Blender has incorporated some of Filmic Blender but...
Information on this topic can be found here:
Are there significant differences between filmic shipped with blender and its development version?
Guru says "Do this simple thing and all your stuff will look better forever"
Instead of relying on dogma, edicts and magic formulas from gurus or priests, it's laudable that you want to take the time to understand how the tools work, and how you can use them for your own purposes. That way you'll get the results you are after, and don't have to settle for whatever you get using unknown parameters.
Read:
Render with a wider dynamic range in cycles to produce photorealistic looking images
So I'm trying to figure out how to make due with standard blender at least for now.
A quick and painless solution is to render your images as EXRs. When you render as EXRs, the final result does not have any color transforms applied to it. That way you can keep all of your scene referred information as linear and undistorted, and then you can import the images into blender (or other apps) and apply the display transforms that you want (including the filmic transforms) before saving them to display referred images (tif, jpg, png or video endoding).
Why we can't just simulate photons around, record their wavelengths, and call it a day
Because image making requires interpretation. There is no universal or "correct" way of determining what should be black or what should be white, or how to interpret all of the tones in between. The same scene can look totally different if you choose to prioritize different brightness levels, or if the scene is visualized through a filter. It all depends on the intentions of the person making the image.
A render engine determines the values for each pixel based on the light reaching the object and the qualities of the materials that make the object. But how those values are represented in the final image, depends on the view transforms through which the image is processed.
In other words: The render engine generates scene referred data, and the values can be in a scale that is linear and infinite, as in real life. In order to represent those values in an image they have to be converted to the limited values that a monitor or projector or paper can display. The values are usually modified in a way that is no longer linear and looks "correct" or "pleasing" (as our brain does not perceive brightess linearly). The data is converted thus to display referred. The tools to convert from scene referred to display referred are the display transforms, which are a series of rules based on LUTs (Look Up Tables) and other algorithms to convert values.
how do the Display Device, Render View, Look, and Color Space even interact?
The way to combine or compound different transforms is by using the Display, View and Looks options in the color management settings.
Display Device will be sRGB If you are using a regular monitor. No need to use any other option (and if you are using a monitor with a wider color space, like a DCIP3 retina display or a HDR monitor you are in for a new set of problems...) The rule of thumb is not to use any setting that you do not know how to use... Do you have a compelling reason to display XYZ? If you don't, then don't use it.
Render View Select Filmic or sRGB to render finished images, Non-Colour data for normal maps or bump maps.
Looks Are meant as additional transforms to control contrast within the log encoded images created by filmic blender.
What are the dozens of film types listed in look for?
They come from configurations for other software that somehow made its way to blender. They are largely useless and can safely be ignored. Read: What is the actual film manufacturer reference for the film response data?
How to learn more? There are plenty of posts on this site. And if you get stuck ask for help in a concise way.
