Re: [-empyre-] Aesthetics of single user works / versus multi user works

Hi Adam -

In my experience of abstract and semi-abstract 3D worlds the act of navigation helps the user tremendously to understand the structure of the 3D space, an aid to which passive viewers - even if they're physically present - don't have access unless they are directly hovering over the user and kibbitzing ("try going left - further, further - now go forward" etc.) So for them the experience can be confusing because it seems no different from a film strip, since they have no experience of cause and effect and little sense of spatial relationships in the 3D space.

In my very first multi-user VR world, for an impatient and 3D-naive client, this unfortunately meant that we canned the abstract worlds we started with and gave them realism instead - not because the ultimate "users" - kids - needed it, but because the ones paying for it couldn't deal with it and weren't going to touch the mouse themselves.

I would guess that the localized 3D sound you use in your piece would contribute substantially as an orientation help, however - what is your experience here?

- tamiko

Adam Nash wrote:
Tamiko said:

I think to make a successful multi-user work you really need to understand what "added value" is provided by other users in the space, and orient your piece towards this. In the best of circumstances the other users are an enrichment; all too often they are an annoying distraction, because they are there with some other agenda. ("Hey I'm new to VRML how did you build this" is a fine conversation to build the bonds of the VRML user crowd, but it doesn't add to your experience of the virtual artwork.

Hi Tamiko, everyone,

This is true of works that are conceived as single-user, and most art whether virtual or not, is indeed single-user, even if it takes place in front of a room full of people. As you note, a work like Beyond Manzanar probably wouldn't benefit from being made multi-user. As for online 3D works designed for multi-user, most of these until now have really been chat rooms with 3D graphics added on. In this case, the multi-user aspect is crucial but it is the 3D itself that is of questionable use.

I'm interested in the multi-user space as a live performance medium - whilst this is obviously not a new idea, not very much work has been done in that area. Something I quickly found to be problematic was the avatar-as-human idea. Trying to treat the space as if it were a regular physical space in which bodies perform introduces massive problems of two major types. First the time and effort required on the modeling and animation (with its attendant problem of large filesize), and second the audience expectation. It seemed natural to remove these problems completely, and use the space for its strengths rather than weaknesses. When trying to emulate physical space, the lack of gravity and economy of geometry is a weakness, but when the piece is abstract these qualities can be used as part of the performance.

It also leads to the questioning of what an avatar is. Again, analogy with physical space quickly falls apart. In the version of VNet that I have been using, switching to OOBE mode ("Out Of Body Experience", ie, one can see one's own avatar from an objective perspective) causes some very interesting effects. Sure, you are looking at your representation in space, but if you move around whilst in this mode, your avatar does not, begging the question of which one is you - the avatar designated as you or the position from which you are looking at the space? There are many other capabilities that in physical space would be considered anomalies, but to me they all point to the conclusion that the virtual 3D space isn't really very similar to physical 3D space, and this is a very freeing realisation from a performative point of view.

In the multi-user version of Memory Plains Returning, that we performed last week, the avatars *become* the space, or at least their visual and sonic manifestation is so large that they seem to. The most common comment offered by audience members logging in at the start was that they couldn't see anything. This was the equivalent of an empty stage or a silent concert hall. The action doesn't start till the performers do. I conducted the performers, whose avatars were made up of different sections of the piece, to change the visual appearance at my command. In this way, they are not really avatars in the conventional sense at all, even though it is true that they represent the performers in that if the performer logged out, the avatar would not be there. It was a highly structured performance with little room for the performers to improvise. We are currently working on a piece where the interaction and improvisation between the avatars is crucial to the performance, as I think this will prove to be more interesting. Ultimately, the interest in last week's performance lay only in the fact that it was a "live, multi-user performance".

The piece was also being projected onto a screen for the benefit of a live, physically present audience in Folly Gallery, and Cornerhouse Gallery. This situation has some good and bad points. For the good, at Folly (from where I was logged in), I was 'presenting' in the way that Steve suggested in a post a few days ago - the live mu performance offered some difficult concepts for those new to the form and as a result we were able to have a very interesting discussion about the nature of the performance and the space itself. I was expecting a similar level of questioning from the online audience, but they were surprisingly quiet in the text chat window. In a way I was 'conducting' them also, in that I would suggest to them viewpoints to hook into as they became available according to a given avatar's manifestation. On the negative side, things didn't go so well at Cornerhouse, where one of the performers was having problems with crashing and being logged out. Preoccupied with these problems, he didn't really have the time to engage the audience and take them with him on his, albeit frustrating, journey. This also was to do with notions of 'perfection' or 'maintaining the illusion' that is typical of a lot of live performance. Unfortunately, I hadn't discussed with him sufficiently the implications of having to perform to two different audiences, online and offline, and as a result the audience present in the Cornerhouse gallery were a little mystified as to what was happening. I think this medium, particularly in these early stages, can only benefit from its 'inner workings' being exposed to the audience, as the philosophy behind the performance has so much to do with those inner workings. As I have said before on this list, I don't believe in the myth of 'suspension of disbelief' and I think the audience need to be included in the performance for it to be satisfying for them. In the case of my work, at the very least this means teaching the audience to use the veiwpoint menu for example. Ideally we'll reach a stage when the audience already know what to do when they log in, but for the moment we will have to contend with a bit of teaching. I suppose it is a bit analogous to giving directions to an unusual or distant venue in a real world performance.

The online audience, small as it was, did seem quite content to sit back and enjoy the composition, and there were many who put up with crashing and logging back in again. This crashing seems to be a combination of the flakiness of VNet and the largish amount of geometry. Some interesting temporal effects were achieved by this re-logging in, where some members of the audience (and indeed the performers themselves) were seeing different things - again an 'anomaly' that can be used within the performance. All in all, what was supposed to be a higly composed and structured piece became quite chaotic in the online performance, and whilst I personally enjoyed that very much it was confusing to some audience members. I will need to seriously research this behaviour of VNet over the coming months, although it is usually quite difficult to get more than one or two people to log in for testing. I will be posting notices to empyre and www-vrml and other lists, to try to enlist the cooperation of those 'in the know'.

I think the 3D multi-user space offers a huge potential for live performance, and my current project "Scorched Happiness" is an attempt to really thoroughly investigate the medium's properties and create a performance, unique to it, that is as satisfying as a live performance in any other medium.

Apologies for the length of this post, but I hope it is a useful contribution to the discussion of multi-user space.


-- ---------------------------------- Tamiko Thiel Media Artist


This archive was generated by a fusion of Pipermail 0.09 (Mailman edition) and MHonArc 2.6.8.