Exposure is not how dark or how light an image looks!

For the same processing, I expose differently for optical print and for scanning. The factor here is what I'm going to do with the film.

Probably best to specify then, since film photography remains popular (and at the rate digital camera sales are going, might become the dominant form of photography again).

I can't decide, and having always thought DPR's stuck record rule was a good one, I'm not inclined to argue. There are others who will do that anyway.

I do agree that some photographers (perhaps many photographers) use over- and under-exposure in a way that is outdated (and wasn't necessarily true in the film days). I don't think it's useful to jump all over people (especially in a beginners forum) when they use those terms in place of lightness. Many photographers shoot jpg, and do indeed use jpg as their primary lightness control. This has its compromises on image quality but on a sunny day it makes little difference in apparent picture quality (and by apparent I mean what you see without getting down to the pixel level).

Educate them, yes. Pounce on them DPR-style with WHAT THE HELL DO YOU THINK EXPOSURE MEANS???, no.

As I see it, exposure in digital is a more nebulous concept, and optimal exposure is almost impossible to define, at least without resorting to a large number of variables that make it nearly meaningless anyway. With film, the answer to "What is the optimum exposure for this scene?" has a very narrow range of answers. Under the same circumstances with digital, the true answer is "It depends".

I do think the concept of "nailing the exposure" doesn't have the importance in digital photography than it did in film photography. That is the beauty of digital, you can have a way-less-than-optimal exposure and still get a very usable image. The gap with film is much narrower. Part of the reason I prefer it.

What if you're going to do both?

I am assuming the D in DPRevived stands for Digital so there is no need for me to specify unless i specifically refer to film.

If what you predict eventuates then perhaps this site's domain name could be changed to FPRevived.com :-)

For me the definition of exposure is very clear and simple. Exposure is the amount of light striking the sensor per unit area while the shutter is open.

For me that is simple and clear as well but everyone will have a different definition for optimal exposure according to their needs and goals.
For me, optimal exposure is the maximum exposure (as defined above) within dof and motion blur requirements without clipping important highlights.

Isn’t it reasonable to say that an image that is too light is the result of the ISO being set too high? This can be the result of the exposure being perfect but the ISO being too high.

If you say that an overly light image is always due to over-exposure, you may mislead a beginner into thinking that the solution is to lower exposure. Often the solution to an overly light image is to leave the exposure the same, and to lower the ISO.

Sure it's reasonable, if that's what caused an image to be too light. There are a multiple ways a photographer can create an image that is too light and multiple opportunities, along the imaging chain's path, to make errors.

Then there's no reason for concern, as I didn't say that and therefore beginners won't be misled by what I said.

I suggest beginners pick up a good book which explains photographic fundamentals, The Camera by Ansel Adams has excellent explanations of how camera and lens adjustments function and interact in your creation of images. There are many books of this type, The Camera is a classic example. Any reliable source, online or off, can provide a foundation of understanding upon which to build. For an understanding of exposure I linked this useful article some pages back in this thread.

That's possible. It will be up to the photographer to apply their knowledge and experience to those decisions.

Except for dual conversion gain, read noise difference in 2-3 stops of ISO is mostly unnoticeable. That is why I prefer Jim's approach to exposure:

Fuji GFX 100S exposure strategy, M and A modes

The main problem of being a couple of stops below the metered ISO is that the EVFs become too dark to frame. With Auto ISO in M mode, and if I am above the dual conversion gain bump, I often set 1 to 2 stops negative EC.

That’s fine. Just do what works best for you to set the maximum exposure* within dof and blur constraints without clipping important highlights.

* exposure - amount of light striking the sensor per unit area while the shutter is open.

Yes, this is a beginner's forum. Therefore it is important not to point beginners in the wrong direction when they have a problem.

Suppose a beginner takes a shot at f/8, 1/60, ISO 1600, and the camera produced JPEG comes out looking much too light. If you tell the beginner that this is the result of "Over Exposure", you are strongly implying that the cause of their problem is that the exposure is too high. The obvious solution to an overly high exposure is to reduce the lighting on the subject, use a faster shutter, or use a smaller f/stop. Reducing the exposure will make the JPEG darker, but has the side effect of increasing image noise.

On the other hand if you tell them the ISO was too high for the exposure, you have made it clear that they also have the option of reducing the ISO setting.

Addressing the issue by reducing the ISO is probably the better solution.

Now, if we are in an experts forum, then you likely won't mislead readers by saying "over exposure" when you mean the image it too light. However, in an experts forum it is more likely that nuances are being discussed, and that's much easier if you use the correct terms.

Not my case.
I don't think it helps to discuss film here.

I think the point here is to get beginners on the same page from the get-go, with everyone speaking the same language and not continually misconstruing everything being discussed.
It’s absolutely worthwhile (IMO) making an effort to avoid perpetuating and further ingraining all the misinformation and misconceptions already floating about out there with everyone using the same terminology, but not saying the same thing.

Image shadow noise and dynamic range are two sides of the same coin.

There is no "for me", a definition is what it is. Yours seems pretty close, it should be "striking the light-sensitive media". And of course you left out "per unit area". I don't think I need to lecture you about total light (though I believe there are others here, ahem, who have jumped all over others for that same omission).

And that makes my point above about too many variables for a good definition. I could have, say, a photo of a moving racing car in which I have saturated the sensor with light without clipping highlights. An optimal exposure, some might say. But the shutter speed froze the motion of the wheels, and I wanted them blurred. Missing my motion blur requirements, as you stated above. So I could say the photo is underexposed. Someone else might like the frozen effect and say the photo is perfectly exposed. It's Schroedinger's photo, in that the photo is both optimally and non-optimally exposed. And I could say it's either depending on my desire to impress the photo's viewer.

I don't think we're making different points here, but that is my contention that "optimal exposure" contains too many variables to be pinned down to a solid definition. I'd go more vague... it's the right set of exposure variables for the desired effect.

Well, so much for making these new forums useful to everyone.

Yes, staying on topic, not bringing up what's irrelevant, and not spreading misconceptions IMO makes forums useful for everybody but trolls :)

Is it possible that only one channel is underexposed or overexposed?

I was trying to think of a word describing the relationship between dark and light in a photo. Not exposure. Perhaps "Dynamics," related to the range of light and how it is expressed.

It depends on whether you are discussing at the image as a whole, or only parts of the image.

For instance, if you have a backlit model at the beach, the pixels over the model may be exposed exactly a desired, yet the background pixels may be blown out.

One can also look at only the pixels behind green Bayer Pattern pixels. One could even limit the discussion to only pixels on the left or right side of the image.

So yes, depending on the context of the discussion, one might talk about the image as a whole, or separate out the discussion by classes of pixel.

Dividing the pixels by the corresponding Bayer Pattern color does make sense. As these are usually processed with different parameters (for instance the White Balance setting will affect the relative processing of the different color classes of pixels).