[prev in list] [next in list] [prev in thread] [next in thread] 

List:       kde-devel
Subject:    Re: integrating virtual reality into KDE
From:       Raphael Langerhorst <raphael-langerhorst () gmx ! at>
Date:       2004-08-15 19:49:43
Message-ID: 200408152149.43889.raphael-langerhorst () gmx ! at
[Download RAW message or body]

Hi all,

thanks for your responses. Especially this one holds some questions.

Just some general reply: the G System is less about data transfer (kio slave). 
The KDE integration would more be for the communicational capabilities of the 
G System [in a 3D virtual environment, which would be an interesting addition 
to Kopete]. That's of course not what the G System is limited to but what 
(one of) the purpose of "KDE" integration would be - or the purpose of a 
"online game experience" for the game module - or as a scientific playground 
for a science module... whatever would be of interest then.

About network lags/disconnects/... world content loss / ... We have a few 
thoughts about that which would minimize such failures, but no implementation 
- as the network layer itself is still missing.

Human beings: they would play part in the simulation just as anything else. Of 
course they have more power in terms of individual will, but ARE restricted 
to the laws of the simulation, which are among other things of 
evolutional/... type. Concering evolution: there is not much about evolution 
integrated in the system yet, this will come more and more as the underlying 
infrastructure is provided.

Overall I agree with you that it is just too premature to think about an 
integration. I also wanted to "taste the waters" of the KDEers, what they 
think about this, so thanks to all for their replies & thoughts. We will 
first provide a standalone client, what happens then is not a question now.

Regards,
Raphael


On Friday 13 August 2004 06:23, Allan Fields wrote:
> [ Please split responses to this thread to appropriate lists,
>   it's cross-posted.. ]
>
> On Thu, Aug 12, 2004 at 07:58:02AM +0200, Raphael Langerhorst wrote:
> > Hi all!
> >
> > This relates to this: http://dot.kde.org/1090909286/
> > (For more info: http://www.g-system.at)
>
> This has got to be the most bizarre/geeky kio slave suggestion yet!
> =) I'm surprised no one has come up with a kio slave for toasters.
>
> > What do you think about having a client interface for this virtual
> > reality in KDE? In particular the question "where to put it" also arises
> > [games,edu, a new science module, or network, ..].
>
> But in all seriousness:
>
> Why do you want it integrated with KDE?  For example: do you think
> it's a good match to konqueror in its current state?  If not, do
> you know what you would like to see KDE/konqueror support for
> navigation of abstract "virtual" spaces?  Will a new kio slave alone
> be enough?
>
> > This is what I have in mind:
> >
> > The server system will be a losely distributable system. This means that
> > there probably won't be a central server cluster - which we couldn't
> > afford right now anyway. So it's possible to drive the server system by
> > the community, or "users" - those with a little bit of bandwith
> > available. The world itself is
>
> So in a distributed system you could simply have an irc model with
> multiple servers linked together.  But, we all know the issues with
> such networks, as they can get out of sync and then suddenly the
> world is split.  So synchronization issues are of importance here,
> where presumably you can't pause the world to wait for the rest of
> it to rejoin, otherwise lacking consistency.  How will you deal
> with the disappearance of elements/members from the system?
>
> Wouldn't you want a timestamp/ordering of interactions so that if
> some arrive out of order you can create a hick-up in the fabric of
> space-time?  Or would lost interactions, simply be dropped?
> What are the implications of this taking into account the cascading
> effect in system simulations.
>
> P2P suggests a decentralized organization, and may be better suited,
> but might require built-in redundancies of magnitude matching the
> interlinkage (reference count) on a given particle?  If a very
> central element suddenly disappeared the disturbance would be
> significant and might even throw the system into disequilibrium.
> (Oops, netsplit, the Sun just disappeared.)
>
> > virtual world rather than "classical" online games]
> >   * this would also serve as a meeting place for people - a kind of
> > Kopete.
>
> Are humans seeding actors/divine steering forces?  What form do they take
> in your space?
>
> How does this compare to automata games?
>
> > Regarding the range of purposes this VR (virtual reality) serves I'm not
> > sure where this would go. The base system (and server system) will stay
>
> How about just making one client a separate KDE application at
> first? If it lends naturally to integration, then consider proposing
> how it would fit cleanly into the desktop and other desktop systems
> (even ones starting with G ;).
>
> > What do you think?
>
> Interesting, but very abstract.
>
> > What do you think about this integration in general?
>
> Not to be discouraging, but I would think slightly too premature
> at this point.
>
> > Regards,
> > Raphael
>
> Just my opinions..
 
>> Visit http://mail.kde.org/mailman/listinfo/kde-devel#unsub to unsubscribe <<
[prev in list] [next in list] [prev in thread] [next in thread]