Re: [croquet-user] Important questions for me..

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

Re: [croquet-user] Important questions for me..

Janet Plato
I think you meant Janet and not James, but I could be wrong.

I understand that you need a protocol spec in order to develop a
server or client to implement a protocol.  I understand that this is
all a large problem and will take a lot of work to solve.  But I think
you missed my entire question.

What is the road map for interoperability between protocols and is
anyone working on it?  Initially I see croquet, x3d and libsecondlife
as the protocols of choice, but I am not certain how this will really
work out.

Mailing lists are relatively slow.  I think people need to be able to
download the SDK and connect to a place to go and hang out with other
users in order for a community support network to develop.  IMO, that
support network and community will greatly accelerate world building.
I have no interest in making commercial worlds, or in proof of concept
work.  We are past that.

We need to get people the tools to start building large scale worlds
and then get out of their way.  Much of this is happening, I just want
to know the roadmap so I can determine where I fit in.

Cheers,

Janet

On Dec 3, 2007 1:09 PM, Les <[hidden email]> wrote:

> Hi, James,
>         Without a protocol spec, and especially one that is accepted by the
> various developers, having multiple worlds that you can visit, and
> especially worlds from different developers you can visit is impossible.
> Think of the protocols as the formal specification of the communications
> between the various elements, thus setting the language of common
> discourse.  If the different worlds all speak differently and the
> clients cannot talk to or with them about the worlds they represent,
> then there is no way that one can understand the other, and your avatar
> cannot cross the boundary, even if you can deal with the language
> differences on the other side of the boundary, which requires a whole
> different level of understanding.
>
>         Right now, Croquet has a spec, Second Life has a spec, MMORPG's each
> have their own spec, and one cannot talk to or visit the other.  This
> has to be sorted out at some point for the web to become fully or even
> somewhat virtualized.  We are all just taking baby steps at this time,
> learning how to accomplish the virutalization is but one piece of the
> puzzle, learning to get from world to world is another, learning how to
> populate and set the environment in these worlds is yet another, and
> this doesn't even touch on the issues with external development.  This
> is just the internal issues within each of the various virtual languages
> and structures.  Furthermore, cross content virtualization will not
> likely come until a breakthrough in expression of the necessary
> requirements for virtualization, characteristics of Avatars and
> environments, and the communications necessary to set the communications
> methods, expressions to use, and structures of components that are
> common to all such environments.
>
>         Progress is being made, but this is a big puzzle, with things like how
> to associate gravity with an object (i.e. property lists and how they
> should be declared?)  or perhaps the means to express the date, time,
> location in the overall virtual world of each location, and what will be
> the standards?  Will there be 3, 4, 5, or who knows how many dimensions
> to the ultimate expression of virtual worlds.  And do the objects need
> to express the full dimensionality of their location, or should there be
> a world location, followed by offsets to tell the objects shape and
> geophysical properties?  Some of these issues will be worked out by
> individual developers, but conferences, and meetings of stakeholders
> like Linden Labs, and the Croquet Consortium will be necessary to
> determine these issues and how to present them across platforms.  During
> that stage, the objects will have to change, inherit the new properties
> so that they will be consistent with the virtual world interface and
> protocol that emerges.
>
>         But do not be discouraged.  Early developers, and especially those with
> interesting concepts, well developed platforms, and large user bases
> will all eventually be drawn to support the further development of the
> Virtual Web.  I can foresee addresses perhaps beginning with VW vs WWW
> to help guide virtual visitors across the new world.
>
> Regards,
> Les H
>
> On Mon, 2007-12-03 at 11:59 -0600, Janet Plato wrote:
> > There are folks running private islands and starting on world building
> > using second life.  The results are starting to get good.  Check out
> > OpenSIM and libsecondlife.  You just run the second life client with a
> > command line option to specify the world to login to.
> >
> > Are any of the croquet folks planning on allowing a croquet world to
> > be accessed by a secondlife client, or a croquet client to access a
> > second life world?
> >
> > It seems you need a large user base to really get things going, and
> > supporting multiple protocols would help with that.
> >
> > Imagine a virtual bazarrre where people can connect from multiple
> > clients to a small  server that provides a portal to private servers.
> > The bazarre becomes almost like DNS, google and the yellow pages for a
> > virtual world.
> >
> > Folks have a stable place to connect to, and debug their client
> > implementations, and they can enter a shop or go through a portal to
> > enter other worlds with content they want.
> >
> > Once the worlds get more common, a better search paradigm would
> > emerge, but this solves the problem the web had before search engines,
> > how do I find content.
> >
> > It's just a thought, but I can say for me without places to go, who
> > cares how cool the protocol spec looks...
> >
> > Janet Plato
> >
> > On Dec 3, 2007 10:18 AM, Doug Thompson <[hidden email]> wrote:
> >
> > >  If you're looking at a multi-user environment, I'd tend to concur with
> > > Darius that SL might be the best bet for the time being. We're working on an
> > > application that allows direct links down from a Web site to a private
> > > island, a contained group kind of idea, such that users don't even really
> > > need to know they're "in" Second Life. If they choose to explore out, great.
> > > Otherwise it allows us to control the user experience a little more and is
> > > vertical rather than horizontal – driving users to 'sealed' regions from the
> > > Web rather than to the region from within SL itself – although frankly the
> > > combination is probably best.
> > >
>
>
Reply | Threaded
Open this post in threaded view
|

Re: Re: [croquet-user] Important questions for me..

Baldur Johannsson
H'lo Janet

On 03/12/2007, Janet Plato <[hidden email]> wrote:
> I think you meant Janet and not James, but I could be wrong.
>
> I understand that you need a protocol spec in order to develop a
> server or client to implement a protocol.  I understand that this is
> all a large problem and will take a lot of work to solve.  But I think
> you missed my entire question.
There isnt an server in the usual sense in Croquet other than Croquet routers.
Which only purpose is to provide an pinch that serializes the order of the
replicated messages that all Island replicas must recive to depict the same
state of the Island. That order is needed because otherwise you would
get the Butterfly effect.

The Croquet protocol between participants and router is rather simple.

 An participant opens an TCP connection to an known endpoint (a ip
address/hostname and port that can be gotten from an CroquetPostcard).
 Sends the 128 bits that identifies the TeaParty/Island (more
specificaly the Croquet router which it is synchronized via) and the
Dispatcher hand the connection to that router.
Then the participant tells the router it wants to subscribe to the
sream of replicated messages sent to the router.
Next the participant tells the router it wants to get the lates
snapshot of the Island the router is responsible for. (The router just
forwards this message to an participant that has been trusted with
takeing snapshots of the Island and forwards the reply back)
Now the participant has the snapshot of the Island and is keeping it
up to date with the replicated messages.
So far the participant can only view the TeaParty but isnt in it yet.
The participant then sends the router messages that are destined for the Island.
The router timestamps these messages and broadcasts to all
participants that are subscriped to the sream of replicated messages.

The above protocol could be implemented by Croquet routers written in
any language if
there wasnt the need to deserialize each replicated message, time stamp it and
reserialize it again.

>
> What is the road map for interoperability between protocols and is
> anyone working on it?  Initially I see croquet, x3d and libsecondlife
> as the protocols of choice, but I am not certain how this will really
> work out.
>
Good qestion which I also want to see answeared.

> Mailing lists are relatively slow.  I think people need to be able to
> download the SDK and connect to a place to go and hang out with other
> users in order for a community support network to develop.  IMO, that
> support network and community will greatly accelerate world building.
> I have no interest in making commercial worlds, or in proof of concept
> work.  We are past that.
>
> We need to get people the tools to start building large scale worlds
> and then get out of their way.  Much of this is happening, I just want
> to know the roadmap so I can determine where I fit in.
>
I agree with you on both points heartily.

> Cheers,
>
> Janet
>
Reply | Threaded
Open this post in threaded view
|

Croquet meeting place

Tapple Gao
In reply to this post by Janet Plato
On Mon, Dec 03, 2007 at 02:50:49PM -0600, Janet Plato wrote:
> Mailing lists are relatively slow.  I think people need to be able to
> download the SDK and connect to a place to go and hang out with other
> users in order for a community support network to develop.  IMO, that
> support network and community will greatly accelerate world building.

On irc.freenode.net, we of the #squeak channel get quite a few
questions about croquet, but are never able to answer them.
There is a #croquet channel, but there is not often anyone with
actual croquet experience in there to help newbies out. I think
you could have good discussions via IRC for now. Many projects
use IRC very successfully.

It doesn't sound like there is a single place to go meet croquet
people, and so I often have nowhere to send croquet admirers who
come into #squeak (and nowhere to hang out myself) regarding
croquet.  IRC would be great until there is a better place to
meet croquet people (like the croquet consortium server or an SL
island)

--
Matthew Fulmer -- http://mtfulmer.wordpress.com/
Help improve Squeak Documentation: http://wiki.squeak.org/squeak/808
Reply | Threaded
Open this post in threaded view
|

Re: Croquet meeting place

David Faught
On 12/10/07, Matthew Fulmer <[hidden email]> wrote:

> On Mon, Dec 03, 2007 at 02:50:49PM -0600, Janet Plato wrote:
> > Mailing lists are relatively slow.  I think people need to be able to
> > download the SDK and connect to a place to go and hang out with other
> > users in order for a community support network to develop.  IMO, that
> > support network and community will greatly accelerate world building.
>
> On irc.freenode.net, we of the #squeak channel get quite a few
> questions about croquet, but are never able to answer them.
> There is a #croquet channel, but there is not often anyone with
> actual croquet experience in there to help newbies out. I think
> you could have good discussions via IRC for now. Many projects
> use IRC very successfully.
>
> It doesn't sound like there is a single place to go meet croquet
> people, and so I often have nowhere to send croquet admirers who
> come into #squeak (and nowhere to hang out myself) regarding
> croquet.  IRC would be great until there is a better place to
> meet croquet people (like the croquet consortium server or an SL
> island)

The Croquet Collaborative server was, I think, intended to be a place
like this.  It is still available, and reasonably stable.  There are
several Croquet-ers that have published their contact information on
Second Life, although it might be nice to have such a directory
somewhere (is there one in Second Life?)

At one time, the UMN Croquet Team offered free accounts on its Jabber
server (I believe this is no longer the case) in order to facilitate
in-Croquet communication with other external channels.  As I recall,
they had very few takers and little Croquet activity on the server.
It seems that a lot of the Croquet community would rather work on
their projects than talk about them, which seems a little odd for an
environment whose main claim to fame is collaboration.

Dave
Reply | Threaded
Open this post in threaded view
|

Re: Croquet meeting place

Janet Plato
It has not been my experience that this world is stable or documented.
 I am not trying to sound negative, but I come into this without
understanding the paradigms and metaphors used in this space, and
being somewhat overwhelmed with where to get started.  I strongly
suspect many artists, engineers and world builders exist, but are in a
similiar state of ignorance.

I think that when a project is below a critical mass, having a stable
place to catch the attention of folks who pass by adds some value,
even if it's mostly unused.  I would have been incredibly thrilled to
attach to an empty world with a bill board covered in post-it notes
pointing me to other worlds and other folks current projects.  Think
of google, and then think "how could I make a google like service
using the croquet paradigms"  Think of all the getting started
documents, and then ask "how do I make the getting started documents
on the web point to "installation and connecting" documents, and
moving the rest of it into the world so people will go into the worlds
and get used to doing things there.

Use croquet to teach croquet.

Every time I see something on the web about croquet, there is a reason
I cannot participate, which leads me to conclude other folks are in
the same position.  I see the arts metaverse project and think,
WOW!!!! but then I cannot connect to it, cannot get the datasets used
to make it and cannot get a reply from the folks who did it.

I have a day job, and have been unable to attend the last two meetings
in qwak space. I suspect for most of the people doing this, it's just
a day job.  This whole thing kind of feels like it's mostly about
someone milking the concept for a job or research grant.  I know that
cannot be true because 20 people were in the IRC channel once I found
it....  but imagine if those 20 people were in a croquet world making
new avatars, or writing how-to and read-me articles.

A lot of people want to help, we just have to show them how.

Cheers,

Janet


On Dec 12, 2007 8:25 AM, David Faught <[hidden email]> wrote:

> On 12/10/07, Matthew Fulmer <[hidden email]> wrote:
> > On Mon, Dec 03, 2007 at 02:50:49PM -0600, Janet Plato wrote:
> > > Mailing lists are relatively slow.  I think people need to be able to
> > > download the SDK and connect to a place to go and hang out with other
> > > users in order for a community support network to develop.  IMO, that
> > > support network and community will greatly accelerate world building.
> >
> > On irc.freenode.net, we of the #squeak channel get quite a few
> > questions about croquet, but are never able to answer them.
> > There is a #croquet channel, but there is not often anyone with
> > actual croquet experience in there to help newbies out. I think
> > you could have good discussions via IRC for now. Many projects
> > use IRC very successfully.
> >
> > It doesn't sound like there is a single place to go meet croquet
> > people, and so I often have nowhere to send croquet admirers who
> > come into #squeak (and nowhere to hang out myself) regarding
> > croquet.  IRC would be great until there is a better place to
> > meet croquet people (like the croquet consortium server or an SL
> > island)
>
> The Croquet Collaborative server was, I think, intended to be a place
> like this.  It is still available, and reasonably stable.  There are
> several Croquet-ers that have published their contact information on
> Second Life, although it might be nice to have such a directory
> somewhere (is there one in Second Life?)
>
> At one time, the UMN Croquet Team offered free accounts on its Jabber
> server (I believe this is no longer the case) in order to facilitate
> in-Croquet communication with other external channels.  As I recall,
> they had very few takers and little Croquet activity on the server.
> It seems that a lot of the Croquet community would rather work on
> their projects than talk about them, which seems a little odd for an
> environment whose main claim to fame is collaboration.
>
> Dave
>
Reply | Threaded
Open this post in threaded view
|

Re: Croquet meeting place

nick hemsley
Rome wasn't built in a day ;)

Croquet is in a somewhat painful place at the moment, especially for newbies.

I would love to start using croquet itself an irc-like fashion,
however croquet is not quite ready for this. The project is still in
beta phase, and the project itself is an enabling technology platform,
not an end user application. Qwaq forums & edusim are such end-user
applications, but I dont know if they are prime-time ready either
(qwaq may be though, never used it).

It's really just not a project for newbies yet, if you want to play
with croquet, you will fast find yourself needing to do some
programming.

My own opinions of course.

Cheers Nick

On Dec 15, 2007 2:55 AM, Janet Plato <[hidden email]> wrote:

> It has not been my experience that this world is stable or documented.
>  I am not trying to sound negative, but I come into this without
> understanding the paradigms and metaphors used in this space, and
> being somewhat overwhelmed with where to get started.  I strongly
> suspect many artists, engineers and world builders exist, but are in a
> similiar state of ignorance.
>
> I think that when a project is below a critical mass, having a stable
> place to catch the attention of folks who pass by adds some value,
> even if it's mostly unused.  I would have been incredibly thrilled to
> attach to an empty world with a bill board covered in post-it notes
> pointing me to other worlds and other folks current projects.  Think
> of google, and then think "how could I make a google like service
> using the croquet paradigms"  Think of all the getting started
> documents, and then ask "how do I make the getting started documents
> on the web point to "installation and connecting" documents, and
> moving the rest of it into the world so people will go into the worlds
> and get used to doing things there.
>
> Use croquet to teach croquet.
>
> Every time I see something on the web about croquet, there is a reason
> I cannot participate, which leads me to conclude other folks are in
> the same position.  I see the arts metaverse project and think,
> WOW!!!! but then I cannot connect to it, cannot get the datasets used
> to make it and cannot get a reply from the folks who did it.
>
> I have a day job, and have been unable to attend the last two meetings
> in qwak space. I suspect for most of the people doing this, it's just
> a day job.  This whole thing kind of feels like it's mostly about
> someone milking the concept for a job or research grant.  I know that
> cannot be true because 20 people were in the IRC channel once I found
> it....  but imagine if those 20 people were in a croquet world making
> new avatars, or writing how-to and read-me articles.
>
> A lot of people want to help, we just have to show them how.
>
> Cheers,
>
> Janet
>
>
>
> On Dec 12, 2007 8:25 AM, David Faught <[hidden email]> wrote:
> > On 12/10/07, Matthew Fulmer <[hidden email]> wrote:
> > > On Mon, Dec 03, 2007 at 02:50:49PM -0600, Janet Plato wrote:
> > > > Mailing lists are relatively slow.  I think people need to be able to
> > > > download the SDK and connect to a place to go and hang out with other
> > > > users in order for a community support network to develop.  IMO, that
> > > > support network and community will greatly accelerate world building.
> > >
> > > On irc.freenode.net, we of the #squeak channel get quite a few
> > > questions about croquet, but are never able to answer them.
> > > There is a #croquet channel, but there is not often anyone with
> > > actual croquet experience in there to help newbies out. I think
> > > you could have good discussions via IRC for now. Many projects
> > > use IRC very successfully.
> > >
> > > It doesn't sound like there is a single place to go meet croquet
> > > people, and so I often have nowhere to send croquet admirers who
> > > come into #squeak (and nowhere to hang out myself) regarding
> > > croquet.  IRC would be great until there is a better place to
> > > meet croquet people (like the croquet consortium server or an SL
> > > island)
> >
> > The Croquet Collaborative server was, I think, intended to be a place
> > like this.  It is still available, and reasonably stable.  There are
> > several Croquet-ers that have published their contact information on
> > Second Life, although it might be nice to have such a directory
> > somewhere (is there one in Second Life?)
> >
> > At one time, the UMN Croquet Team offered free accounts on its Jabber
> > server (I believe this is no longer the case) in order to facilitate
> > in-Croquet communication with other external channels.  As I recall,
> > they had very few takers and little Croquet activity on the server.
> > It seems that a lot of the Croquet community would rather work on
> > their projects than talk about them, which seems a little odd for an
> > environment whose main claim to fame is collaboration.
> >
> > Dave
> >
>