My basic grading philosophy and policy is:
1) In a teaching career that began in 1969, and a learning career that began in 1948, I have learned that a teaching and assessment system that is based on ABCDF grading, and in turn on measurements of seat-time and distributions along a "bell curve", is inferior to a system that brings along as many learners as possible, helping them to reach a standard of sufficiency before then challenging them to gain new capabilities and reach new standards.
2) Since we still have to use ABCDF grading, I want the "A" grade to mean what it should: not just that a student "does" what is assigned, or even does it "very well", but that the student distinctly stands out as a learner who can do more than just follow simple directions - someone who can learn and think independently and, eventually, lead others.

That is what is meant by a "6" on the system that is being used more and more in education: exemplary. In terms of ABCDF, we can use this device to help us:

A = a performance that is AmAzing. What you do has to go beyond what is assigned and expected of the ordinary student. You have to come up with something more, and you have to carry it out: not just write a longer paper, not just spend more time on it, but truly extend the boundaries set before you. Yes, there can be some small flaws in your grand design, but the grand design has to be there.

Don't ask me to tell you specifically what will AmAze me. Or rather, if you do ask me, I may give you an example of something that would AmAze me, but then of course you can't just do that, because I wouldn't be AmAzed by it.

As you try to AmAze me, you might think about what a historian of science has proposed as a major factor in advance of science. (Caution: coarse language in next sentence). She proposed that science advances often because somebody sits up late at night thinking, "I'm gonna show that SOB."

So show me. While you're figuring out how to do that you are allowed to call me anything you wish, at least in your thoughts.

I wrote this sitting up late one night, to show you something, oh BS people (=Beloved Students).

:-)

Please note the fundamental differences between the two systems:
1) Scoring guides are always based on standards that define performance in functional terms; letter grades are often based on point scores, with points that may not refer directly to functional competence.
2) Tests and other activities that yield letter grades are often used to generate a "bell curve" that is in turn adjusted up or down to yield a desired distribution of grades. Scoring guides measure how many examinees meet or do not meet a standard, with the score of 4 defining the border. In my courses "sufficient" is a B-; C+ students are considered ready to go on for more learning too, but they are not always comfortable doing so.
3) Scoring guides imply that student learning for a given standard is assessed more than once, if necessary, with revision and rescoring as basic parts of the system. That is why my courses invite you to revise your work, sometimes even your tests, for rescoring and appropriate change of grade.
4) Scoring guides imply that learners take part in their own evaluation and also take more responsibility for their learning. Letter grades may create an atmosphere of rote learning, mysterious tests and grading, and passive learning. That is why may scoring guides for my courses are publicly posted; you are encouraged to use them to guide your learning, even before your performance is assessed.
5) Scoring guides imply that learner placement is important. You are encouraged to ask for placement advice. Scoring guides also imply that learners who can move ahead should do so. Unfortunately, we don't yet have a system in place that allows for fractional moves, such as from 101 to 102 within the same quarter. We're working on it!

Scoring guides are being used more and more as education is modernized to serve a society that needs large numbers of highly-trained citizens. In such a society, learners who do not meet standards are given more attention; those who readily meet standards are encouraged to move on to new learning. The A-F "bell curve" system implies a society where all learners receive the same amount of instruction ("seat time") under the same circumstances. Then a test is given, and elite (or perhaps just privileged) learners are identified and selected for further education and, eventually, privilege and power. The others (the larger group) are confined to "lower" levels of the area, redirected to other subjects, or culled from the educational system. In the agricultural or even industrial past there were more such jobs, and some even paid fairly well.

An economy based on services and information needs a better educational system than one of seat time and the bell curve. And instruction and learning that are defined by standards tend to provide more rewarding educational experiences.