More neutral than almost all other gray cards

[Doug Kerr]

On C-M-Y ratios

The recent, and very interesting, discussion of "recipes" for "nice-looking" skin revolves around what are described as ratios between the C, M, and Y coordinates of the colors involved.

For openers, the "imperatives" are described in terms of "this coordinate should not be greater than that one".

But when we get more subtle, we hear that "this one should preferably be perhaps 25% of that one".

There we run into a complication. The C, M, and Y coordinate values are non-linear. (They derive from a simplistic transformation of the R, G, and B coordinates, which are themselves nonlinear. And the nature of the transform "aggravates" the nonlinearity.)

Thus, for a given chromaticity, the ratios between C, M, and Y will not be fixed, but will vary with the luminance at which we find the color.

Accordingly, for the same skin, and consistent color rendering, the ratios between C, M, and Y will differ with the luminance at which we find this skin area in the image.

For example, the three C,M,Y triples below represent exactly the same chromaticity, but the second has 1/2 the luminance. I have shown C, M, and Y on a 0-255 scale (rather than the 0-100% scale we sometimes see), and after each triple, I have shown the ratios between the C, M, and Y values, normalized to the greatest value (that is, the greatest of the values there is arbitrarily made 100%).

15,46,64 (23%,72%,100%)

80,102,116 (69%,88%,100%)

I'm delighted to see the entry of some basic "color science" into this overall discussion, but we need to be a bit careful in tossing about the concept of "target" or "typical" "C-M-Y ratios" for various genres of human skin.

Best regards,

Doug

 

[Doug Kerr]

Hi, Drew,

Of course, none of this changes the basic equations I provided above.

Having only an engineer's understanding of basic statistical concepts, and not that of a marketeer or political operative, I remain wholly unimpressed by "evaluation" equations in which the weighting factors do not add up to 1.0 (especially when that's true of only one of the equations representing competing scenarios).

Best regards,

Doug

 

[Michael Tapes]

I'm delighted to see the entry of some basic "color science" into this overall discussion, but we need to be a bit careful in tossing about the concept of "target" or "typical" "C-M-Y ratios" for various genres of human skin.

Well said, Doug.

 

[Andrew Rodney]

Well said, Doug.

Yup, this ancient and really not too useful "rule" of getting "acceptable" skin tones by using what is a highly output dependent color space, based on something most users have differently setup needs to die. It dates back to old time drum scan operators who's RGB devices could only deliver output ready CMYK for a specific print process. Years allowed them to work "by the numbers" without a display at all! The numbers were refined based upon a highly specific CMYK press. Why anyone would use a modern workflow, based on RGB working spaces (being Quasi-Device Independent), and profiled displays, then try to fix something using an output color space based on some printing press when ratio's (or rules) can be created using only three values (RGB) in a working space is beyond me. Its like the old "convert to Lab to sharpen" silliness of 1990's Photoshop techniques that needs to be buried from modern workflows.

 

[Andrew Rodney]

Yet another article on the subject:

http://www.ppmag.com/web-exclusives/2008/11/product-comparison-white-balan-1.html#more

I like the conclusions regarding the Melitta Junior Basket coffee filters for WB. That's multi-tasking!

 

[Drew Strickland]

I remain wholly unimpressed by "evaluation" equations in which the weighting factors do not add up to 1.0 (especially when that's true of only one of the equations representing competing scenarios).

Best regards,

Doug

Doug,

I made the equation out of whack intentionally to try and get you to consider it from a mathematical perspective for a moment, but then obviously reject it, and focus back on the concept I was trying to convey.

Bart's equation would not have more than one component: (Q) Image Quality.

As a manufacturer of a tool that I want to actually sell- because real people will actually use it- I had to factor in the other two variables of (S)peed and (E)ase of use while also still delivering a very high value for (Q).

Bart apparently has a lot of time on his hands and actually enjoys digging deep into these sorts of issues and learning all about them. A trait that I certainly applaud as a former instructor.

In fact, I would suggest that both you and he probably enjoy this aspect of photography much more than actually taking photographs. Again, this is a great thing. But, it is far from everything. It takes all types to make the world go 'round.

 

[Doug Kerr]

Hi, Andrew,

Yet another article on the subject:

http://www.ppmag.com/web-exclusives/2008/11/product-comparison-white-balan-1.html#more

I like the conclusions regarding the Melitta Junior Basket coffee filters for WB. That's multi-tasking!

Indeed. Our earliest WB diffuser (after the plastic one that came with the Sony MVC-FD91 started to turn yellow) was made of two layers of generic coffee filter sandwiched between two Canon brand UV filters (which were so wretched they weren't good for anything else)

But I liked best the part of the tests with morning fog as a subject!

Best regards,

Doug

 

[Drew Strickland]

Hi, Andrew,



Indeed. Our earliest WB diffuser (after the plastic one that came with the Sony MVC-FD91 started to turn yellow) was made of two layers of generic coffee filter sandwiched between two Canon brand UV filters (which were so wretched they weren't good for anything else)

But I liked best the part of the tests with morning fog as a subject!

Best regards,

Doug

I'll have you know that those filters were premium brand name filters.

Some people I know balked at the lack of consistency of the surface.

Boy, it's a good thing we're not still trying to sell only a neutral product (Michael). Especially when it seems to be so bad it can almost never turn in a decent result on skin.

But in all seriousness. That article is pretty much useless. Fog indeed. And ExpoDisc as the most neutral. Sorry. Not.

 

[Drew Strickland]

Why anyone would use a modern workflow, based on RGB working spaces (being Quasi-Device Independent), and profiled displays, then try to fix something using an output color space based on some printing press when ratio's (or rules) can be created using only three values (RGB) in a working space is beyond me. Its like the old "convert to Lab to sharpen" silliness of 1990's Photoshop techniques that needs to be buried from modern workflows.

Well, Andrew. With all due respect to your color knowledge, I think you'll want to take this up with SmugMug, Dan Margulis, Lee Varis and many others.

From the jacket of Dan's book:

Dan Margulis was one of the first three individuals- and the only writer- to be named as a member of the Photoshop Hall of Fame.

I'm pretty sure Elmo knows Lee. When he gets back from his walkabout maybe I'll see if we can get all of you together for a pow-wow. Might be good times.

 

[Ben Rubinstein]

Drew, in that last image you posted, having your target turned towards the light has resulted in a significant amount of glare. I know Mr Tapes specifically warns against doing that when using his WhiBal and from experience I agree with him. Does it bother you or are you getting accurate enough anyway not to have to watch out for it?

 

[Asher Kelman]

Yeah, could be. Except considering we have a pending patent on the whole "skin tone selector using wb" concept I doubt he would want to go there.

So far discussion is challenging, even contentious with some satire, I must admit. However this idea of using a colored card to shift skin color has been around since Kodachrome II! My late father in law used that 30 years ago! This is all common knowledge. We don't want to talk about "patents", since that is only degrees of freedom away from a Hedes, where one's efforts are sucked in like to a black hole. So lets not migrate rhetoric in that direction!

Look, Drew, your latest device might indeed be the most practical tool to help quickly obtain pleasing skin color in all required circumstances, despited vagaries of mixed light, each of different intensity, handling racial differences with aplomb. Personally, I'd like to try one and get unflummoxed by the science!

So far, the logic in terms of CMY explanations does not appear take into account changes in K and the entire gamut of colors of other objects. All that might be unimportant if it indeed somehow works for getting pleasing skin colors as claimed! The Kabbalah movement to which Madonna belongs, sells Kabbalah bottled water that also has claims of superiority. The molecular structure is explained to be held together in a more advantageous manner, apparently by a blessing's influence on quantum physics. That unexamined claim is either a lie or else a delusion which, in the end, doesn't detract from such water quenching thirst or sustaining life for beautiful orchids of every delicate color. I personally have observed these results After all, water is water!

Here, however, amongst us, I'd aver, we have enough technical knowledge, engineers and color specialists as well as experts in skin hue so that scientific claims are examined. This is bound to be educational! Just everyone keep this discussion in the bounds of courtesy and reasonable contention.

Asher

As an aside, I have admiration for the simple devices such as WhiBal™, the Expodisc™, the unheralded otehrs and even a cut out from a plastic milk carton over a lens to help achieving desired color. Drew's product may be even more useful. The world is huge and there's enough bread here for everyone's table. This is OPF, be nice!

 

[Drew Strickland]

Thus, for a given chromaticity, the ratios between C, M, and Y will not be fixed, but will vary with the luminance at which we find the color.

Best regards,

Doug

Yep! And there's a whole lot more once you really get into it. I know you guys didn't believe me.

Remember I said the tool was only loosely related to these concepts.

Again, back to (E)ase of Use and (S)peed. I initially rejected the notion of even attempting to make this sort of tool because the perfectionist in me said the tool would be way too complicated.

I told those crazy people who kept asking for simple and easy skin tone that at the bare minimum I would need about 5-7 different ratio "families" to even begin to cover the gamut of ethnicities. Then I also complained to them that for each "ratio family" I would need at least ten different "saturation/ luminance" levels.

By the time people got through clicking all 49 patches they would never use the tool again.

But once I started testing the thing with a reduced set of ratios, I found that really, brown isn't all that different. Pleasing is a recipe, not an absolute. A little more salt here, a little less pepper there can make a difference, but the biggest difference is just getting those darned skin tones to not be overly red.

I know you guys are only interested in hacking the perceived and real inconsistencies in the science and will continue to do so with gusto. I hope.

However, the only judge of this tool to me comes from statements such as the one you made earlier:

And certainly the result in the example is pleasing.

AWB is the real enemy here. Just as it is the main competitor for the core ColorRight tools (not WhiBal (sorry Michael I'm not really competing with you), ExpoDisc or any other competitor)).

Giving people a Simple and Easy tool to wean them off AWB is the goal. Then I'll leave it to you other experts to teach them they can possibly go even further!

 

[Andrew Rodney]

Well, Andrew. With all due respect to your color knowledge, I think you'll want to take this up with SmugMug, Dan Margulis, Lee Varis and many others.

From the jacket of Dan's book:

Dan Margulis was one of the first three individuals- and the only writer- to be named as a member of the Photoshop Hall of Fame.

You don't take things up with Dan, he's quite the religious zealot in terms of image processing beliefs. I've know him for years. We regally speak at the same NAPP venues. We disagree on all kinds of things (the use of wide gamut spaces, the need for high bit workflows etc). And neither he, nor SmugMug (or Lee who I also know) have defined why using a set of CMYK values which have NOTHING to do with the current digital file is useful or why using the same rules with an RGB color space isn't easier and more consistent.

Your values above can easily change by simply altering the CMYK profile in your Photoshop color settings. Your RGB working space values do not.

So why would anyone practicing modern image processing use a color space and a technique that date back to scanning chromes at a time in history when there were virtually no color displays? Why use an output color space that has absolutely no relationship to the current set of pixels in front of your very eyes?

As for the Photoshop Hall of Fame, there are far more members (myself included) that severely disagree with a large number of Dan's image processing belief systems.

 

[Drew Strickland]

Drew, in that last image you posted, having your target turned towards the light has resulted in a significant amount of glare. I know Mr Tapes specifically warns against doing that when using his WhiBal and from experience I agree with him. Does it bother you or are you getting accurate enough anyway not to have to watch out for it?

Hi Ben,

Yes. I agree with you and Michael. It is an issue.

The lighting in all of the examples is very poor. I took them this way to try and represent what you might wind up with at a typical wedding shoot running around in various poor light sources with little time to set exposure properly.

Hope this helps.

 

[Drew Strickland]

You don't take things up with Dan, he's quite the religious zealot in terms of image processing beliefs. I've know him for years. We regally speak at the same NAPP venues. We disagree on all kinds of things (the use of wide gamut spaces, the need for high bit workflows etc). And neither he, nor SmugMug (or Lee who I also know) have defined why using a set of CMYK values which have NOTHING to do with the current digital file is useful or why using the same rules with an RGB color space isn't easier and more consistent.

Your values above can easily change by simply altering the CMYK profile in your Photoshop color settings. Your RGB working space values do not.

So why would anyone practicing modern image processing use a color space and a technique that date back to scanning chromes at a time in history when there were virtually no color displays? Why use an output color space that has absolutely no relationship to the current set of pixels in front of your very eyes?

As for the Photoshop Hall of Fame, there are far more members (myself included) that severely disagree with a large number of Dan's image processing belief systems.

Was pretty sure you did. Yes, I know his beliefs are somewhat controversial. Just wanted to help set the context for those reading the thread who don't follow these things as much. I still think having all you guys together over here discussing your views might be very educational for all of those following this thread. I'm sure I could learn something.

But, again, as I have been saying all along. I'm not here to fight that battle.

At the end of the day, I could really care less if people believe his theories or not. His theories/ recipes produce a key nugget of understanding. Namely, that people don't like seeing too much red/ magenta in their skin. Even then, I don't care if people believe that. I'm not here to fight an ideological battle at all.

I'm here to show that this tool delivers what almost all viewers would describe as more "pleasing" skin tone. That's it. Yeah, I know, not too glamorous. But, it works.

 

[Asher Kelman]

Re: "little time to set exposure properly", wedding in various poor light conditions!

Drew, in that last image you posted, having your target turned towards the light has resulted in a significant amount of glare. I know Mr Tapes specifically warns against doing that when using his WhiBal and from experience I agree with him. Does it bother you or are you getting accurate enough anyway not to have to watch out for it?

Hi Ben,

Yes. I agree with you and Michael. It is an issue.

The lighting in all of the examples is very poor. I took them this way to try and represent what you might wind up with at a typical wedding shoot running around in various poor light sources with little time to set exposure properly.

Hope this helps.

Ben, Drew and Michael,

There's never enough time when one does not take time!

Talking of sloppy methods is unacceptable in serious photography. I just don't accept excuses! "Good enough isn't!", not for a pro and not for here! If one says one has a useful "gadget" that's maybe fine. That has, as yet, not been demonstrated or fully examined.

Anyway, many pros take photographs and send the film or digital files to their trusted lab. The lab will get it "perfect" with just the right pleasing look the photographer is famous for, anyway. Some photographers swear by jpg, especially Nikon and Fuji users. Most grey card standards will give a good in-camera reference. The final look is by eye!

For Raw, (thank you Andrew, Raw and not RAW! ), spend the effort to get the readings before, then or later. The world is not coming down! A professional either can correct by eye or else takes care of this, with a standard shot or a great lab but never excuses!

This is what I do at an event. I walk from area to area where there is different light and in each area I take several pictures with a WhiBal™ target to later batch process according to where I'll susequently take pictures. That simple!

If one likes a particular shift towards the red-yellow for example, one can have a standard set of curves to add to all one's PSD files. Try out which curve gives the most pleasing effect after WB and then "Bob's your Uncle!

Asher

 

[Andrew Rodney]

But, again, as I have been saying all along. I'm not here to fight that battle.
.

Its less to do about battles versus advising newer users to techniques that make sense that don't perpetrate old, sloppy, outdated and non useful techniques.

Take a look at your own CMYK examples here. Try this. Go into the Photoshop Color Settings, select Custom CMYK and pick the "default" which is SWOP with a medium GCR. Now open your RGB document with skin and put a sample point on the skin and have the readout set to CMYK.

In the example of an image I have (in ColorMatch RGB), the values I get are:

15/44/34/4

Now simply go back into the CMYK setup and change just the GCR to Max (nothing else). The new values are:

0/38/25/19!

The ratio's are shot. There's a difference in Cyan of 15, magenta of 6, yellow of 9 and black of 15. Which of the two sets of CMYK (both based on SWOP) are correct? I would submit, none are.

However, the RGB values haven't changed a lick. Why deal with a ratio of colors that uses four colors based on a ink and press colorant when you could simply build a set of ratio rules using RGB? It wouldn't matter one bit how a user sets up his/her color settings. The color values are the color values? And this ignores the idea of simply LOOKING on a calibrated display what the skin tones look like (or building a small visual library of known, well output skin tones as a reference when correcting an image).

This CMYK technique is simply an old, brain dead way of working numerically, based on Photoshop prior to color management, calibrated and profiled displays and well behaved RGB working spaces. It needs to go away (like Adobe Gamma).

 

[Drew Strickland]

Gotta go stand in line to vote and then work on some server installations.

Will try to check back in later.

Hope everyone has a great day.

 

[Doug Kerr]

Hi, Drew,

. . . a key nugget of understanding. Namely, that people don't like seeing too much red/ magenta in their skin.

Yes, and that has been a very useful revelation to me.

Even more fascinating, and valuable on many fronts, is to realize that almost all we H. sapiens have brown skin (even my favorite Cherokee).

And I'm grateful to have the benefit of a "recipe", imperfect though it is, to guide my amateurish efforts at skin tone tweaking.

The part I'm still working on is, given the basic fact you mention above, why does should putting in more red make a "more pleasing" result?

Perhaps it is that having M less than Y, even a little bit is the most potent requirement (more so than don't have too much red), so the benefit of adding Y overcomes the seeming disadvantage of adding R at the same time. (The chromaticity shift vectors we might expect from the use of the non-neutral patches of the ColorRight MAX are mostly somewhat more aligned with the Y direction than the R direction.)

I appreciate that it is of course more complex and subtle than this. Still, I like to have the salient pieces fit somewhat together, or at least not collide.

Best regards,

Doug

 

[Michael Tapes]

I'm here to show that this tool delivers what almost all viewers would describe as more "pleasing" skin tone. That's it.

Ah..confirms that you are here to market and promote your product for free. Just checking.

<g>

Have a good day.

MT

 

[Michael Tapes]

Ben, Drew and Michael,

This is what I do at an event. I walk form area to area where there is different light and in each area I take several pictures with a WhiBal™ target to later batch process according to where I'll susequently take pictures. That simple!

Asher

That is what I do as well. It *is* that simple..

MT

 

[Michael Tapes]

Its less to do about battles versus advising newer users to techniques that make sense that don't perpetrate old, sloppy, outdated and non useful techniques.

Andrew,

Thanks for your sanity..

Talk about beating a dead horse!

MT

 

[Doug Kerr]

Hi, Michael,

That is what I do as well. It *is* that simple..

I think both you guys are thinking that white balance color correction is for white balance color correction.

Best regards,

Doug

 

[Drew Strickland]

This CMYK technique is simply an old, brain dead way of working numerically, based on Photoshop prior to color management, calibrated and profiled displays and well behaved RGB working spaces. It needs to go away (like Adobe Gamma).

Hi Andrew,

I get that you don't like the CMY method. I knew that before I ever posted over here.

I posted your arguments against this method to one of these folks who was pushing for the tool a long time ago.

{snipped}
"... Import images with known, good skin tone appearance (ideally something you've output successfully using proper color management). View the RGB percentages. Start jotting down the arguably variable values (skin tone is all over the map). If you do this long enough, you'll find useful ratio's that are no different in theory than using CMYK or any other set of numbers (scales). Or just look at the images as you (and I) suggest and render to taste. But there's nothing super impressive here with the old CMYK trick. Someone spent time using ONE (unfortunately undefined) CMYK output color space and said "use this and that ratio of CMY and K values and you'll get good skin tones" (hopefully adding "Your mileage may vary").
{Snipped}

LOL!!! Yes, this is vintage Digital Dog. It's also what I said to Peter: do this long enough in (insert your favourite RAW converter) and you will get a sense for what's going on anyway.

As for Andrew's thoughts on color, well, I see DNGs might now have input profiles in the DNG too in the next version of Lightroom This was also something else Andrew thought was pointless; it'll be interesting to see what he thinks of the next version of Lightroom (but I already know )

And I mentioned before you could can colour correct in RGB "by the numbers", but the numbers / ratios are different in every single RGB space. LOL! Not exactly easy, unless you, say, stick to sRGB and be done with it.

If I understand correctly, this is not true for CMY, because CMY is RGB independent. Except for black levels and ink limits (which have absolutely nothing to do with CMY values), there is only ONE CMY space in Photoshop, not many (IOW, because of the way presses work, you can have many CMYK output profiles, and Dan's book will tell you why they're a good thing to know about too; that doesn't mean the CMY values themselves change).

I suppose you could change the gamma of a CMY profile as well, but that wouldn't change the CMY values enough to affect the ratios.

So adjusting the ratios in CMY is very nearly the same as adjusting in LAB; the values just don't change that much compared with an RGB space.

Sorry, I think Andrew's quite wrong on this. He may make good profiles, but he's not an expert on colour correction the way Margulis is.

(and of course this is an old trick! What part of "they used to do this with color film prints" makes it an invalid process again?!)

Again, the point is pleasing. Is it more pleasing to most people? Yes. Argument done. At least as far as I'm concerned.

I stood up for you man.

This is what people close to Adobe (Andrew Rodney) say over at the Lightroom 2 forum:

(skin tone is all over the map).

But there's nothing super impressive here with the old CMYK trick. Someone spent time using ONE (unfortunately undefined) CMYK output color space and said "use this and that ratio of CMY and K values and you'll get good skin tones" (hopefully adding "Your mileage may vary").
Vintage Andrew Rodney. He is none too complimentary of this skin tone shortcut method.

Good old Andrew! For once we agree and he can't even be here to see it. Isn't that how it always goes.

Ok, maybe the second or third time we've agreed on something.

FWIW Jamie, I can see some value in the method. If it works for you. It works.

This tool is for working pros who want to please their clients in a quick and convenient way. It does that.

Gotta go for real now.

 

[Drew Strickland]

Talk about beating a dead horse!

MT

Yes, the horse is dead.

To summarize:

1) Neutral does not work well with skin tones quite often.

2) The ColorRight MAX corrects this problem quickly and easily.


It was dead a long time ago. All you had to do was look with your eyes at the results.

But, you guys want science. So, I give you science. You don't like what you hear. So, you disagree.

Dead, I tell you, dead.

We believe what we want to believe. Again, on that note, gotta go vote what I believe.

Hope everyone else (in the US) is ahead of me and has already done their duty.

 

[Andrew Rodney]

I get that you don't like the CMY method. I knew that before I ever posted over here.

Then please stop using it! Unless you've got a rebuttal that suggests its useful based on what I've said about it.

I don't like lots of things that have cropped into the imaging language despite my attempts to get folks to stop using them:

DPI for image resolution (images have pixels, output devices make dots).
RAW instead of Raw (its NOT an acronym).
MegaPixel for "describing" what a camera supposedly produces (not all the so called megapixel sensor spec's produce image data).
False Profile and "Ultra Wide" gamut spaces (Dan terms), stuff he's made up that only he uses and makes no sense (there's no such thing as a false profile).
Invisible colors (if you can't see it, its not a color).

We can add all kinds of useless Photoshop techniques or techniques that should be conducted at Raw rendering and not at all in Photoshop.

This CMYK skin tone "trick" needs to go away, you can start by not referencing it.

As the Chinese proverb says: The first step towards genius is calling things by their proper name. And not perpetuating outdated and useless techniques!

 

[Michael Tapes]

Yes, the horse is dead.

To summarize:

1) Neutral does not work well with skin tones quite often.

2) The ColorRight MAX corrects this problem quickly and easily.

I guess it took just one more summary (read ad) to really get that horse to lay down <g>

Gee... what if someone had their *own* forum so they could advertise for free there. Hmmmmmnmnnnn.

Makes one wonder.....

Good thing it is a free country and a free forum! let's all rock the vote.

MT

 

[Asher Kelman]

Getting terms right in Color and Digital Photography!

Then please stop using it! Unless you've got a rebuttal that suggests its useful based on what I've said about it.

I don't like lots of things that have cropped into the imaging language despite my attempts to get folks to stop using them............

Invisible colors (if you can't see it, its not a color).............

HI Andrew,

The only thing that I would caution is that some colors are invisible under certain common circumstances. The color may be remapped to make it look something close to the real color the file will yield, but there's no way of seeing colors that the monitor cannot display. Thus, for example, some real colors one can see on an Epson print are invisible to the eye looking at most monitors. The monitor is essentially an attempt at an accurate window to what is to be accurately printed. At that time we are looking through this window, we fail to see the true color.

So I hope, just in that important context, you'll allow the use of an "invisible color", unless you insist on "a printer achieved color that is not visible on the monitor"!

Asher

 

[Andrew Rodney]

HI Andrew,

The only thing that I would caution is that some colors are invisible under certain common circumstances.

In the truest sense its not a color if you can't see it.

Color, is a perceptual property. So if you can't see it it's not a
color. Color is not a particular wavelength of light. It is a
cognitive perception that is the end result of the excitation of
photoreceptors followed by retinal processing and ending in the
visual cortex. We define colors based on perceptual experiments.

A coordinate in a "colorspace" outside the spectrum locus is not a
color. We often refer to these as "imaginary colors" but this is by
and large also erroneous (you can't map an imaginary color from one
colorspace to another as the math (and experimental data) for each
colorspace breaks down outside the spectrum locus.

 

[Doug Kerr]

On the soapbox

Most photography can be thought of as technique (and technology) in pursuit of art.

And for the most part, a technique being considered for repeated use is evaluated according to these two criteria:

- Does it allow or assist us to produce a result that is "desirable" to the relevant constituency (which might include only ourself, or perhaps a paying client, or a publisher, or the jury of a competition, or the lookers at a gallery)? [Note that "desirable" here may have a definition with both objective and subjective components.]

-Is it practical to utilize?

Of course, any technique actually works through the exercise of various physical principles, which "we" may or may not understand. Fortunately, understanding the principles is often not necessary for the successful employment the technique. But in some cases, it can be helpful, and may be especially useful to refining the technique, or making it "repeatable".

The outlook given above in blue does not seem to require defense, or validation, but sometimes it gets it. I think often this is the result of fear that art and science, pragmatism and idealism, faith and fact, are somehow incompatible, and thus we must pick one of each, and in its honor, denounce the other.

In any case, I find it offensive whenever someone seems to find it necessary to "validate" this outlook by ridiculing scientific inquiry. I, for one, will never suffer that gladly. Restez en garde.

Best regards,

Doug