From doherty;at;msc.edu Wed Feb 2 13:46:39 1994 Received: from noc.msc.edu for doherty _-at-_)msc.edu by www.ccl.net (8.6.4/930601.1506) id NAA06284; Wed, 2 Feb 1994 13:15:59 -0500 Received: from uh.msc.edu by noc.msc.edu (5.65/MSC/v3.0.1(920324)) id AA06292; Wed, 2 Feb 94 12:15:58 -0600 Received: by uh.msc.edu (5.65/MSC/v3.1r(920220)) id AA09358; Wed, 2 Feb 94 12:15:57 -0600 From: doherty _-at-_)msc.edu (David C. Doherty) Message-Id: <9402021815.AA09358[ AT ]uh.msc.edu> Subject: Re: CCL:Electronic Publishing To: 100012.1163#* at *#CompuServe.COM (Rainer Stumpe 100012 1163) Date: Wed, 2 Feb 1994 12:15:55 -0600 (CST) Cc: chemistry /at\ccl.net In-Reply-To: <940201201440_100012.1163_BHB75-1 ^at^ CompuServe.COM> from "Rainer Stumpe 100012,1163" at Feb 1, 94 03:14:41 pm X-Mailer: ELM [version 2.4 PL22] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 2391 Rainer Stumpe wrote: > >All this discussion started with the question of color plates. >Let me comment my personal view. And I am limiting myself to >sciences (not arts, not medicine). Color may distract attention >or focus it on a minor issue (I use color transparencies in >presentations if I have something to hide). Black-and-white >line drawings have their advantage over multicolored area >graphs (not only in sciences): the viewer cannot easily be >cheated, the psychology is better known (e.g. using logarithmic >scale). I do believe, color should be used sparsely and applied >only where necessary. Would you prefer watching the movie >"Casablanca" in color? > > Colorized Bogart aside, this is a specious argument against the use of color. It makes the assumption that the mapping of color to the quantities of interest is indeterminate. This need not be the case. If the color scheme is carefully chosen (and described to the audience), a color presentation can reveal important aspects of a set of data which cannot be seen otherwise. In my own work (on phase transitions of crystalline polymers), I have found the use of color visualization tools to be invaluable in identifying phenomena (soliton wave propagation as an example) which I couldn't find in other ways. The use of gray-scale version of a figure in a recent paper of mine was necessitated by outrageous (IMHO) page and reprint charges for color expected by the journal (PRL). This is unfortunate for the reader because the color version was much more effective in conveying the same point. It is true that color can be used to "hide" or overemphasize certain aspects of the data. Intentional use in this manner is simply unscrupulous; naive misuse should be identified and corrected. But I don`t believe that color needs to be singled out in these instances any more than a poorly described (or applied) theory. The use of color to present data requires that its use be appropriate, well-described and understood. This is no different than anything else that we do. Dave Doherty ps. I can provide postscript versions of the aforementioned figure upon request. Inspection of the gray-scale and color versions should demonstrate my points. Of course, reading the paper to understand what they represent is highly advised ;-). -- David C. Doherty Minnesota Supercomputer Center doherty;at;msc.edu