here's something i'm prepared to argue at great length with anyone
interested to give it the time of day:
"the only way to create a truly dynamic and compelling virtual world
scenario is through extensive and effective use of MOBILE CODE."
any other approach will be relatively static and boring, however
cleverly the APIs are designed, and will not survive in the long term.
not only data, but (emergent) behavior as well must travel around
the net. objects (as in object-oriented programming) consist of
both data and functionality, or better, state and behavior, or best
of all: "body and spirit". it's objects that have to travel.
---
Gerald de Jong, Beautiful Code B.V.
Rotterdam, The Netherlands, Tel. +31655893940
http://www.beautifulcode.nl
Gerald de Jong writes:
>4. Controversial statement of the day
here's something i'm prepared to argue at great length with anyone
interested to give it the time of day:
"the only way to create a truly dynamic and compelling virtual world
scenario is through extensive and effective use of MOBILE CODE."
any other approach will be relatively static and boring, however
cleverly the APIs are designed, and will not survive in the long term.
not only data, but (emergent) behavior as well must travel around
the net. objects (as in object-oriented programming) consist of
both data and functionality, or better, state and behavior, or best
of all: "body and spirit". it's objects that have to travel.
Gerald,
Could you say more about why you think mobile code is
essential? I agree with you, but my reasoning overlaps in some ways
and is different in others, I think.
I'm focussing right now on some requirements for defining object
behavior, so I realize there are a lot of things on the wish list
that I'm leaving out. Basically, I think an oworld needs:
1. Support for world objects to change structure and behavior over
time
2. A distributed implementation (including client<->server
distribution as well as server<->server)
3. Some distinction between system-level programming and end-user
programming
For me, it's #1 and #2 that imply a need for mobile code.
More detail:
A key need in a multiuser virtual world is the ability to modify
the world on-the-fly, without bringing it down. This includes not
just objects and geometry, but behavior too. And this means not
just introducing new classes of behavior, but modifying ones that
people already use. This requires the ability to change both the
objects and their internal structure.
This might be accomplished in different ways. For instance, one
might be able to modify an object directly. Or one might be able
to create a new object of a new class, copy information over, and
then replace all references to the old object with the new one.
Both would meet the requirement to allow change, although the
differences impact a lot of other design decisions for the world
infrastructure.
A second key need is that it must be convenient for people to
build on objects and behaviors created by others, and that the
environment support the kinds of problems that occur over time
when this is done. Examples are interfaces changing, objects that
are depended on disappearing, non-interface characteristics of
implementations changing (e.g., performance). Version control is
one technique for managing this, but as Microsoft knows with its
system DLL's, it can become a nightmare on a large scale if done
poorly.
Programmable MUDs (e.g., MOO) are one example of a solution to
this. (In that world, everything's dynamic and cooperation and
social controls help people work things out.)
System-level vs end-user programming
MUDs generally make a distinction between system-level programming
and end user world scripting. The deep roots of the world are
done in some conventional (and usually strongly typed) programming
language, the end users use an untyped scripting language with
small scripts attached to objects. Any script can be changed at
any time.
Some MUDs have a 2 level structure: system-level programming is
done in C by one or a handful of people, and end-user scripting is
done by everybody. Some MUDs have a 3 level structure. The core
system is handled by one or a few people. A larger group of
people write extensions, still in a conventional language, and
then end-users write in the scripting language. I would argue a 3
level structure for an oworld.
Distributed Implementation
Now, to mobile code. I think the world infrastructure should be
distributed for several reasons.
The server has to be scalable. It must be possible for
individuals or organizations to put a world or a portion of a
world up on their own hardware and control it. (ActiveWorlds is
scalable, for instance, by putting up many worlds on many servers
and supporting travel between them.)
Multiple servers plus people writing behaviors for their objects,
and the fact that people will want to build on each other's work
in some way, means that code will move somehow from server to
server.
The simplest way to do this is for people to copy the code from
one place to another. This is a start, but it's lame over the
long run. If people really build on each other's work, then the
sharing and versioning becomes a nightmare. In the MUD world,
building on work across servers is *much* harder than within a
single server. (At least it was until packages for automatically
moving code were written, and these still aren't integrated into
the infrastructure well enough for any but MUD wizards to use them,
to my knowledge.)
Why not just go for the endpoint of this progression and build on
a fully distributed, persistent object system?
Now, for the client/server argument. I would like clients to be
able to run behaviors like a server because I want end-user
scripting of the realtime dynamic behaviors of some objects. One
kind of "mini-world" I'd like to be able to create is, say, a
simulation of the space shuttle's Canada arm. I'd like a teacher
somewhere to be able to model the dynamics of the arm and a
satellite that the astronauts are trying to grasp and create an
experience where folks can try to do what some astronauts do.
Another example is a world like Mario 64 (with some educationally
justifiable content :^) where the dynamics of the character moving
matter.
Basically, I'd like to think of the client and the server as being
built on the same foundation. It's just that clients have GUI
code loaded while servers don't, and servers may have code loaded
for talking to persistent object stores, databases and other
things while clients probably won't.
Anyway, those are the thoughts your controversial question kicked
up. Comments?
- Mark

Folks another book on VEs..
>"Networked Virtual Environments - Design and Implementation"
>By Sandeep Singhal and Michael Zyda
>
>Cool book. Table of contents available at
>http://www.npsnet.nps.navy.mil/zyda/NVEBook/Book.html
>
>and ordering info available at
>http://www.npsnet.nps.navy.mil/zyda/NVEBook/SinghalZyda2.pdf
>
>Site extracts follow. Highly recommended.
>Disclaimer: I get fired if I don't really really like this book. :)
>
>Networked Virtual Environments (Net-VEs or NVEs)
>allow multiple users to interact in real time, even though
>participants may be located in different places around the
>world. Providing realism for users by incorporating 3D graphics and
>stereo sound, NVEs create an immersive experience in shared virtual
>space via the Internet.
>
>Networked Virtual Environments, by Sandeep Singhal and Michael
>Zyda, teaches the design and implementation of NVEs to commercial
>NVE infrastructure and applications developers in the military,
>in the aerospace industry, in entertainment, and in other commercial
>areas such as engineering design, scientific research, and even
>retail business. Demonstrating how to build and operate interactive
>3D graphics sites online, the book includes information on ways to
>overcome the problems and limitations encountered in real-time
>interactive system development. Readers find out how to build tools
>for NVE technology, learn the underlying principles and practical
>techniques for making successful NVEs, and glimpse future direc-tions
>in this growing field.
>
>The authors plan an associated web site that will show how to build
>a usable networked virtual environment.
>
>“I love this book. It is technical, comprehensive, well-written, and
>deals with an area which is gaining popularity rapidly, both in
>academia and industry.”
>—Dr. Gurminder Singh, Director (Learning Lab), Kent Ridge Digital Labs,
>Singapore
>
>“[The] first comprehensive reference on the subject of Networked
>VR. Very readable, with a mix of the history and the science on
>the topic.”
>—Michael Macedonia, Chief Scientist and Technical Director, US Army
>Simulation, Training, and Instrumentation Command (STRICOM)
>
>“The book provides a good road map of the field, pointing to
>recent, relevant research in the field ... it identifies all the critical
>topic areas and provides pointers for in-depth study.”
>—Katherine L. Morse, Senior Computer Scientist, Science Applications
>International Corporation
>
>
>all the best, Don
>--
>Don Brutzman Naval Postgraduate School, Code UW/Br Root 200 work
831.656.2149
> Monterey California 93943-5000 USA fax
831.656.3679
>Virtual worlds/underwater robots/Internet
http://www.stl.nps.navy.mil/~brutzman
>-------------------------------------------------------------------------
>for list subscription instructions,
>send email to www-vrml-request@web3d.org with text "info"
>
>

++ The Contact Consortium ++
A Forum for Contact, Culture and Community in Digital Space
Visit us at: http://www.ccon.org
or reach us at the Consortium by email at: via our Webmaster
/- Living, Learning and Working in Virtual Worlds -/

Hello,
I have subscribed to OWorlds list. My interests are mostly on
simulation. I made a simple linear accleration simulation example some
years ago, but the platform broke it. In the interim I have been
researching simulation methods as well as 3D technologies in general. I
have a skeletal document set of an API for simulation, and hope that as
I can finish and expand it it will become a general framework for
simulation. For now, it is nothing but a set of ideas. Here is a link
to what information on this project is available on the web:
http://www.tomco.net/~raf/sim/
This is a personal endeavor but one I hope to see established over the
years as computational power becomes sufficient and I learn enough to
implement it as it should be done.
I have done some small work with Avatars and am bitter about Universal
Avatars. I am angry at Moses Ma for no particular reason except he had
all these offers and then disappeared.
I was going to post this to OWorlds maybe I will.
Ross A. Finlayson

<html>
Hi Ross, you should post this directly to the list (yes it is of interest) by sending email to.. 
oworld@lists.best.com 
 
You might want to format your post with our original framework: 
 
Hello OWorld, my name Bruce Damer, moderator of this list (actually a 
pretty silent moderator). I think we have enough folks signed up (over 40 
now) so it is time to make introductions. I think we can all stick to a 
pretty simple formula: 
 
1. Where do I come from? 
2. Why did I join OWorld (ie: why do I believe in open community efforts to 
create virtual worlds and platforms) 
3. What am I working on/would like to work on 
4. Controversial statement of the day 
 
1. Where do I come from? 
****************************** 
I came from a background of ten years building graphical OS environments 
for Xerox and others, having worked on some of the original windows and 
icons stuff based on the STAR system (which the Mac and Windows copied). 
See <a href="http://www.damer.com/background/cv.htm" eudora="autourl">http://www.damer.com/background/cv.htm</a> for details. I got tired of the 
2D page-based metaphors of interaction and in the early 1990s left a 
software project I was helping direct in Prague to retun to live in 
Northern California and plug into the then emergent visual Internet. I 
learned in MUDs and read Snow Crash and dreamed of a visual community 
metaphor that would punch through the documents of the web and make 
cyberspace into a true place. So I co-founded the Contact Consortium in 
1995 with anthropologist Jim Funaro. The Consortium is the rather organic 
virtual organization that provides OWorld a home and seeks to catalyze the 
development of shared visual virtual worlds in cyberspace. We have held 5 
conferences and a gazillion other events around the Earth and built many 
experimental worlds in four years. The Consortium membership is large and 
very diverse and knows one heck of a lot about multi user worlds and what 
keeps users coming back. See <a href="http://www.ccon.org/" eudora="autourl">http://www.ccon.org</a> for the Consortium. I also 
run a small company that builds worlds for real customers, see 
<a href="http://www.digitalspace.com/" eudora="autourl">http://www.digitalspace.com</a>. 
 
2. Why did I join OWorld (ie: why do I believe in open community efforts to 
create virtual worlds and platforms) 
****************************** 
Well at the 2nd annual "Digital Biota" conference (Contact Consortium and 
Cyberlife) in Cambridge UK in sept 1998, there seemed to be a great brain 
trust of folks who were passionate about bringing biological metaphors into 
cyberspace. The OWorld list and SIG were born at that conference. I would 
like to see worlds where biological processes can express themselves, like 
Tierra, Nerve Garden and Karl Sim's evolving creatures (see 
<a href="http://www.biota.org/" eudora="autourl">http://www.biota.org</a>). Heck I want to live to see a "cambrian explosion in 
cyberspace". These worlds have to be inhabited by users to provide the 
stimulus for the organisms. 
 
3. What am I working on/would like to work on 
****************************** 
Well I am not a programmer anymore, but can be a good ringleader. 
 
4. Controversial statement of the day 
****************************** 
Can open source platform efforts work, or will they produce bloated kludges 
that are Xanadu-ed into long delayed nonexistance? Java or ActiveX.. is 
Windows the only client side platform that matters now? 
Hey, I also wrote a book on avatars and virtual worlds too! Here is a 
shameless plug for it.. find ALL the virtual worlds you could ever want to 
download at: 
<a href="http://www.digitalspace.com/avatars" eudora="autourl">http://www.digitalspace.com/avatars 
</a>and YOU are up next.. 
Bruce Damer 
(DigiGardener) 
 
 
At 11:53 PM 3/3/99 -0500, you wrote: 
>Hello, 
> 
>I have subscribed to OWorlds list.  My interests are mostly on 
>simulation. I made a simple linear accleration simulation example some 
>years ago, but the platform broke it.  In the interim I have been 
>researching simulation methods as well as 3D technologies in general.  I 
>have a skeletal document set of an API for simulation, and hope that as 
>I can finish and expand it it will become a general framework for 
>simulation.  For now, it is nothing but a set of ideas.  Here is a link 
>to what information on this project is available on the web: 
> 
><a href="http://www.tomco.net/~raf/sim/" eudora="autourl">http://www.tomco.net/~raf/sim/</a> 
> 
>This is a personal endeavor but one I hope to see established over the 
>years as computational power becomes sufficient and I learn enough to 
>implement it as it should be done. 
> 
>I have done some small work with Avatars and am bitter about Universal 
>Avatars.  I am angry at Moses Ma for no particular reason except he had 
>all these offers and then disappeared. 
> 
>I was going to post this to OWorlds maybe I will. 
> 
>Ross A. Finlayson 
> 
</html>
Hi Bruce,
OK it will take a while. I don't want to seem like I just subscribed and sent it
which I would have done if I sent it immediately.
I hope you take this as a joke, the name "Damer" is spelled differently than the
serial killer but that is the easiest connection. I say this in all respect of
your heritage and name. Perhaps that gives some insight into me as I would think
of serial killers.
About the Simulation API, I think it has some good concepts. The most work I had
done on it is simulating Newtonian mechanics, it's still nowhere near any kind of
workability. It seems logical to me that a rule and event based framework for
general simulation would be able to simulate about anything. Hopefully at some
point there will be learning neural net inputs to it so the simulation could
simply read world data and then simulate it. This is more fantasy than fact at
this point.
I am only one person and I hope to avoid overextending myself because in the past
it has overextended me. I worry sometimes that people will say stuff when I am
not there more freely than if I am there.
All work on my web site or with my name on it is solely done by me. Finlayson
Consulting is me. There is another Ross Finlayson,
http://xenon.stanford.edu/~rsf/, who is not me but he is very smart.
I have been frustrated many times throughout the VRML process because things
happen that are not logical. I guess that is what I say about that.
Have a nice day, bonne chance,
Ross A. Finlayson
Bruce Damer wrote:
> Hi Ross, you should post this directly to the list (yes it is of interest) by
> sending email to..
> oworld@lists.best.com
>
> You might want to format your post with our original framework:
>
> Hello OWorld, my name Bruce Damer, moderator of this list (actually a
> pretty silent moderator). I think we have enough folks signed up (over 40
> now) so it is time to make introductions. I think we can all stick to a
> pretty simple formula:
>
> 1. Where do I come from?
> 2. Why did I join OWorld (ie: why do I believe in open community efforts to
> create virtual worlds and platforms)
> 3. What am I working on/would like to work on
> 4. Controversial statement of the day
>
> 1. Where do I come from?
> ******************************
> I came from a background of ten years building graphical OS environments
> for Xerox and others, having worked on some of the original windows and
> icons stuff based on the STAR system (which the Mac and Windows copied).
> See http://www.damer.com/background/cv.htm for details. I got tired of the
> 2D page-based metaphors of interaction and in the early 1990s left a
> software project I was helping direct in Prague to retun to live in
> Northern California and plug into the then emergent visual Internet. I
> learned in MUDs and read Snow Crash and dreamed of a visual community
> metaphor that would punch through the documents of the web and make
> cyberspace into a true place. So I co-founded the Contact Consortium in
> 1995 with anthropologist Jim Funaro. The Consortium is the rather organic
> virtual organization that provides OWorld a home and seeks to catalyze the
> development of shared visual virtual worlds in cyberspace. We have held 5
> conferences and a gazillion other events around the Earth and built many
> experimental worlds in four years. The Consortium membership is large and
> very diverse and knows one heck of a lot about multi user worlds and what
> keeps users coming back. See http://www.ccon.org for the Consortium. I also
> run a small company that builds worlds for real customers, see
> http://www.digitalspace.com.
>
> 2. Why did I join OWorld (ie: why do I believe in open community efforts to
> create virtual worlds and platforms)
> ******************************
> Well at the 2nd annual "Digital Biota" conference (Contact Consortium and
> Cyberlife) in Cambridge UK in sept 1998, there seemed to be a great brain
> trust of folks who were passionate about bringing biological metaphors into
> cyberspace. The OWorld list and SIG were born at that conference. I would
> like to see worlds where biological processes can express themselves, like
> Tierra, Nerve Garden and Karl Sim's evolving creatures (see
> http://www.biota.org). Heck I want to live to see a "cambrian explosion in
> cyberspace". These worlds have to be inhabited by users to provide the
> stimulus for the organisms.
>
> 3. What am I working on/would like to work on
> ******************************
> Well I am not a programmer anymore, but can be a good ringleader.
>
> 4. Controversial statement of the day
> ******************************
> Can open source platform efforts work, or will they produce bloated kludges
> that are Xanadu-ed into long delayed nonexistance? Java or ActiveX.. is
> Windows the only client side platform that matters now?
> Hey, I also wrote a book on avatars and virtual worlds too! Here is a
> shameless plug for it.. find ALL the virtual worlds you could ever want to
> download at:
> http://www.digitalspace.com/avatars
> and YOU are up next..
> Bruce Damer
> (DigiGardener)
>
> At 11:53 PM 3/3/99 -0500, you wrote:
> >Hello,
> >
> >I have subscribed to OWorlds list. My interests are mostly on
> >simulation. I made a simple linear accleration simulation example some
> >years ago, but the platform broke it. In the interim I have been
> >researching simulation methods as well as 3D technologies in general. I
> >have a skeletal document set of an API for simulation, and hope that as
> >I can finish and expand it it will become a general framework for
> >simulation. For now, it is nothing but a set of ideas. Here is a link
> >to what information on this project is available on the web:
> >
> >http://www.tomco.net/~raf/sim/
> >
> >This is a personal endeavor but one I hope to see established over the
> >years as computational power becomes sufficient and I learn enough to
> >implement it as it should be done.
> >
> >I have done some small work with Avatars and am bitter about Universal
> >Avatars. I am angry at Moses Ma for no particular reason except he had
> >all these offers and then disappeared.
> >
> >I was going to post this to OWorlds maybe I will.
> >
> >Ross A. Finlayson
> >
>
> DigitalSpace Corporation
> 343 Soquel Avenue
> Suite 70
> Santa Cruz CA 95062 USA

At 11:20 AM 3/3/99 -0800, you wrote:
>Gerald de Jong writes:
> "the only way to create a truly dynamic and compelling virtual world
> scenario is through extensive and effective use of MOBILE CODE."
Mark Shirley writes:
>Gerald,
> Could you say more about why you think mobile code is essential?
the core reasoning is embarrassingly simple: when data moves around
the net the behavior is wholly dependent on the software used to
realize or render that data. when objects (data and code) move around
the net, behavior flows around and the realization engine need not
be as many-faceted since it no longer is the bottleneck. it allows for
simplicity in the realization engine, and simplicity is a survival trait in
the domain of software.
>1. Support for world objects to change structure and behavior over
> time
think also of behavior that is dependent on stimuli. there are any
number of ways of implementing this, and any solution will probably
involve several at once. it's of the utmost important that the mobile
code be of the "general-purpose" variety so that anything can be
implemented (scripting, neural-nets, things we haven't thought of
yet) on top of it. that's why Java is the best candidate.
>2. A distributed implementation (including client<->server
> distribution as well as server<->server)
a truly forward-looking virtual world implementation will involve
abandoning the notions of client and server and replacing them
with "participant" or something. a participant is always both client
and server.
>3. Some distinction between system-level programming and end-user
> programming
a clear distinction! the realization mechanism will necessarily span
the spectrum from fast native code (C/Assembler) to actual hardware
solutions. there need not even be a hugely wide "API" connecting this
to the "user programming" which should be done using general purpose
code that is mobile.
>For me, it's #1 and #2 that imply a need for mobile code.
i include #3 as well.
>A key need in a multiuser virtual world is the ability to modify
>the world on-the-fly, without bringing it down.
everything one encounters should effectively be capable of
modifying the world in the sense of introducing wholly novel
behaviors.
>A second key need is that it must be convenient for people to
>build on objects and behaviors created by others, and that the
>environment support the kinds of problems that occur over time
>when this is done.
versioning is an issue, as well as clarity with respect to the origin
of the mobile code. each participant should be able to introduce
new behaviors.
Java RMI provides for the transportation of objects along with
the URL from which their code originates. versioning is really
then only a concern in terms of individual code sources.
>MUDs generally make a distinction between system-level programming
>and end user world scripting. The deep roots of the world are
>done in some conventional (and usually strongly typed) programming
>language, the end users use an untyped scripting language with
>small scripts attached to objects. Any script can be changed at
>any time.
i'm making a plea for using a strongly typed language as the basis of
scripting, with a variety of untyped scripting strategies implemented
on top of this. the means of scripting should not be coded into the
underlying plaform but rather isolated from it through a general-
purpose mobile code mechanism, namely Java, with its inherent
security model.
>Some MUDs have a 2 level structure: system-level programming is
>done in C by one or a handful of people, and end-user scripting is
>done by everybody. Some MUDs have a 3 level structure. The core
>system is handled by one or a few people. A larger group of
>people write extensions, still in a conventional language, and
>then end-users write in the scripting language. I would argue a 3
>level structure for an oworld.
yes. there's the realization engine, the mobile code layer, and the
scripting layer above. much scripting can be done visually then
(ie without writing code) which invites a new and creative audience
of people not so hip on the idea of "coding".
>The server has to be scalable.
i believe that everyone should be client and server, with the
opportunity for heftier participants who have big iron and big
pipes capable of hosting lots of other participants.
the notions of "the server" and "the client" don't have the
longevity that i'm looking for. in a few years we'll think
differently, and i want to anticipate that.
>The simplest way to do this is for people to copy the code from
>one place to another. This is a start, but it's lame over the
>long run.
yes, it should not involve any effort. it must be spontaneous.
>Why not just go for the endpoint of this progression and build on
>a fully distributed, persistent object system?
bingo. but you have to further explore what is implied by
"fully distributed".
>Basically, I'd like to think of the client and the server as being
>built on the same foundation. It's just that clients have GUI
>code loaded while servers don't, and servers may have code loaded
>for talking to persistent object stores, databases and other
>things while clients probably won't.
clearly we're thinking in the same general directions, which is
good. we probably have to move to some newer metaphors like
those found in something like Jini: the things with no GUI are
really just "persistency services" (hard disks?).
>Anyway, those are the thoughts your controversial question kicked
>up. Comments?
glad you decided to pursue the issue.
---
Gerald de Jong, Beautiful Code B.V.
Rotterdam, The Netherlands, Tel. +31655893940
http://www.beautifulcode.nl
1. Where do I (Alastair Channon) come from?
******************************
After a dull year as a support programmer, followed by a Cambridge
Mathematics degree, I was convinced that I should direct my future
towards something far more interesting. I'd had a long-running (but
shallow: Eliza, AI-games, L-Systems, etc.) interest in AI and ALife
and found the idea of artificial evolution fascinating. So I took
one of the excellent M.Sc. courses at COGS, Sussex University (UK).
It was called KBS (Knowledge Based Systems) but many of us chose mainly
ALife-related options, which were being developed for the current EASY
(Evolutionary and Adaptive SYstems) M.Sc.. My dissertation title
was ``The Evolutionary Emergence route to Artificial Intelligence''.
My long-term aim was and is to use natural selection (as opposed to
artificial selection, which most GA/GP/... work uses despite claims of
natural selection) to generate open-ended evolution, as a route AI-Life.
Thus my work is related to systems such as Tierra and PolyWorld. (Yes,
Digital Biota II was an excellent experience - thanks Steve, Bruce and
others.) I'm currently in the final year of my Ph.D. at Southampton
University (UK) and planning to continue with the same research route,
although I have not yet planned where.
Should anyone wish, there is more on my work at www.soton.ac.uk/~adc96r
2. Why did I join OWorld (ie: why do I believe in open community efforts to
create virtual worlds and platforms)
******************************
I now believe in taking a behavioural route to spotting emergent phenomena
within evolving artificial worlds. It is simply to hard to work out what
an organism's emergent behaviours are by looking at its inner workings,
not least because you don't know what you are looking for. At least
that's what I've found in my work, where the inner workings are artificial
neural messworks (networks that work but I cannot tell how). Karl Sim's
block-creatures work (albeit artificial-selection based) shows how easy it
can be for a human to interpret evolutionary emergent behaviours (eg. the
hockey-player) when they are embedded within a 3D virtual world. Hence I'm
interested in all virtual world ideas that might be of use in such a
venture, including block-creatures, Demetri Terzopoulos' artificial
fish, Gerald de Jong's Struck, and anything I may hear about on this
list. Oh, and I do believe in open community efforts to create virtual
worlds and platforms, but that's not the main reason I joined.
3. What am I working on/would like to work on
******************************
See 2. I know exactly what I want to do. All I need now is somewhere to
do it (ie. finance).
4. Controversial statement of the day
******************************
The main thing that I see missing from virtual worlds is autonomous
evolution, ie. artificial organisms evolving not in the sense of being
developed by humans but by interaction with each other. For Bruce's
"cambrian explosion in cyberspace" to take place, a good deal of evolution
will be needed. Human interaction alone will be totally (completely
and utterly ... more superlatives ...) insufficient. If this is to be
realised, then there are implications for the design of future worlds.
Best wishes all,
Alastair.
--
Alastair Channon
Image, Speech & Intelligent Systems (ISIS) Research Group (Building 1)
Department of Electronics & Computer Science
University of Southampton, SO17 1BJ. United Kingdom
http://www.soton.ac.uk/~adc96r/
Bruce,
(Sorry if I missed this.) Are these messages getting archived
somewhere? I don't see mention of this on the web page
(http://www.ccon.org/lists/oworld.html).
Here are some suggestions for some initial topics we might talk about:
1. What do we want to do with our virtual worlds? What dreams or
visions do we have? What are the broad categories of
technologies that are going to have to be developed and fit
together? Where do we want to go?
2. What do people like about existing virtual worlds? What do
people feel is insufficient and can be improved upon? The idea
is to take advantage of people's experience in compiling
suggested do's and don'ts for any actual work we might do.
3. What specific steps can we take to start constructing an open
source virtual world infrastructure? Or more than one? Who
wants to help? What existing code is good and might be built
on top of? What language(s)? How can the task be decomposed so
that folks have pieces to work on?
- Mark

<x-html>
<!doctype html public "-//w3c//dtd html 4.0 transitional//en">
<html>
Mark -
I'm not Bruce but I''ll bite.
  
"Mark H. Shirley" wrote:
[snip]
<blockquote TYPE=CITE>
<pre>3. What specific steps can we take to start constructing an open
     source virtual world infrastructure?</pre>
</blockquote>
 Infrastructure -- At the risk of stating the obvious it has to have
a stable core, and yet be open and extensible. (The problem seems to me
to be more a problem in Operating System design than Application architecture.
)
IMHO, the guiding principle of developing cyberspace should be one of
dynamic evolution not rigid specification (except at the kernel).
The growth of such a system could be modelled on A-Life concepts. Code
as DNA, the most suitable implementations win. Easier said than done of
course ;^)
Also, and very importantly v-world infrastructure needs to be platform
and application agnostic. To wit, It's unlikely that anything as universal
virtual reality will be created by any one company or group of individuals.
Opening up the application and platform space to the galaxy best ensures
the kind of development that resulted in such universally applicable technologies
as the Internet itself, TCP/IP and Linux.
  
<blockquote TYPE=CITE>
<pre>Or more than one?</pre>
</blockquote>
 As to the number of infrastructures, I have give some thought to
that and the problem I see with more than one public infrastructure is
not competition but bifurcation and the inevitable need to bring the standards
back together if more than one is generally adopted. On the other hand,
it would be very interesting to be able to test out different infrastructure
approaches and then adopt the best of the bunch (or scrap em all and start
over).
  
<blockquote TYPE=CITE>
<pre>Who wants to help?</pre>
</blockquote>
 I know I do. I think that if the charter and underlying philosophy
of the organization was compelling enough and well enough advertised you'd
have a lot of takers (not necessarily a good thing, tho).
Anyway, just a few thoughts.
  
  
  </html>
</x-html>
Attachment Converted: "d:\eudora\attach\agern.vcf"
Andrew Gerngross wrote:
> Mark -
> I'm not Bruce but I''ll bite.
Oops. I meant the question about archiving for Bruce.
I didn't mean to limit the rest of the questions to him 8^)
- Mark

I've just joined this list, so if I revisit well trodden ground then
please excuse me...
>"Mark H. Shirley" wrote:
>> 3. What specific steps can we take to start constructing an open
>> source virtual world infrastructure?
>
>Infrastructure -- At the risk of stating the obvious it has to have a
>stable core, and yet be open and extensible. (The problem seems to me to
>be more a problem in Operating System design than Application
>architecture. )
>
>IMHO, the guiding principle of developing cyberspace should be one of
>dynamic evolution not rigid specification (except at the kernel).
Do you mean dynamic evolution of the infrastructure or of the
content?
>Also, and very importantly v-world infrastructure needs to be platform and
>application agnostic. To wit, It's unlikely that anything as universal
>virtual reality will be created by any one company or group of
>individuals. Opening up the application and platform space to the galaxy
>best ensures the kind of development that resulted in such universally
>applicable technologies as the Internet itself, TCP/IP and Linux.
Platform independent is one thing. Application agnostic is far more
tricky. As a general rule of thumb, anything generic will be less
efficient than something (even slightly) more specific.
Infrastructure is important, but if you want to achieve something in
a reasonable timescale you should make it application-driven. Not
just one application, of course, the more the merrier, but at some point
the design will have to make a compromise between genericity and
efficiency.
>> Or more than one?
>
>As to the number of infrastructures, I have give some thought to that and
>the problem I see with more than one public infrastructure is not
>competition but bifurcation and the inevitable need to bring the standards
>back together if more than one is generally adopted. On the other hand, it
>would be very interesting to be able to test out different infrastructure
>approaches and then adopt the best of the bunch (or scrap em all and start
>over).
This approach is effectively what has happened in academic DVE research
over the years. Unfortunately, given the nature of this kind of research,
these systems rarely see the light of day. There's a wealth of
information about these past attempts, it would be wise to consult them
for a good head start. If the best ideas/concepts can be extracted then
it makes sense to work as a whole on a single infrastructure given their
inevitable complexity.
Rych
===================================================================
Rycharde Hawkes Tel/Voice: +44 117 92 28487
HP Labs, Bristol Fax: +44 117 92 28003
===================================================================

Hi
I'm Jane Prophet.
> 1. Where do I come from?
> ******************************
I'm a British artist, living in London, I makes installation, web sites
and
cdroms. Interested in space, real and virtual, and performance. Started
TechnoSphere with Gordon Selley in 1995 (design an alife creature on the
web
and get email which tell you of it's activities). Rych Hawkes who has
already
posted here wrote the current WWW alife engine for TechnoSphere. When
he's not
at HP, Rych and I are currently designing a sassy Bot called Chrissy for
a new
art piece that I'm working on. I work with programmers, scriptwriters,
composers etc and so I practice the open community approach to working.
In
keeping with this TechnoSphere has been developed by welcoming input
from a
wide range of people.
> 2. Why did I join OWorld (ie: why do I believe in open community efforts to
> create virtual worlds and platforms)
> ******************************
Because I'd like to be part of making an online alife inhabited 3D World
as
part of developing TechnoSphere. I met Sue Wilcox and Bruce Damer in
Banff
last summer and then a whole bunch of alife heroes at DB2, loved the
atmosphere
and the networking and sharing of ideas. The beer drinking wasn't bad
either,
and for me the meat and bone people and social life behind the virtual
community are most important. The open exchange of ideas was most
refreshing, I
left with an aching, stimulated brain and some new friends.
> 3. What am I working on/would like to work on
> ******************************
We install a real time 3D version of TechnoSphere at the National Museum
of
Photography, Film and TV in Bradford, UK in a couple of weeks, it'll be
there
for 5 years. Planning an online world and a PC game. TechnoSphere is
driven by
Gordon and my focus on visual appearance (he started out in design) and
now
with the Museum piece, 3D audio.
We want to make a virtual world that looks gorgeous and is populated by
non-organic fantastic looking beasties using Gordon's new alife engine
which
uses a complex energy model based on carbon organisms. Humour is
essential (not
only to collaborative working) but within TechnnoSphere and the
behaviour of
the creatures.
Going to be at UCLA for April, May and June so maybe I can meet some of
you
(again) face to face then.
>
>
> 4. Controversial statement of the day
> ******************************
Does the O in OWorld stand for Orgasm?
Sue's points about the possible convergence of diverse groups from
military to
porn struck a chord with me. Especially as the majority of sex workers
are
women, and while DB3 was a really great event, women were very much in
the
minority.
I use a Macintosh everyday.
-------------------------
www.technosphere.org.uk
www.ucl.ac.uk/slade/swarm

Just a thought - might it be an idea to try and use the platform
independence of Java?
Additionally the Object-Oriented nature makes it rather good for
cooperative development.
A nice task for someone (less busy than myself, naturally...) would be
to see what is available already as Open Source, and to collate sites
containing (eg.) the VRML spec.
Danny.
--
Intermittent Web site :
http://members.xoom.com/dayers
or
http://www.danja.free-online.co.uk

Alternate email :
Danny_Ayers@yahoo.com

Rycharde Hawkes wrote:
> >IMHO, the guiding principle of developing cyberspace should be one of
> >dynamic evolution not rigid specification (except at the kernel).
>
> Do you mean dynamic evolution of the infrastructure or of the
> content?
(Not trying to speak for Gerald) I think this means both. For me,
however, the near-term emphasis is on evolution of the
infrastructure. Dynamic evolution of the content is absolutely
basic and critical. Supporting dynamic evolution of the
infrastructure is also critical over the long term, but much
easier to lose among engineering compromises.
ActiveWorlds is a good example both of getting this very right and
a little wrong, in my opinion. (If I'm mischaracterizing AW,
someone please correct me.) ActiveWorlds supports building by
normal users and the worlds' content seems to be constantly
growing and changing. The infrastructure (server and client) can
be modified by the activeworlds programmers and changes propagated
without the users having to do anything. All of this is great.
The limitation is that no one, outside the small group of activeworlds
programmers, can add fundamentally new kinds of objects. (Their
bot api will fix this somewhat, but not to the degree I'm hoping for.) I think
lots of people having the capability to do this, and on their
running worlds, will be important to the long term evolution of
this stuff.
Gerald de Jong wrote:
> On the other hand, it
> >would be very interesting to be able to test out different infrastructure
> >approaches and then adopt the best of the bunch (or scrap em all and start
> >over).
>
> This approach is effectively what has happened in academic DVE research
> over the years. Unfortunately, given the nature of this kind of research,
> these systems rarely see the light of day. There's a wealth of
> information about these past attempts, it would be wise to consult them
> for a good head start.
I agree completely. Can you point us at some specific projects?
Do you know of any "lessons learned" papers, newsgroup discussions
or email? (I've run across some too, and I'll go back and look for them.)
- Mark

At 09:13 AM 3/5/99 -0800, Mark Shirley wrote:
> Gerald de Jong wrote:
sorry, but i didn't write anything here. (i don't use caps).
>
>> On the other hand, it
>> >would be very interesting to be able to test out different infrastructure
>> >approaches and then adopt the best of the bunch (or scrap em all and start
>> >over).
>>
>> This approach is effectively what has happened in academic DVE research
>> over the years. Unfortunately, given the nature of this kind of research,
>> these systems rarely see the light of day. There's a wealth of
>> information about these past attempts, it would be wise to consult them
>> for a good head start.
>
>I agree completely. Can you point us at some specific projects?
>Do you know of any "lessons learned" papers, newsgroup discussions
>or email? (I've run across some too, and I'll go back and look for them.)
> - Mark

---
Gerald de Jong, Beautiful Code B.V.
Rotterdam, The Netherlands, Tel. +31655893940
http://www.beautifulcode.nl

Danny Ayers wrote:
> Just a thought - might it be an idea to try and use the platform
> independence of Java?
Agree completely. Just came back from VRML99 where I saw the Java3D VRML
loader in action. It looks like the best approach to developing a platform
independent browser. Also, the loader architecture is quite open and
already fairly extensively used.

> Additionally the Object-Oriented nature makes it rather good for
> cooperative development.
Agree completely. Also, Java's class loader and typing allow for safe
modularity, reuse and cooperation.
>
> A nice task for someone (less busy than myself, naturally...) would be
> to see what is available already as Open Source, and to collate sites
> containing (eg.) the VRML spec.
I believe java3D and the VRML loader are open source. At the very least
the source is available for both.
-Andrew Gerngross
Attachment Converted: "d:\eudora\attach\agern3.vcf"

Rycharde Hawkes wrote:
> I've just joined this list, so if I revisit well trodden ground then
> please excuse me...
Ditto for me...
>
> >IMHO, the guiding principle of developing cyberspace should be one of
> >dynamic evolution not rigid specification (except at the kernel).
>
> Do you mean dynamic evolution of the infrastructure or of the
> content?
>
Mostly, I mean the content. However, some "content" in a system that encourages
wide reuse naturally becomes canonical. Such content could be adopted into the
core infrastructure.
The more stable the infrastructure the better. The infrastructure should, IMHO,
be open (to a core development team) and dynamic at first and through testing,
experience and iterative specification and implementaion gain stability.
This naturally brings to mind the question of what is meant by infrastructure
in a large and (potentially) dynamic system. Has anyone considered establishing
a policy stating what defines the infrastructure for oworlds?

> >Also, and very importantly v-world infrastructure needs to be platform and
> >application agnostic. To wit, It's unlikely that anything as universal
> >virtual reality will be created by any one company or group of
> >individuals. Opening up the application and platform space to the galaxy
> >best ensures the kind of development that resulted in such universally
> >applicable technologies as the Internet itself, TCP/IP and Linux.
>
> Platform independent is one thing. Application agnostic is far more
> tricky. As a general rule of thumb, anything generic will be less
> efficient than something (even slightly) more specific.
> Infrastructure is important, but if you want to achieve something in
> a reasonable timescale you should make it application-driven. Not
> just one application, of course, the more the merrier, but at some point
> the design will have to make a compromise between genericity and
> efficiency.
Re genericity and speed: true enough. A well-designed finely grained component
based architecture, however, could allow for application-specific enhancements.
Do you agree? Despite many years of developing OO software I do not know of any
large scale efforts where this ideal has come true. Small scale efforts
including some of my own have acheived this ideal in part.

>
> >> Or more than one?
> >
> >As to the number of infrastructures, I have give some thought to that and
> >the problem I see with more than one public infrastructure is not
> >competition but bifurcation and the inevitable need to bring the standards
> >back together if more than one is generally adopted. On the other hand, it
> >would be very interesting to be able to test out different infrastructure
> >approaches and then adopt the best of the bunch (or scrap em all and start
> >over).
>
> This approach is effectively what has happened in academic DVE research
> over the years. Unfortunately, given the nature of this kind of research,
> these systems rarely see the light of day. There's a wealth of
> information about these past attempts, it would be wise to consult them
> for a good head start. If the best ideas/concepts can be extracted then
> it makes sense to work as a whole on a single infrastructure given their
> inevitable complexity.
>
I have very little real experience with competing infrastructures. My thoughts
stemmed from idle ruminations and corollary but not identical experiences. I
agree wholeheartedly that the more is known about similiar attempts the better
-- do you (or anyone else reading this for that matter) know anyone who has
experience in developing large system infrastructures (successfully or
otherwise) who could add to this discussion?

Attachment Converted: "d:\eudora\attach\agern1.vcf"
Gerald de Jong wrote:
> a truly forward-looking virtual world implementation will involve
> abandoning the notions of client and server and replacing them
> with "participant" or something. a participant is always both client
> and server.
>
Does this means that a multi-user server would not be necessary for
running multi-user worlds? One thing which really annoys me with
multi-user worlds is this dependance to a professional server. even if
the server is free, it's mostly impossible to run it for a hobbyist, who
has only access to web space provided by his ISP, but has no acccess to
any server side programming. How do you imagine a multi-user world
without multi-user server ? Would it be possible ? easy to create ?
Remi
Holodesk contains a combined client/server package and uses a connection
service to find other users on the net instead of requiring a central world
server. This strength is also a weakness in that we don't support more than
10 users in the same world currently. But, we are standard VRML and
implement a Living-Worlds style interface.
We're in beta, download and try us out at: http://www.holodesk.com
JEM
> -----Original Message-----
> From: Remi Sussan [SMTP:rsussan@geocities.com]
> Sent: Friday, March 05, 1999 2:52 PM
> To: oworld@lists.best.com
> Subject: oworld: abandoning server
>
> Gerald de Jong wrote:
>
> > a truly forward-looking virtual world implementation will involve
> > abandoning the notions of client and server and replacing them
> > with "participant" or something. a participant is always both client
> > and server.
> >
> Does this means that a multi-user server would not be necessary for
> running multi-user worlds? One thing which really annoys me with
> multi-user worlds is this dependance to a professional server. even if
> the server is free, it's mostly impossible to run it for a hobbyist, who
> has only access to web space provided by his ISP, but has no acccess to
> any server side programming. How do you imagine a multi-user world
> without multi-user server ? Would it be possible ? easy to create ?
>
> Remi
At 08:52 PM 3/5/99 +0100, you wrote:
>Gerald de Jong wrote:
>> a truly forward-looking virtual world implementation will involve
>> abandoning the notions of client and server and replacing them
>> with "participant" or something. a participant is always both client
>> and server.
Remi:
>Does this means that a multi-user server would not be necessary for
>running multi-user worlds?
the way i see it, if i want to visit your world, i actually go into your
computer. this is dangerous without an excellent security mechanism,
of course. the current Java 2 platform has just such a thing.
call it a Javatar. :)
---
Gerald de Jong, Beautiful Code B.V.
Rotterdam, The Netherlands, Tel. +31655893940
http://www.beautifulcode.nl

"Mark H. Shirley" wrote:
> Rycharde Hawkes wrote:
>
> > >IMHO, the guiding principle of developing cyberspace should be one of
> > >dynamic evolution not rigid specification (except at the kernel).
> >
> > Do you mean dynamic evolution of the infrastructure or of the
> > content?
>
> (Not trying to speak for Gerald) I think this means both. For me,
> however, the near-term emphasis is on evolution of the
> infrastructure. Dynamic evolution of the content is absolutely
> basic and critical. Supporting dynamic evolution of the
> infrastructure is also critical over the long term, but much
> easier to lose among engineering compromises.
Actually, that would be me, Andrew not Gerald you'd be speaking for (but thanks for
the compliment :^) ).
I agree that dynamic evolution of the content is critical. Reality, at least as I
remember it, wouldn't be reality without dynamism. The infrastrucure or core could
conceivably benefit from an evolutionary development approach. But don't you agree
that core stability (with modularity albeit) is equal in importance to the
evolutionary dynamism of the content? It seems that if the core were to evolve with
the same rapidity as the content that stability would be sacrificed.
>
-Andrew Gerngross
Attachment Converted: "d:\eudora\attach\agern2.vcf"
Gerald said (this time, I've got it right!):
the way i see it, if i want to visit your world, i actually go into your
computer. this is dangerous without an excellent security mechanism, of
course. the current Java 2 platform has just such a thing.
I see this slightly differently. I think that, when several
people are together in a world, their computers should work
together to create the illusion of a shared world for them. Which
computer does which work should be negotiated between those
computers: some might be done on my computer, some might be done
on yours and some might be done on a dedicated world server. How
that negotiation comes out might depend upon the relative power of
the computers involved and whether there's a dedicated world
server involved at all, maybe its current load, and maybe things
like restrictions on code migration to personal computers
established by the individuals involved.
I guess I'd like the infrastructure to be a soup of objects, some
of which have special access to limited resources like displays
and keyboards, and maybe persistent stores, and network access.
How these objects are mapped to the processors available should
be, I think, completely flexible, and we would need to work out
policies for that mapping.
So I also don't think that a multi-user server is strictly
necessary for running multi-user worlds. Eric Mason just pointed
out an example (Holodesk) without a centralized server. I think
we should strive for an infrastructure that can be used that way
as well as the way, say, ActiveWorlds is used now, and any
variation in between.
- Mark

Remi wrote:
> One thing which really annoys me with
>multi-user worlds is this dependance to a professional server. even if
>the server is free, it's mostly impossible to run it for a hobbyist, who
>has only access to web space provided by his ISP, but has no acccess to
>any server side programming. How do you imagine a multi-user world
>without multi-user server ? Would it be possible ? easy to create ?
Yes a serverless framework could be possible,
but there is a scalabitlity / bandwidth problem if the clients are directly
communicating
peer to peer.
There are currently many existing ways to make worlds multi user,
e.g. ask some body running a vnet host for example, to host your world.
Probably its also possible for Active Worlds, DeepMatrix etc.
For blaxxun its quite simple to make your world temporarily multi-user:
Goto to
http://www.blaxxun.com/cgi-bin/frameset/frameset.pl
and enter the URL of your VRML world. Many people are using this feature.
On www.colonycity.com you can now submit your own VRML world for MU-hosting
and/or you can create your own club.
-- Holger
----------------------------------------------------------------------
Holger Grahn holger@blaxxun.com
blaxxun interactive http://www.blaxxun.com

Andrew Gerngross wrote:
"Mark H. Shirley" wrote:
Actually, that would be me, Andrew not Gerald you'd be speaking for (but thanks for
the compliment :^) ).
Apologies to both of you. I'm chagrined.
I agree that dynamic evolution of the content is critical. Reality, at least
as I remember it, wouldn't be reality without dynamism. The infrastrucure
or core could conceivably benefit from an evolutionary development
approach. But don't you agree that core stability (with modularity albeit)
is equal in importance to the evolutionary dynamism of the content? It seems
that if the core were to evolve with the same rapidity as the content that
stability would be sacrificed.
I agree with your last sentence completely.
What I'm trying to argue is that the core should be constructed so
that the difference between changing stuff inside it and changing
stuff outside it is minimal.
Now, I should be more careful with the terms. MOO is an example,
I think, of a system that walked this line between core stability
and end-user innovation very well. Here's how it works. There
are 5 levels (sort of):
1. The MOO server
This is a C program maintained by one person who
makes infrequent, well-thought out changes some of which
are submitted by a small number of hackers (~20?). This
server provides a bunch of persistent objects with
behavior, object inheritence, a scripting language and
ownership and security mechanisms.
2. The World Core
A set of objects used to start new worlds. These objects
define places, people (players), object containment
(player A can be in room B), what happens when someone
connects to the server and so on. This set of objects was
contributed to by many over time. It changes slowly.
There is more than one core in the MOO community, but not
as many as one-per-world.
3. Fertile Objects
Fertile objects correspond to superclasses in Java.
(Subclassing is handled via object delegation. Objects
that can have children, e.g., are fertile, are those whose
behaviors can be inherited. Many more people create and
maintain fertile objects. Many of these eventually make
there way into a world core, but not all.
4. Objects w/ behavior
End users can create objects with behaviors (e.g., run
this script when someone in the same as the object room
says "foo"). Still more people create these.
5. Objects w/o behavior
End users can create objects that just look a certain
way. This corresponds to most activeworlds construction,
except that the look of an objects is just its text
description in MOO, rather than its geometry. Everybody
creates these.
In this system, only a small part of the infrastructure (the C
program) is truly at center and off limits to virtually everybody.
(It's off limits only in a practical sense. The server's source
is openly available, but most everyone found it better to work at
the higher levels.) Almost all of what corresponded to a shared
core was defined at level #2. And those objects were different
from level #4 only in they were agreed upon by the community.
There is this nice learning curve that draws people in and
encourages experimentation and innovation. I think this is a great
example that we can learn from.
- Mark

At 08:52 PM 3/5/99 +0100, Remi Sussan wrote:
>Gerald de Jong wrote:
>
>> a truly forward-looking virtual world implementation will involve
>> abandoning the notions of client and server and replacing them
>> with "participant" or something. a participant is always both client
>> and server.
>>
>Does this means that a multi-user server would not be necessary for
>running multi-user worlds? One thing which really annoys me with
>multi-user worlds is this dependance to a professional server. even if
>the server is free, it's mostly impossible to run it for a hobbyist, who
>has only access to web space provided by his ISP, but has no acccess to
>any server side programming. How do you imagine a multi-user world
>without multi-user server ? Would it be possible ? easy to create ?
Been lurking here for a few days...
I have some notes on the distributed system architecture I've been working
on for a few years now available at my website, it's a work in progress, a
bad doc to HTML translation, and needs another thousand hours or so of
typing, but I've got to go public with it sooner or later. It's at
<www.vscape.com/new/whitepaper.htm>, please ignore the rest of the site,
it's lame, hasn't been touched in over a year, and doesn't reflect my
decision to move to open source.
Some of the sentences in this document are longer than the one above, I
need to find me a good technical writer one of these days. :)
The part I've actually typed in describes an architectural approach to
deploying a distributed system for VR. I'm currently typing in the section
that describes how the data itself is distributed across load balancing
servers. FWIW, the implementation is much further along than the
documentation.
Comments, questions, scathing criticisms, ect. are all appreciated, if
nothing else it's one idea on how to solve the problems.
Kerry

Hi,
"Kerry L. Bonin" wrote:
> At 08:52 PM 3/5/99 +0100, Remi Sussan wrote:
> >Gerald de Jong wrote:
> >
> >> a truly forward-looking virtual world implementation will involve
> >> abandoning the notions of client and server and replacing them
> >> with "participant" or something. a participant is always both client
> >> and server.
> >>
> >Does this means that a multi-user server would not be necessary for
> >running multi-user worlds? One thing which really annoys me with
> >multi-user worlds is this dependance to a professional server. even if
> >the server is free, it's mostly impossible to run it for a hobbyist, who
> >has only access to web space provided by his ISP, but has no acccess to
> >any server side programming. How do you imagine a multi-user world
> >without multi-user server ? Would it be possible ? easy to create ?
>
> Been lurking here for a few days...
>
> I have some notes on the distributed system architecture I've been working
> on for a few years now available at my website, it's a work in progress, a
> bad doc to HTML translation, and needs another thousand hours or so of
> typing, but I've got to go public with it sooner or later. It's at
> <www.vscape.com/new/whitepaper.htm>, please ignore the rest of the site,
> it's lame, hasn't been touched in over a year, and doesn't reflect my
> decision to move to open source.
>
> Some of the sentences in this document are longer than the one above, I
> need to find me a good technical writer one of these days. :)
>
> The part I've actually typed in describes an architectural approach to
> deploying a distributed system for VR. I'm currently typing in the section
> that describes how the data itself is distributed across load balancing
> servers. FWIW, the implementation is much further along than the
> documentation.
>
> Comments, questions, scathing criticisms, ect. are all appreciated, if
> nothing else it's one idea on how to solve the problems.
>
> Kerry
There are two open-source IP-Multicasting based multi-user 3D browsers
products out there: YouBuildItVR (http://www.ncsa.uiuc.edu/SDG/Software/Java3D)
and Bang. The IP-Multicasting solution is serverless but requires one to be
connected to the MBONE or to be on the same sub-net as other connected
Avatars.
Basically, it is much more efficient to skip the server, you move your avatar and
the movement happens closer to RealTime than if all packages have to go trough
a server.
Warm regards,
Róbert Viðar Bjarnason
robofly@bang.is
bang space inc.
http://the.bang.is/

At 05:13 PM 3/5/99 -0800, you wrote:
>There are two open-source IP-Multicasting based multi-user 3D browsers
>products out there: YouBuildItVR
(http://www.ncsa.uiuc.edu/SDG/Software/Java3D)
>and Bang. The IP-Multicasting solution is serverless but requires one to be
>connected to the MBONE or to be on the same sub-net as other connected
>Avatars.
>
>Basically, it is much more efficient to skip the server, you move your
avatar and
>the movement happens closer to RealTime than if all packages have to go
trough
>a server.
Hello Róbert!
I've seen these systems, although admittedly not in much depth recently.
I've avoided "serverless" systems that rely on solely peer-peer for two
reasons - security and scalability.
I have implemented systems (in the games industry where I worked last
~10yrs) that scale into the low hundreds of users using entirely peer-peer
topologies. I do agree that the latency reduction of eliminating a hop is
compelling, but managing state properly for such a world can easily become
a O(N^2) problem.
I've taken a hybrid approach instead to get as many benefits as possible
from both topologies. I use a model that looks like client/server
peripherally, except that at I allow message/stream routing/broadcasting
information to be propagated back towards the clients so time critical
information can arrive as soon as practical given bandwidth loads.
I still retain the concept of a server to handle things that need to be
single-point, such as state decisions or ownership transfers. And then the
client/server indirection can be taken advantage of to provide hot failover
server nodes for a more robust architecture.
And I will take another look at Bang, it's been a while... :)
Kerry

Hello OWorld again, erstwhile moderator Bruce Damer would like to welcome
all newbies, 25 of whom have signed up in the past 3 days. The list is
really jumping! Newbies, even if you are planning to lurk, kindly consider
introducing yourself so the active members can know you or your works. We
are using a kind of format for this below..
1. Where do I come from?
2. Why did I join OWorld (ie: why do I believe in open community efforts to
create virtual worlds and platforms)
3. What am I working on/would like to work on
4. Controversial statement of the day

Bruce Damer
(DigiGardener)

++ The Contact Consortium ++
A Forum for Contact, Culture and Community in Digital Space
Visit us at: http://www.ccon.org
/- Living, Learning and Working in Virtual Worlds -/

Bruce Damer wrote:
>
> Hello OWorld again, erstwhile moderator Bruce Damer would like to welcome
> all newbies, 25 of whom have signed up in the past 3 days. The list is
> really jumping! Newbies, even if you are planning to lurk, kindly consider
> introducing yourself so the active members can know you or your works. We
> are using a kind of format for this below..
Hi,
this is a good point to stop lurking :-)
>
> 1. Where do I come from?
I'm working for / with geometrek and develop an open-source multi-user system
called
DeepMatrix. thas why I'm interested in all this discussions here.
> 2. Why did I join OWorld (ie: why do I believe in open community efforts to
> create virtual worlds and platforms)
I don't really know why I believe in virtual environments ! I guess I read
the right books at the right age, when I was easily impressed and formed.
however, I like open community efforts and think this is a good way to work.
I would like to see a very wide spread infrastructure for VEs, where everyone
can add their own stuff. we might end up with something as complicated and
sometimes hard to use as the web, but i think its worth the risk.
> 3. What am I working on/would like to work on
still developing DeepMatrix further. we have a lot of ideas floating around
and have somehow to keep track of what we want to do next. a central point
of discussion ( also on this list ) is the communication structure of such a
system, basically the old server/client versus peer to peer. I believe a hyprid
solution can give the best results. use peer to peer where you have the
processing
power to handle all the updates and the bandwidth ( might be less of an issue
with multicast ) but keep servers for slow connected, slow CPU clients. here
the servers could provide some reductions / fitlering etc. of the information
for the clients.
> 4. Controversial statement of the day
cyberspace will be different from everything we imagine today.

bye,
Gerhard
>
> Bruce Damer
> (DigiGardener)
>
> ++ The Contact Consortium ++
> A Forum for Contact, Culture and Community in Digital Space
> Visit us at: http://www.ccon.org
>
>
> /- Living, Learning and Working in Virtual Worlds -/

--
Gerhard Reitmayr mailto:e9325167@student.tuwien.ac.at \\\
http://stop.at/gerhard L\\\\
DeepMatrix http://www.geometrek.com/products/ \\\\
I feel I am diagonally parked in a parallel universe
Howdy people,
Just subscribed and wants to know if there exists an archive of this
mailing list somewhere. If not, maybe someone that uses NS can zip the
mailfolder and send it to me (but please send a notification to the list
also so that I don't get 2000 of them:)
Looking forward to read some intersting posts...
Cheers,
/Niclas
--
niclas olofsson christian doegl virtual real-estate gmbh
mailto:no@dc.co.at breite gasse 3, a-1070 vienna
http://www.dc.co.at fon: ++43 1 526 29 67 fax: ++43 1 526 29 67 11

Back to Consortium Home Page

Please send any comments to Nancy Zuidema