LWN: Comments on "Web Boosts the Cause of Free Color (Internet Evolution)"

Or just use sRGB

PO8 — Mon, 14 Dec 2009 21:15:58 +0000

"I don't know very much about colour printing." Don't mean to be rude, but yes, you're right. I'm afraid that sRGB isn't at all suitable as a one-size-fits-all color standard, especially for printing. The reasons are kind of complex. Let's just say that people doing professional color reproduction aren't idiots; their willingness to saddle themselves with a solution like Pantone speaks volumes to the difficulty of the problem.

My favorite site for educating oneself about color is handprint.com, which has a beautifully-done comprehensive guide to the subject.

I would love to see an open Pantone replacement. My big concern is that the "Open Color Standard" appears to be one badly-maintained wiki page with no obviously viable plan. The suggestion to get Toyo Ink Group involved would be much more exciting if the link to Color Finder on the wiki page actually worked, or was even spelled correctly. Sadly, I think this project is non-news until someone cares enough to really take it on and move forward with it.

Or just use sRGB

njs — Mon, 14 Dec 2009 20:15:10 +0000

> I don't know very much about colour printing, but I rather get the impression that the people advocating this "Open Color Standard project" don't know much more than I do.

Not sure where you're getting that -- I don't know how expert they are on digital color (though I'd assume quite a bit more than the average bear), but they're clearly experienced with the realities of prepress. They're talking about making physical swatch books and finding color-matched inks usable for spot color. That's what Pantone does that makes it more than just a weird discretized version of L*a*b*.

Or just use sRGB

edmundo — Mon, 14 Dec 2009 18:02:40 +0000

I don't know very much about colour printing, but I rather get the impression that the people advocating this "Open Color Standard project" don't know much more than I do.

Colour is a complex thing. You can easily spend a day or two just getting to grips with what Wikipedia has to say on the subject.

Fortunately, there is now an established standard for how 8-bit RGB values should be interpreted: sRGB. Of course sRGB isn't really suitable for printing: firstly it only covers a subset of visible colours, and secondly it's designed for specifying the colour of light rather than the colour of a surface. However, most people most of the time don't really care. Increasingly they are dealing with files that just JPEG files with no additional colour information. Also, if you're specifying the colour of a logo, or anything else that will be displayed on the web, then you're going to have to reduce the colours to sRGB eventually, in any case, so you might as well start off with sRGB and tell your printers to match #D81E05, say, rather than tell your web designer to match Pantone 485. So I think Pantone will become less and less important and sRGB will become the standard for most things.

There are plenty of (unofficial) Pantone-sRGB translation tables on the web. A properly defined translation from Pantone to sRGB would take account of the lighting conditions (see http://en.wikipedia.org/wiki/Standard_illuminant). However, in practice you usually don't know what conditions printed material will be viewed in.