Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!sdd.hp.com!zaphod.mps.ohio-state.edu!uakari.primate.wisc.edu!dali.cs.montana.edu!milton!schraudo%beowulf@ucsd.edu From: schraudo%beowulf@ucsd.edu (Nici Schraudolph) Newsgroups: sci.virtual-worlds Subject: Re: "Space" Message-ID: Date: 13 Aug 90 00:39:31 GMT References: <9007250107.AA01311@hitl.vrnet.washington.edu> Sender: hlab@milton.u.washington.edu Lines: 74 Approved: hitl@hardy.u.washington.edu wex@dali.pws.bull.com (Buckaroo Banzai) writes: >I disagree on two points. First off, people do pretty well with flight >simulators, so we *can* go to 3D (in fact, we spend most of our waking lives >navigating in 3D). Wumpus is a kind of weird space; not very intuitive. Sorry, but humans *don't* do well in general 3D. We spend most of our waking lives on a 2D surface, and have evolved to deal with that. Sure, there is altitude, but gravity and visual clues make the orientation problem along that dimension trivial: you always know which way the ground is. Wumpus is played in a 3D space with additional contraints: the only moves allowed are those between vertices of a dodekaeder. Thus Wumpus is an instantiation of the general 3D navigation problem, and it follows that your performance at general 3D navigation can't be better than at Wumpus. I agree that the constraints are counterintuitive - indeed my whole point is that you get the best mileage by imposing intuitive constraints that are gleaned from nature. >From a cognitive perspective our natural environment might best be described as "2 + 1"D, and this concept is also beneficial when thinking about cyber- space implementations: adding a grid in the zero plane for one of the three coordinates (in case of a semantic space, preferably for the coordinate that represents the quality most "different" from the other two) will aid navigation through the VR, as will simulated gravity effects (eg. a tendency to slowly turn "right side up" in the absence of movement commands). ---- >Second, who says cyberspace should be devoid of cognitive artifacts? I >think there should be lots. Again in a past paper (with Kim Fairchild), we >talk about navigating cyberspace. To do this, we hypothesize the >construction of whole new classes of artifacts to act as navigational aids. Sure. What I'm saying is the less artifacts you need to navigate a given data structure, the better your system, because you are obviously tapping more of the intuitive human navigation potential. Thus the zero-grid suggested above is preferable to having to consult a 3D-compass to find out which way you're headed. ---- >You're confusing theory with implementation. One can implement sparse >arrays in a number of compact, useful forms. Why not do the same for >spaces? Remember, the prime purpose of semantic-space theory is to provide >a natural locative system for abstract data. Sorry, but you're the one confusing abstract data structure with implemen- tation: a sparse array is an *implementation* of a matrix. This might explain why I interpreted your first post as advocating a sparse array *visualization* approach, which I am sure you agree is the least "compact, useful form". It has become obvious now that this is not your approach. ---- >Time-honored computer science data structures just don't deal well with >huge volumes of data, and hierarchical divide-and-conquer strategies only go >so far. I'm attempting to suggest something to supplement these things. Wait a minute - time-honored computer science data structures have been (and are being) invented precisely for the purpose of dealing with huge volumes of data! Implementing visual representations of these structures is a very valuable undertaking in cognitive engineering, and will give a performance advantage by exploiting intuitive human skills such as navigation. However, you seem to claim that your work provides ways of structuring data that are new in a more fundamental, abstract sense. This is a big claim that I have yet to be convinced of. We could play a little game: you name a feature in your system, I name the data structure it is a visual imple- mentation of. I'll be pleasantly surprised if I can't answer. -- Nicol N. Schraudolph, C-014 nici%cs@ucsd.edu University of California, San Diego nici%cs@ucsd.bitnet La Jolla, CA 92093-0114 ...!ucsd!cs!nici