steve@apple.com (Steve Savitzky) (12/08/90)
I found Arnaud's view of CyberSpace to be very interesting; it meshes in many ways with my own. But I would like to argue strongly against his contention that everyone should share the same view of virtual objects. I believe that such things as fonts, document formats, icons, images, and whatever should be *hints*, not demands. For example, I am left-handed and nearsighted. I would like to be able to view a document that presents itself as a virtual book by holding it in my right hand and flipping pages with my left, which means that I want page 1 to be in the "back". I want the font to be 12-point, not the 10-point that is more usual for books. I don't like sans-serif fonts in which I can't distinguish "i", "I", "l", and "1". And so on. A blind person would have even more drastic preferences. Similarly, when meeting people in a virtual environment, I would generally want to see them in the form they intend to present, but I might occasionally prefer to present a particular politician with the head of a toad. Since I am poor at remembering names, I would like to display virtual nametags on the people I meet. Different people might want to see different information on such nametags. Preference arguments apply even more strongly to commands (gestures, magic words, virtual tools, and the like). I currently have a rather large and very idiosyncratic collection of private shell aliases and emacs key-bindings; I would expect no less in a virtual environment. In short, a virtual reality should, as much as possible, separate content from appearance. This has the desirable side-effect of greatly decreasing bandwidth, and putting the burden of presentation on the user's local equipment, thus giving users the the option of paying for only the level of presentation they actually need. (We should neither exclude blind users, nor expect them to acquire eyephones they can't use and image data they don't need.) This said, there are of course cases in which a shared view will be expected. It would probably be disconcerting if two people sharing a virtual office disagreed about where things were in it, although they might indeed prefer different colors on the walls. Games and other simulations would probably enforce a uniform view as well. Sometimes I might even want to view a document with all of its author's formatting hints, especially if I knew that that author's designs were particularly innovative or especially beautiful. -- \ --Steve Savitzky-- \ ADVANsoft Research Corp \ REAL hackers use an AXE! \ \ steve@advansoft.COM \ 4301 Great America Pkwy \ #include<disclaimer.h> \ \ arc!steve@apple.COM \ Santa Clara, CA 95954 \ 408-727-3357 \ \__ steve@arc.UUCP _________________________________________________________
brucec%phoebus.labs.tek.com@RELAY.CS.NET (Bruce Cohen;;50-662;LP=A;) (12/11/90)
In article <12618@milton.u.washington.edu> arc!steve@apple.com (Steve Savitzky) writes: > I found Arnaud's view of CyberSpace to be very interesting; it meshes > in many ways with my own. But I would like to argue strongly against > his contention that everyone should share the same view of virtual > objects. > I agree with Steve's argument completely. There are simply too many differences between individual people to expect that one view of reality will fit all of us. Hell, we don't really share one view of reality even when talking about consensus reality! Another strong argument against an (enforced) common interface is that there will be pressure to keep it from changing, even if there are good reasons for it to do so. I doubt very much that any standards body (and I've been on several of them) will be able to specify an interface which can grow in unexpected directions. And this sort of evolution is exactly what's making VR possible in the first place. > This said, there are of course cases in which a shared view will be > expected. It would probably be disconcerting if two people sharing a > virtual office disagreed about where things were in it, although they > might indeed prefer different colors on the walls. Games and other > simulations would probably enforce a uniform view as well. Sometimes > I might even want to view a document with all of its author's > formatting hints, especially if I knew that that author's designs were > particularly innovative or especially beautiful. Yes, but notice that the final choice is the user's, NOT the author's or the programmer's. The way to deal with this problem is to develop protocols which allow an interface designer to specify a recommended level of commonality, or even several, with enumerations of the attributes which should be used as given at each level. -- ------------------------------------------------------------------------ Speaker-to-managers, aka Bruce Cohen, Computer Research Lab email: brucec@tekchips.labs.tek.com Tektronix Laboratories, Tektronix, Inc. phone: (503)627-5241 M/S 50-662, P.O. Box 500, Beaverton, OR 97077
gourdol@imag.imag.fr (Gourdol Arnaud) (12/15/90)
In article <12618@milton.u.washington.edu> arc!steve@apple.com (Steve Savitzky) writes: >I believe that such things as fonts, document formats, icons, images, >and whatever should be *hints*, not demands. For example, I am >left-handed and nearsighted. I would like to be able to view a >document that presents itself as a virtual book by holding it in my >right hand and flipping pages with my left, which means that I want >page 1 to be in the "back". I want the font to be 12-point, not the >10-point that is more usual for books. I don't like sans-serif fonts >in which I can't distinguish "i", "I", "l", and "1". And so on. Ok, ok ! I agree with you. You have a point. But I think that the examples you are giving are good, and some other may be not so good. Let's admit that you can wear some virtual glasses that can "correct" in some way your view of the world, but not drastically change it, and always leave you a way to quickly and easily get back to the original message. >A blind person would have even more drastic preferences. I don't think he could use visual CyberSpace as we imagin it today, although we might use big-sized braille pads, or have some totally different environnement, maybe based on sound which is a sense that blind people have usually well developped. Steve says he wants to put labels on people with their names, prefered brand of beer and the like (I extrapolate)] It's Ok. Don't you think that a pair of glasses would do the trick ? As to really change the appearance I disagree, I find it unfair, it's like using the y-ray glasses that allow you to see people naked even when they're clothed. >Preference arguments apply even more strongly to commands (gestures, >magic words, virtual tools, and the like). [...] Once again, it's ok. I do not say that we must have an uniformized common virtual reality. Have your own tool set. You use it, you're happy with it. Fine. My point is that the view of objects we share should be the same. If I send you a message in a given form, receive it that way. [Ok, now I'm a refering to an article posted by someone else beuuuuuh but being NOT an UNIX/rn/e-macs/tty (I don't know which is the culprit (and don't say it's me :-> (and by the way having no memory either (I would surely need glasses wuth nametags), I am not able to post parts of this article here, not to cite his author] I do not think (IMHO) that we should have "laws" or "rules" or "security mechanisms" to protect our messages from being viewed differently. If you want by some magical/electronical effect change the picture of myself I put in my letter as a picture of myself nude and send it to alt.sex.bestiality (where it probably belong :-) I am not going to charge you in law (is that correct english ?, I mean ask justice to suspend your CyberSpace account and copy one thousand time (by hand) "I shall not play with my little friend's picture".) Well, as long as it's funny, and non diffamatory (does this word exists ? I feel great tonigth). If had not been such a UNIX hostile I would have put some part of your message there. I don't think you would have flammed me for this, neither if I used 9point italic instaead of you 12 point. The point is: If I send you some information that is non-textual but embedded in some textual information (page layout, etc) the CyberSpace should send it to you that way, and you should receive it that way, even if you alter it after (on whatever purpose you like). BTW, I think that doing such kind of "automatic" translation will be very-very difficult. Think of it: We can't do that today as soon as it's just a bit complicated. An example: A friend of mine tried to import a Word RTF document into WriteNow for the Mac. The text was here but all the formatting was... strange. The fonts were incorrect, etc. Think of it: Today there is no good way for blind people to use graphic-based computers. We are just unable to translate all those nice graphics into something they can perceive. The only solution I see for CyberSpace is to work very hard in healt science or electronic implants so that there will be no blind when CyberSpace will really exist. (However, I sincerely hope that we will not need such excuses to make those progresses). [Now, I'm slighly disagreeing with Steve's example] [ Sorry if I can't put it there, I am really NOT an e-macs guru and I'm lost if I do not have cut/copy/paste] If we are living in the same virtual office, let's have the walls the same color, the furniture the same place... Imagine you ask me "Where is the backup of this article you wrote about Antarctica ?". And I reply "Oh, it's in the file cabinet near the window with blue curtains (Private joke for ML)", and that in your view there IS a windows with blue curtains but that's not the same as mine. Imagine what a mess !! >I might even want to view a document with all of its author's >formatting hints, especially if I knew that that author's designs were >particularly innovative or especially beautiful. I agree very much with you here. >-- >\ --Steve Savitzky-- \ ADVANsoft Research Corp \ REAL hackers use an AXE! \ > \ steve@advansoft.COM \ 4301 Great America Pkwy \ #include<disclaimer.h> \ > \ arc!steve@apple.COM \ Santa Clara, CA 95954 \ 408-727-3357 \ > \__ steve@arc.UUCP _________________________________________________________ [Plus an anonymous author to which I apologise ] Arnaud.-- /======================//==========================================/ / Arnaud Gourdol. // On the netland : gourdol@imag.fr / /======================//==========================================/