This might be a dream but what do you think about an OWL solution combining the following:
- Possibility of using low-end PC clients to access OWL servers - Possibility of using any tablet to access OWL servers - An OWL which would be fully web-based, possibly using clouds, proxies, etc... - Cancelation of all current constraints to open ports on clients sites - And of course a solution providing same functionality, performance, scalability, usability, .... as with conventional heavy Java OWL clients connected to servers
I am looking for a ballpark number in terms of required man-years of engineering
I have time thinking in implement a webgl client using GWT to compile the
actial OWL client and websockes to communicate, but I am still joining the
pieces, and estimating maybe two years of full work until a such client
reach the users.
The lowend PC clients and tablets I have herad people that are using VNC
for this, but it should be formalized in order that ordinary users can use
it
But sure I want to go in this way too
On Mon, Jul 23, 2012 at 11:57 AM, micheldenis <michel.m.de...@gmail.com>wrote:
> This might be a dream but what do you think about an OWL solution
> combining the following:
> - Possibility of using low-end PC clients to access OWL servers
> - Possibility of using any tablet to access OWL servers
> - An OWL which would be fully web-based, possibly using clouds, proxies,
> etc...
> - Cancelation of all current constraints to open ports on clients sites
> - And of course a solution providing same functionality, performance,
> scalability, usability, .... as with conventional heavy Java OWL clients
> connected to servers
> I am looking for a ballpark number in terms of required man-years of
> engineering
> I have time thinking in implement a webgl client using GWT to compile > the actial OWL client and websockes to communicate, but I am still > joining the pieces, and estimating maybe two years of full work until > a such client reach the users.
> The lowend PC clients and tablets I have herad people that are using > VNC for this, but it should be formalized in order that ordinary users > can use it
> But sure I want to go in this way too
> On Mon, Jul 23, 2012 at 11:57 AM, micheldenis > <michel.m.de...@gmail.com <mailto:michel.m.de...@gmail.com>> wrote:
> This might be a dream but what do you think about an OWL solution
> combining the following:
> - Possibility of using low-end PC clients to access OWL servers
> - Possibility of using any tablet to access OWL servers
> - An OWL which would be fully web-based, possibly using clouds,
> proxies, etc...
> - Cancelation of all current constraints to open ports on clients
> sites
> - And of course a solution providing same functionality,
> performance, scalability, usability, .... as with conventional
> heavy Java OWL clients connected to servers
> I am looking for a ballpark number in terms of required man-years
> of engineering
> Thanks,
> -michel
> No virus found in this message.
> Checked by AVG - www.avg.com <http://www.avg.com>
> Version: 2012.0.2197 / Virus Database: 2437/5149 - Release Date: 07/23/12
I've often thought of writing a WebGL based "Open Wonderland
Player"...basically a client without the world-building aspects. As a
first step, I think acquiring the information to draw from the server
would be a straight-forward task, though much of the java client
connections we use would need to be implemented in the web application
(server-side). I think it would be reasonable to see such a work by
Halloween or so. Though this would have no world building features, no
audio, no good avatar implementation (avatars would likely be a
picture of a figure rather than an actual 3D figure), and no
synchronized state....a user could walk through a world but not see
any changes made by other users...a purely single-player experience.
That's by halloween.
By year's end, I think I could see a more synchronized client, one
where a user can see other users walking around and changes that they
might make. Still no good avatar implementation, no audio, and no good
world building features. Given another 6 months after that, *maybe* we
might see an intermediate avatar implementation away from the images
used prior. That gets us to about a year, maybe a little less from
now.
From there, I think I would split the player into an additional
version, one with world-building capabilities and do another 6 months
of GUI work to better develop the tools for the engine usage so one
can create rich artwork, meshes, and effects. Beyond that, I think I
would see it released under the foundation's supervision and see where
that goes.
All in all, this is just a dream-sequence. It would take a lot of
expertise, and more than one (maybe even several) developer(s) to
produce such a design. It's really not something you can just plan on
the forum and say..."wouldn't it be cool if..." and expect it to be
done.
On Mon, Jul 23, 2012 at 12:27 PM, Roland Sassen <sas...@thinsia.com> wrote:
> Webgl is not a web-based or cloud-based but HTML 5 based solution.
> Roland
> On 23-7-2012 19:04, Carlos Rafael Ramirez wrote:
> I have time thinking in implement a webgl client using GWT to compile the
> actial OWL client and websockes to communicate, but I am still joining the
> pieces, and estimating maybe two years of full work until a such client
> reach the users.
> The lowend PC clients and tablets I have herad people that are using VNC for
> this, but it should be formalized in order that ordinary users can use it
> But sure I want to go in this way too
> On Mon, Jul 23, 2012 at 11:57 AM, micheldenis <michel.m.de...@gmail.com>
> wrote:
>> This might be a dream but what do you think about an OWL solution
>> combining the following:
>> - Possibility of using low-end PC clients to access OWL servers
>> - Possibility of using any tablet to access OWL servers
>> - An OWL which would be fully web-based, possibly using clouds, proxies,
>> etc...
>> - Cancelation of all current constraints to open ports on clients sites
>> - And of course a solution providing same functionality, performance,
>> scalability, usability, .... as with conventional heavy Java OWL clients
>> connected to servers
>> I am looking for a ballpark number in terms of required man-years of
>> engineering
>> Thanks,
>> -michel
> No virus found in this message.
> Checked by AVG - www.avg.com > Version: 2012.0.2197 / Virus Database: 2437/5149 - Release Date: 07/23/12
> This might be a dream but what do you think about an OWL solution > combining the following:
> - Possibility of using low-end PC clients to access OWL servers
> - Possibility of using any tablet to access OWL servers
> - An OWL which would be fully web-based, possibly using clouds, > proxies, etc...
> - Cancelation of all current constraints to open ports on clients sites
> - And of course a solution providing same functionality, performance, > scalability, usability, .... as with conventional heavy Java OWL > clients connected to servers
> I am looking for a ballpark number in terms of required man-years of > engineering
> Thanks,
> -michel
> No virus found in this message.
> Checked by AVG - www.avg.com <http://www.avg.com>
> Version: 2012.0.2197 / Virus Database: 2437/5149 - Release Date: 07/23/12
Sounds like an enhanced client/server protocol to remove the data
independence issue that using serialized java classes creates and a opengl
(es) renderer that displays based on the content of the world rather than
just rendering would be in order. Maybe someone already did that.
On Mon, Jul 23, 2012 at 1:48 PM, Roland Sassen <sas...@thinsia.com> wrote:
> Only WonderSchool can do this at the moment,
> I think,
> cheers Roland
> On 23-7-2012 18:57, micheldenis wrote:
> This might be a dream but what do you think about an OWL solution
> combining the following:
> - Possibility of using low-end PC clients to access OWL servers
> - Possibility of using any tablet to access OWL servers
> - An OWL which would be fully web-based, possibly using clouds, proxies,
> etc...
> - Cancelation of all current constraints to open ports on clients sites
> - And of course a solution providing same functionality, performance,
> scalability, usability, .... as with conventional heavy Java OWL clients
> connected to servers
> I am looking for a ballpark number in terms of required man-years of
> engineering
As we know it for a long time now, OWL is more advanced than most of the VW platforms on the marketplace, in a number of areas.
Still we know that the future for VW is the Web, and platforms (e.g. gaming ones) already exist in this area, but missing collaborative tools that OWL has for a long time now.
The architecture foundations of OWL enable all this ...
Carlos Rafael Ramirez wrote:
> I have time thinking in implement a webgl client using GWT to compile > the actial OWL client and websockes to communicate, but I am still > joining the pieces, and estimating maybe two years of full work until > a such client reach the users.
> The lowend PC clients and tablets I have herad people that are using > VNC for this, but it should be formalized in order that ordinary users > can use it
> But sure I want to go in this way too
> On Mon, Jul 23, 2012 at 11:57 AM, micheldenis > <michel.m.de...@gmail.com <mailto:michel.m.de...@gmail.com>> wrote:
> This might be a dream but what do you think about an OWL solution
> combining the following:
> - Possibility of using low-end PC clients to access OWL servers
> - Possibility of using any tablet to access OWL servers
> - An OWL which would be fully web-based, possibly using clouds,
> proxies, etc...
> - Cancelation of all current constraints to open ports on clients
> sites
> - And of course a solution providing same functionality,
> performance, scalability, usability, .... as with conventional
> heavy Java OWL clients connected to servers
> I am looking for a ballpark number in terms of required man-years
> of engineering
> Sounds like an enhanced client/server protocol to remove the data > independence issue that using serialized java classes creates and a > opengl (es) renderer that displays based on the content of the world > rather than just rendering would be in order. Maybe someone already > did that.
> Morris
> On Mon, Jul 23, 2012 at 1:48 PM, Roland Sassen <sas...@thinsia.com > <mailto:sas...@thinsia.com>> wrote:
> Only WonderSchool can do this at the moment,
> I think,
> cheers Roland
> On 23-7-2012 18:57, micheldenis wrote:
>> This might be a dream but what do you think about an OWL solution
>> combining the following:
>> - Possibility of using low-end PC clients to access OWL servers
>> - Possibility of using any tablet to access OWL servers
>> - An OWL which would be fully web-based, possibly using clouds,
>> proxies, etc...
>> - Cancelation of all current constraints to open ports on clients
>> sites
>> - And of course a solution providing same functionality,
>> performance, scalability, usability, .... as with conventional
>> heavy Java OWL clients connected to servers
>> I am looking for a ballpark number in terms of required man-years
>> of engineering
> No virus found in this message.
> Checked by AVG - www.avg.com <http://www.avg.com>
> Version: 2012.0.2197 / Virus Database: 2437/5149 - Release Date: 07/23/12
When you have time, please could you tell about your feelings about such a project ? From all I've read about OWL, it has everything in its foundations to enable what I suggest, maybe with lots of engineering but .. on the principle is it true or false ?
-michel
Le lundi 23 juillet 2012 20:40:37 UTC+2, micheldenis a écrit :
> As we know it for a long time now, OWL is more advanced than most of the > VW platforms on the marketplace, in a number of areas.
> Still we know that the future for VW is the Web, and platforms (e.g. > gaming ones) already exist in this area, but missing collaborative tools > that OWL has for a long time now.
> The architecture foundations of OWL enable all this ...
> -michel
> Carlos Rafael Ramirez wrote:
> I have time thinking in implement a webgl client using GWT to compile the > actial OWL client and websockes to communicate, but I am still joining the > pieces, and estimating maybe two years of full work until a such client > reach the users.
> The lowend PC clients and tablets I have herad people that are using VNC > for this, but it should be formalized in order that ordinary users can use > it
> But sure I want to go in this way too
> On Mon, Jul 23, 2012 at 11:57 AM, micheldenis <michel.m.de...@gmail.com>wrote:
>> This might be a dream but what do you think about an OWL solution >> combining the following:
>> - Possibility of using low-end PC clients to access OWL servers
>> - Possibility of using any tablet to access OWL servers
>> - An OWL which would be fully web-based, possibly using clouds, proxies, >> etc...
>> - Cancelation of all current constraints to open ports on clients sites
>> - And of course a solution providing same functionality, performance, >> scalability, usability, .... as with conventional heavy Java OWL clients >> connected to servers
>> I am looking for a ballpark number in terms of required man-years of >> engineering
>> Thanks,
>> -michel
> -- > Michel DENIS
> Président, Internet 3 Solutions SAS
> Tél: 06.72.87.17.50michel.de...@internet3solutions.comhttp://www.internet3solutions.com/
> Sounds like an enhanced client/server protocol to remove the data
> independence issue that using serialized java classes creates and a opengl
> (es) renderer that displays based on the content of the world rather than
> just rendering would be in order. Maybe someone already did that.
> Morris
> On Mon, Jul 23, 2012 at 1:48 PM, Roland Sassen <sas...@thinsia.com> wrote:
>> Only WonderSchool can do this at the moment,
>> I think,
>> cheers Roland
>> On 23-7-2012 18:57, micheldenis wrote:
>> This might be a dream but what do you think about an OWL solution
>> combining the following:
>> - Possibility of using low-end PC clients to access OWL servers
>> - Possibility of using any tablet to access OWL servers
>> - An OWL which would be fully web-based, possibly using clouds, proxies,
>> etc...
>> - Cancelation of all current constraints to open ports on clients sites
>> - And of course a solution providing same functionality, performance,
>> scalability, usability, .... as with conventional heavy Java OWL clients
>> connected to servers
>> I am looking for a ballpark number in terms of required man-years of
>> engineering
Carlos Rafael Ramirez wrote:
> I have time thinking in implement a webgl client using GWT to compile > the actial OWL client and websockes to communicate, but I am still > joining the pieces, and estimating maybe two years of full work until > a such client reach the users.
> The lowend PC clients and tablets I have herad people that are using > VNC for this, but it should be formalized in order that ordinary users > can use it
> But sure I want to go in this way too
> On Mon, Jul 23, 2012 at 11:57 AM, micheldenis > <michel.m.de...@gmail.com <mailto:michel.m.de...@gmail.com>> wrote:
> This might be a dream but what do you think about an OWL solution
> combining the following:
> - Possibility of using low-end PC clients to access OWL servers
> - Possibility of using any tablet to access OWL servers
> - An OWL which would be fully web-based, possibly using clouds,
> proxies, etc...
> - Cancelation of all current constraints to open ports on clients
> sites
> - And of course a solution providing same functionality,
> performance, scalability, usability, .... as with conventional
> heavy Java OWL clients connected to servers
> I am looking for a ballpark number in terms of required man-years
> of engineering
Mmm good point I think eventually will be sopported but in 18 months i
doubt. Google Chrome comes with a software implementation of WebGL to help
a little. Ipad now supports WebGL, so i think is mean of time
On Mon, Jul 23, 2012 at 3:05 PM, Michel DENIS <michel.m.de...@gmail.com>wrote:
> Do you know whether WebGL be supported on all PC browsers and tablets'
> interfaces in the coming 18 months ? That would help.
> Thanks,
> -michel
> Carlos Rafael Ramirez wrote:
> I have time thinking in implement a webgl client using GWT to compile the
> actial OWL client and websockes to communicate, but I am still joining the
> pieces, and estimating maybe two years of full work until a such client
> reach the users.
> The lowend PC clients and tablets I have herad people that are using VNC
> for this, but it should be formalized in order that ordinary users can use
> it
> But sure I want to go in this way too
> On Mon, Jul 23, 2012 at 11:57 AM, micheldenis <michel.m.de...@gmail.com>wrote:
>> This might be a dream but what do you think about an OWL solution
>> combining the following:
>> - Possibility of using low-end PC clients to access OWL servers
>> - Possibility of using any tablet to access OWL servers
>> - An OWL which would be fully web-based, possibly using clouds, proxies,
>> etc...
>> - Cancelation of all current constraints to open ports on clients sites
>> - And of course a solution providing same functionality, performance,
>> scalability, usability, .... as with conventional heavy Java OWL clients
>> connected to servers
>> I am looking for a ballpark number in terms of required man-years of
>> engineering
>> Thanks,
>> -michel
> --
> Michel DENIS
> Président, Internet 3 Solutions SAS
> Tél: 06.72.87.17.50michel.de...@internet3solutions.comhttp://www.internet3solutions.com/
> Mmm good point I think eventually will be sopported but in 18 months i > doubt. Google Chrome comes with a software implementation of WebGL to > help a little. Ipad now supports WebGL, so i think is mean of time
> On Mon, Jul 23, 2012 at 3:05 PM, Michel DENIS > <michel.m.de...@gmail.com <mailto:michel.m.de...@gmail.com>> wrote:
> Hi Carlos,
> Do you know whether WebGL be supported on all PC browsers and
> tablets' interfaces in the coming 18 months ? That would help.
> Thanks,
> -michel
> Carlos Rafael Ramirez wrote:
>> I have time thinking in implement a webgl client using GWT to
>> compile the actial OWL client and websockes to communicate, but I
>> am still joining the pieces, and estimating maybe two years of
>> full work until a such client reach the users.
>> The lowend PC clients and tablets I have herad people that are
>> using VNC for this, but it should be formalized in order that
>> ordinary users can use it
>> But sure I want to go in this way too
>> On Mon, Jul 23, 2012 at 11:57 AM, micheldenis
>> <michel.m.de...@gmail.com <mailto:michel.m.de...@gmail.com>> wrote:
>> This might be a dream but what do you think about an OWL
>> solution combining the following:
>> - Possibility of using low-end PC clients to access OWL servers
>> - Possibility of using any tablet to access OWL servers
>> - An OWL which would be fully web-based, possibly using
>> clouds, proxies, etc...
>> - Cancelation of all current constraints to open ports on
>> clients sites
>> - And of course a solution providing same functionality,
>> performance, scalability, usability, .... as with
>> conventional heavy Java OWL clients connected to servers
>> I am looking for a ballpark number in terms of required
>> man-years of engineering
>> Thanks,
>> -michel
> -- > Michel DENIS
> Pr�sident, Internet 3 Solutions SAS
> T�l: 06.72.87.17.50
> michel.de...@internet3solutions.com <mailto:michel.de...@internet3solutions.com>
> http://www.internet3solutions.com/
> No virus found in this message.
> Checked by AVG - www.avg.com <http://www.avg.com>
> Version: 2012.0.2197 / Virus Database: 2437/5149 - Release Date: 07/23/12
Basically, it is unlikely that a non-java client is going to be able to
understand the protocol that OWL uses (serialized java classes that can
change at any time). Also, to make any of the ways of rendering that you
are talking about (webgl, etc) someone will have to create an opengl based
client.
Morris
On Mon, Jul 23, 2012 at 4:01 PM, Carlos Rafael Ramirez
<crrami...@gmail.com>wrote:
>> Sounds like an enhanced client/server protocol to remove the data
>> independence issue that using serialized java classes creates and a opengl
>> (es) renderer that displays based on the content of the world rather than
>> just rendering would be in order. Maybe someone already did that.
>> Morris
>> On Mon, Jul 23, 2012 at 1:48 PM, Roland Sassen <sas...@thinsia.com>wrote:
>>> Only WonderSchool can do this at the moment,
>>> I think,
>>> cheers Roland
>>> On 23-7-2012 18:57, micheldenis wrote:
>>> This might be a dream but what do you think about an OWL solution
>>> combining the following:
>>> - Possibility of using low-end PC clients to access OWL servers
>>> - Possibility of using any tablet to access OWL servers
>>> - An OWL which would be fully web-based, possibly using clouds, proxies,
>>> etc...
>>> - Cancelation of all current constraints to open ports on clients sites
>>> - And of course a solution providing same functionality, performance,
>>> scalability, usability, .... as with conventional heavy Java OWL clients
>>> connected to servers
>>> I am looking for a ballpark number in terms of required man-years of
>>> engineering
On Mon, Jul 23, 2012 at 3:38 PM, Morris Ford <morrishf...@gmail.com> wrote:
> Basically, it is unlikely that a non-java client is going to be able to
> understand the protocol that OWL uses (serialized java classes that can
> change at any time). Also, to make any of the ways of rendering that you
> are talking about (webgl, etc) someone will have to create an opengl based
> client.
> Morris
> On Mon, Jul 23, 2012 at 4:01 PM, Carlos Rafael Ramirez <
> crrami...@gmail.com> wrote:
>> Morris sorry but I couldn't understand what you said
>> On Mon, Jul 23, 2012 at 1:44 PM, Roland Sassen <sas...@thinsia.com>wrote:
>>> Sounds like an enhanced client/server protocol to remove the data
>>> independence issue that using serialized java classes creates and a opengl
>>> (es) renderer that displays based on the content of the world rather than
>>> just rendering would be in order. Maybe someone already did that.
>>> Morris
>>> On Mon, Jul 23, 2012 at 1:48 PM, Roland Sassen <sas...@thinsia.com>wrote:
>>>> Only WonderSchool can do this at the moment,
>>>> I think,
>>>> cheers Roland
>>>> On 23-7-2012 18:57, micheldenis wrote:
>>>> This might be a dream but what do you think about an OWL solution
>>>> combining the following:
>>>> - Possibility of using low-end PC clients to access OWL servers
>>>> - Possibility of using any tablet to access OWL servers
>>>> - An OWL which would be fully web-based, possibly using clouds,
>>>> proxies, etc...
>>>> - Cancelation of all current constraints to open ports on clients sites
>>>> - And of course a solution providing same functionality, performance,
>>>> scalability, usability, .... as with conventional heavy Java OWL clients
>>>> connected to servers
>>>> I am looking for a ballpark number in terms of required man-years of
>>>> engineering
When we started Wonderland, we made a big bet: that mobile Java code was
the best way to share the behavior of objects in a virtual world. Any time
we talk about porting to a different client architecture, that means
re-writing all the client-side behavior in some other system. That isn't
trivial -- the core of Wonderland is about 300k lines of client code, which
would all have to be re-written (for reference, there are only about 50k
lines of server code). If you go with a native solution, the code
potentially has to be rewritten for every platform. Even when we had a
full development team at Sun working on Wonderland, this was not an
affordable option.
In terms of your specific requirements:
- Possibility of using low-end PC clients to access OWL servers
A low-end PC with bad 3D graphics won't be able to display a complex 3D
world, regardless of what technology you choose. The only solutions in this
case are to offload the rendering to a server (which has serious
scalability issues) or to create a massively simplified view of the world
(which is a large project, and requires lots of versions of all artwork).
> - Possibility of using any tablet to access OWL servers
Tablets actually have decent 3D graphics these days. Running on an iPad
would be a major challenge, since the entire client would have to be
rewritten. Android might be more feasible, but still a large project.
> - An OWL which would be fully web-based, possibly using clouds, proxies,
> etc...
This should be a reasonable goal. The first step would be to port all
networking in Wonderland to use websockets instead of proprietary TCP
connections. This should fix most firewall problems. Audio scalability
could become an issue though, since it will be using TCP instead of UDP.
There are also lots of optimizations around distributing code and content
using a real cloud distribution system. This is more of a configuration
change than a full re-write.
> - Cancelation of all current constraints to open ports on clients sites
Porting to use websockets would have a reasonable chance of achieving this,
assuming most firewalls and proxies support websockets.
> - And of course a solution providing same functionality, performance,
> scalability, usability, .... as with conventional heavy Java OWL clients
> connected to servers
Putting all these requirements together, I could imagine a more web-centric
version of Wonderland. The client behavior would still be defined in a
cross-platform way (probably Javascript), but the underlying system would
have to be re-written for different hardware and operating systems. This is
very similar to PhoneGap, a system of developing software for lots of
different types of mobile phones using Javascript. Building a system like
that would be a very large undertaking, probably comparable to the original
Wonderland effort (~20 man-years).
On Mon, Jul 23, 2012 at 11:52 AM, micheldenis <michel.m.de...@gmail.com>wrote:
> When you have time, please could you tell about your feelings about such a
> project ? From all I've read about OWL, it has everything in its
> foundations to enable what I suggest, maybe with lots of engineering but ..
> on the principle is it true or false ?
> -michel
> Le lundi 23 juillet 2012 20:40:37 UTC+2, micheldenis a écrit :
>> As we know it for a long time now, OWL is more advanced than most of
>> the VW platforms on the marketplace, in a number of areas.
>> Still we know that the future for VW is the Web, and platforms (e.g.
>> gaming ones) already exist in this area, but missing collaborative tools
>> that OWL has for a long time now.
>> The architecture foundations of OWL enable all this ...
>> -michel
>> Carlos Rafael Ramirez wrote:
>> I have time thinking in implement a webgl client using GWT to compile the
>> actial OWL client and websockes to communicate, but I am still joining the
>> pieces, and estimating maybe two years of full work until a such client
>> reach the users.
>> The lowend PC clients and tablets I have herad people that are using
>> VNC for this, but it should be formalized in order that ordinary users can
>> use it
>> But sure I want to go in this way too
>> On Mon, Jul 23, 2012 at 11:57 AM, micheldenis <michel.m.de...@gmail.com>wrote:
>>> This might be a dream but what do you think about an OWL solution
>>> combining the following:
>>> - Possibility of using low-end PC clients to access OWL servers
>>> - Possibility of using any tablet to access OWL servers
>>> - An OWL which would be fully web-based, possibly using clouds, proxies,
>>> etc...
>>> - Cancelation of all current constraints to open ports on clients sites
>>> - And of course a solution providing same functionality, performance,
>>> scalability, usability, .... as with conventional heavy Java OWL clients
>>> connected to servers
>>> I am looking for a ballpark number in terms of required man-years of
>>> engineering
> When we started Wonderland, we made a big bet: that mobile Java code > was the best way to share the behavior of objects in a virtual world. > Any time we talk about porting to a different client architecture, > that means re-writing all the client-side behavior in some other > system. That isn't trivial -- the core of Wonderland is about 300k > lines of client code, which would all have to be re-written (for > reference, there are only about 50k lines of server code). If you go > with a native solution, the code potentially has to be rewritten for > every platform. Even when we had a full development team at Sun > working on Wonderland, this was not an affordable option.
> In terms of your specific requirements:
> - Possibility of using low-end PC clients to access OWL servers
> A low-end PC with bad 3D graphics won't be able to display a complex > 3D world, regardless of what technology you choose. The only solutions > in this case are to offload the rendering to a server (which has > serious scalability issues) or to create a massively simplified view > of the world (which is a large project, and requires lots of versions > of all artwork).
> - Possibility of using any tablet to access OWL servers
> Tablets actually have decent 3D graphics these days. Running on an > iPad would be a major challenge, since the entire client would have to > be rewritten. Android might be more feasible, but still a large project.
> - An OWL which would be fully web-based, possibly using clouds,
> proxies, etc...
> This should be a reasonable goal. The first step would be to port all > networking in Wonderland to use websockets instead of proprietary TCP > connections. This should fix most firewall problems. Audio scalability > could become an issue though, since it will be using TCP instead of UDP.
> There are also lots of optimizations around distributing code and > content using a real cloud distribution system. This is more of a > configuration change than a full re-write.
> - Cancelation of all current constraints to open ports on clients
> sites
> Porting to use websockets would have a reasonable chance of achieving > this, assuming most firewalls and proxies support websockets.
> - And of course a solution providing same functionality,
> performance, scalability, usability, .... as with conventional
> heavy Java OWL clients connected to servers
> Putting all these requirements together, I could imagine a more > web-centric version of Wonderland. The client behavior would still be > defined in a cross-platform way (probably Javascript), but the > underlying system would have to be re-written for different hardware > and operating systems. This is very similar to PhoneGap, a system of > developing software for lots of different types of mobile phones using > Javascript. Building a system like that would be a very large > undertaking, probably comparable to the original Wonderland effort > (~20 man-years).
> On Mon, Jul 23, 2012 at 11:52 AM, micheldenis > <michel.m.de...@gmail.com <mailto:michel.m.de...@gmail.com>> wrote:
> Jonathan,
> When you have time, please could you tell about your feelings
> about such a project ? From all I've read about OWL, it has
> everything in its foundations to enable what I suggest, maybe with
> lots of engineering but .. on the principle is it true or false ?
> -michel
> Le lundi 23 juillet 2012 20:40:37 UTC+2, micheldenis a �crit :
> As we know it for a long time now, OWL is more advanced than
> most of the VW platforms on the marketplace, in a number of areas.
> Still we know that the future for VW is the Web, and platforms
> (e.g. gaming ones) already exist in this area, but missing
> collaborative tools that OWL has for a long time now.
> The architecture foundations of OWL enable all this ...
> -michel
> Carlos Rafael Ramirez wrote:
>> I have time thinking in implement a webgl client using GWT to
>> compile the actial OWL client and websockes to communicate,
>> but I am still joining the pieces, and estimating maybe two
>> years of full work until a such client reach the users.
>> The lowend PC clients and tablets I have herad people that
>> are using VNC for this, but it should be formalized in order
>> that ordinary users can use it
>> But sure I want to go in this way too
>> On Mon, Jul 23, 2012 at 11:57 AM, micheldenis
>> <michel.m.de...@gmail.com <mailto:michel.m.de...@gmail.com>>
>> wrote:
>> This might be a dream but what do you think about an OWL
>> solution combining the following:
>> - Possibility of using low-end PC clients to access OWL
>> servers
>> - Possibility of using any tablet to access OWL servers
>> - An OWL which would be fully web-based, possibly using
>> clouds, proxies, etc...
>> - Cancelation of all current constraints to open ports on
>> clients sites
>> - And of course a solution providing same functionality,
>> performance, scalability, usability, .... as with
>> conventional heavy Java OWL clients connected to servers
>> I am looking for a ballpark number in terms of required
>> man-years of engineering
>> Thanks,
>> -michel
> -- > Michel DENIS
> Pr�sident, Internet 3 Solutions SAS
> T�l: 06.72.87.17.50
> michel.de...@internet3solutions.com <mailto:michel.de...@internet3solutions.com>
> http://www.internet3solutions.com/
> No virus found in this message.
> Checked by AVG - www.avg.com <http://www.avg.com>
> Version: 2012.0.2197 / Virus Database: 2437/5153 - Release Date: 07/24/12
