My question is the same as the OP's:
That differentiation can only be achieved if each term is defined separately and then the difference pointed out. Plain English helps the rest of us understand experts.
Geez, like pulling teeth to get
you to
ask a simple question.... As for
the rest of us, you mean (thus far) you! Best speak for yourself.
Calibration is a process whereby a device is placed into some predetermined condition or behavior. An example is when a user calibrates a display. We want to set some parameters that can be controlled, such as the display’s white point, brightness contrast, and TRC gamma. This calibration of the device creates a condition that can be standardized and is repeatable. This allows similar devices in multiple locations to behave the same way. Since a device such as a display is in a state of flux over time, it is necessary to calibrate the device. This calibration returns the device to the original aim point. Calibration is something we need to do on a regular basis. Particularly with such devices as a display, which varies considerably over time. In conventional photographic processes, this would be similar to running control strips and adjusting chemistry to achieve target densities. If we understand that a profile describes the behavior of a device, we should be aware that if the device changes its behavior, the profile is no longer valid. Calibration returns the device back to the original condition, maintaining the integrity of the profile. If the device can no longer reach the original aim point, a new aim point within the capability of the device needs to be created.
Linearization
Some profile-building products offer an optional step they usually call prelinearization. The idea is to output a target with a small subset of patches, usually CMY and K in various steps from light to very dark as seen in Fig. 6-9. This linearization target is measured and the software uses that information to produce an optimal target for profiling based upon the information gathered from the linearization step. This means that the profiling process becomes a two-step procedure. Some devices are quite nonlinear in how they reproduce color. The linearization step can aid in producing quality profiles from such devices. At the very least, linearization allows a good profile to be generated with an initially smaller number of patches. This is possible because the secondary patches generated from the linearization data is better optimized for the printer.
Some products support this prelinearization process and some do not. Products that do support prelinearization usually ask the user if they wish to use this option. If you know the printing is very nonlinear, it’s worth testing. On the accompanying CD is a TIFF file called InkDensityTest.tif, which can be useful for visually evaluating if the output device is nonlinear. If most of the steps block up in color and don’t show good tonal separa- tion, the printer is exhibiting this nonlinear behavior. You can try a different driver setting if available. However, this can often alter the resulting color gamut of the printer. Alternatively, you can try conduct- ing a prelinearization process, but even this can’t produce miracles. A profile can do only so much to overcome the limitations of poor printer behavior (see the sidebar, “Printer Drivers and Their Effect on Quality and Color Gamut”).
Some packages also offer a postlinearization process. This can be quite useful for devices that change their behavior. After an initial profile is built, the user prints a special postlinearization target. This target is measured and the profile can be updated to account for some changes in the output device. This postlinearization process can’t work miracles on output devices whose behavior has greatly shifted. In such a case, building a new profile from a standard target is the only viable solution. However, for some devices where the drift of a printer isn’t large, this postlinearization process is a useful feature. Some users running commercial labs print out a postlinearization target every day and update their existing profile to account for slight device drift.
And DO try to read the materials provided already!
http://digitaldog.net/files/LinearityandGamma.pdfParagraph 2 for example.