Archive of UserLand's first discussion group, started October 5, 1998.
Re: ZDNet Fell for DigiScents
Author: Jeremy Bowers Posted: 10/15/1999; 6:36:49 AM Topic: Today's scriptingNews Outline Msg #: 12062 (In response to 12061) Prev/Next: 12061 / 12063
It's sort of funny. If I remember my bio-psychology class correctly from a year ago, there are thought to be 1000-10000 distinct types of smell detectectors in your nose, with very little overlap, and that a whopping number of them, as high as the low double digits, (that's sarcasm) have been actually identified, and thus possibly available for such an application.Note that those identified are all extremely unpleasent, as those are the easiest to get a strong reaction from. Therefore, the only thing we could build now would deliver an extremely small array of extremely unpleasent smells, or would simply have to be loaded with every type of scent you want to use, which is hardly a breakthrough.
That's why this is impossible: There are 4 basic types of detectors in the human eye, so we can fool it with 3 colors and some mixing of them on your monitor. The ear basically detects one things and detects it well, and it happens we can fairly easily generate that one thing. The nose detects thousands of independent chemicals, and how much mixing occurs versus how much is just which things are triggered is totally unknown. Thus, you can't just get away with having 3 or 4 chemicals and mixing them, you need a full organic synthesis factory on your desktop, and we still can't effectively generate som common scents anyhow, even if a factory fit on your desktop, which it doesn't.
Waiting for the hoax admission... or, alternatively, the implementation complete with
Scratch here to smell a rose! tags...(BTW, I'm having some trouble posting... it says the document contains no data, yet my posts and the changes are going through...)
This page was archived on 6/13/2001; 4:53:04 PM.
© Copyright 1998-2001 UserLand Software, Inc.