On Thursday 22 November 2007 20:18:14 Sandman wrote: > Hi, I would like to ask for the development team's consideration for > implementing the OpenPalette format for their palettes. > > OpenPalette is really hoping for a medium with interoperability and > flexibility. The file format is XML-based; there's a DTD and XSD already > written and ready for business. There's a lot more information at the site > (http://openpalette.uk.to), if you are willing to look into it. You can > also download a PDF concerning OpenPalette here. > > Thanks; if you have any questions, I'm ready to answer. My first concern is that it does not appear to have a way to specify the actual color. That is it allows the "color" to be specified by a hex value that appears to be limited to FFFFFF and/or an RGB value which appears to be limited to 255,255,255 which is really equal to FFFFFF. But what do these values mean? On the Usage page there is an example of a color named "red" that has an RGB value of 255,0,0. Depending on the color space this could be any number of actual real world colors. For example, this red would be significantly different if the color space is sRGB as compared to ProPhotoRGB or my monitors native color space or what would be printed on my printer. So which red are we talking about? There needs to be a way to specify colors in a way that is not ambiguous and this format as far as I was able to determine does not do that. Since I was not able to find the PDF document perhaps I am missing something? Open source software that uses this typs of data is now, or is in the process of becomming, color management aware and should be able to deal with color being specified using an unambigous format such as specifying the CIE Lab or CIE XYZ value of the desired color. Another way to do this would be to specify both a colorspace such as sRGB, BetaRGB or ProPhotoRGB along with the RGB or CMY(K) values of the color. Also what about other color space types such other than RGB such as CMY(K)? And what about other representations such as floating point or 16 bit or 32 bit channel values? Hal _______________________________________________ Gimp-developer mailing list Gimp-developer@xxxxxxxxxxxxxxxxxxxxxx https://lists.XCF.Berkeley.EDU/mailman/listinfo/gimp-developer
