A few things I found while I’ve been debugging some of my own profile creation tools:
2D/3D plot list + menu has an entry for “Active”. When it is enabled, it shows 2 or more blank entries. Selecting any of them gets an error “Error: additem not setup with Add menu correctly”.
Clicking the 2D and 3D graph buttons in the profile inspector frequently does nothing, or simply opens an empty (axes only) grapher window. Sometimes I get lucky and it actually graphs the selected profile, but usually I have to add it to the list manually.
Profile Inspector shows tag sizes in Kilobytes, which isn’t very useful when inspecting an 11 byte tag (that should have been 12 bytes) and it all shows 0.01KB. I would prefer to see an exact byte count instead of a bad approximation of a user friendly value.
ColorThink does not graph 2D and 3D plots for devicelink or abstract profiles. Abstract could show a cube mapping (see Apple’s ColorSync Utility) or a gamut hull, device link probably should should a cube mapping.
Abstract and DeviceLink profiles show the 2D and 3D graph buttons in the Profile Inspector without graying them out, but they do nothing.
Profile Inspector has the Curves tab enabled for devicelink and abstract profiles, but shows nothing for the curves in those profiles. Abstract should have obvious content. Devicelink should switch from L* to device coordinate curves.
Profile Inspector Statistics tab shows nothing useful for devicelink or abstract profiles. Gamut Volume for both shows -NA,N(0,00). Abstract profiles have an obvious gamut volume. Devicelinks are less obvious, but should switch to device dependent measurements.
On MacOS, all file paths are still given in Carbon “:” separated form. We moved to OS X a few years ago now, and it’s time to adopt Unix style file paths.
Similarly, MacOS identifies profiles by extension not just by filetype “prof”, so having that as a warning is kind of pointless now. Classic is now long gone, and no longer a concern for most users (ok, except those still running G5’s).
And requiring that filenames and profile descriptions match is kind of old-school as well. I want the filename to be unique, but I don’t necessarily want the description to be so terse or filesystem centric – that’s why the ICC specified the description separately from the filename (well, that and 8.effing3).