Path: utzoo!utgpu!news-server.csri.toronto.edu!rutgers!uwm.edu!ogicse!milton!schraudo%beowulf@ucsd.edu From: schraudo%beowulf@ucsd.edu (Nici Schraudolph) Newsgroups: sci.virtual-worlds Subject: Re: "Space" Message-ID: Date: 28 Aug 90 20:02:15 GMT References: <9007250107.AA01311@hitl.vrnet.washington.edu> Sender: hlab@milton.u.washington.edu Lines: 105 Approved: hitl@hardy.u.washington.edu wex@dali.pws.bull.com (Buckaroo Banzai) writes: >In article schraudo%beowulf@ucsd.edu (Nici Schraudolph) writes: > there is altitude, but gravity and visual clues make the orientation > problem along that dimension trivial: you always know which way the > ground is. >This is more or less the point Meredith Bricken made. I see your point, but >I don't know any of these thing when I'm playing the various flight >simulators (particularly in those modes that remove external cues like the >horizon), and I don't do too badly. Ah, but even then your instrumentation will have a "2+1 D" setup, with a compass for two dimensions and an altimeter for the third. Or do you have a simulator that allows you to navigate by 3-D compass alone? -------- >Besides, what's to stop us putting in all kinds of cognitive artifacts to >help us out? In fact, I advocate doing just that. Landmarks are just one >example. Oh, I agree totally - in fact my virtual horizon suggestion is just another such artifact. My point is that artifacts that humans know intuitively how to deal with are preferable to those that come with a 500-page manual. Isn't one of the main selling points of VR that it allows you to make very abstract operations concrete and intuitive? > 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. >Again, I'm dubious about any claim that includes the word "obvious." What >data do you base this idea on? I bet I can make a cybercompass that's >better than your zero-grid, particularly when we're trying to find a place >or object that's described by more than three interesting properties. You're right in that artifacts can of course make navigation easier, after all that's the whole point of having them. What I was trying to say is that if I can find data from a DB using complex navigation artifacts with interface A, but can do the same job with less or no artifacts using interface B, I'd say that B is better than A. I am not suggesting that you shouldn't have all sorts of artifacts, just that if B is so powerful even without them it will be even better with them. -------- > 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! >Sorry, but that's simply not the case. Flowcharts, trees, graphs, et al >were invented *long* before there were million LOC programs. Indeed, at MCC >we did field studies and found engineer after engineer who had been in the >business for decades and who were growing increasingly frustrated that their >time-tested methods simply couldn't help them with the increasingly complex >problems they were facing. > 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. >I don't find anything intuitive about a flowchart or a dataflow diagram or >an entity-relation diagram. If you've got some other meaning of intuitive, >or if you're talking about some other "representations" or "structures" >please let me know. OK, I see the problem... to me as computer science type a "data structure" is an abstract mathematical entity, namely a set of data objects and a set of operations defined on these objects. To me, any kind of diagram is not a data structure since no operations are defined on it. Thus when I say "graph" I mean a set of nodes and edges, plus operations to traverse the structure, not some kind of diagram with circles and arcs. That's just visualization. > 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. >OK, let's start with fisheye views. If that's too easy for you, try >generalized degree-of-interest functions. These also seem to be data transformations rather than full-blown data structures in their own right to me. In particular, you have toimplement them on top of some existing data structure in order to access the data you want to transform in the first place. >In a sense it's a losing game for me in that if I ever hope to implement my >theories I'm going to have to do it in terms of computationally-realizable >structures, any of which are amenable to multiple conventional and >unconventional representations. Exactly. It seems that we were mostly having terminology problems, which have caused both of us to read claims into each other's posts that were not there... peace? :-) -- Nicol N. Schraudolph, C-014 "Big Science, hallelujah. University of California, San Diego Big Science, yodellayheehoo." La Jolla, CA 92093-0114 - Laurie Anderson. nici%cs@ucsd.{edu,bitnet,ucsd}