1. Match high expectations with high levels of support.
Duh. A bunch of empty business words/bromides.
2. Include evidence of teaching and student learning from multiple sources.
Again, Duh. At least they are admitting that test scores don't help much. (And of course, they are relying on "gains" being the sole result of the teacher. The largest influences (parents and the students themselves) are still left out of the equation.)
3. Use information to provide constructive feedback to teachers, as befits a profession, not to shame them.
I support this.
4. Create confidence in the quality of teacher development and evaluation systems and the schools ability to implement them reliably.
Ugh. No system will ever work well because you're trying to measure something that is next to impossible to measure. And it's far too variable to measure with any repeatability. There are two paths: a generic system that isn't particularly useful or a very specific system that also isn't useful. If you want a "system" that works, you need to recognize that someone teaching 2nd grade reading is going to have to do things very differently than a high school chemistry teacher. But since they want a standard evaluation, you need to go with the generic system. Instead, we get ridiculous garbage like the Marzano system.
5. Align teacher development and evaluation to the Common Core State Standards.
Why? How is this going to improve education? Is there ANY evidence it will significantly change things (other than making it easier to sell standardized tests) for the better?
6. Adjust the system over time based on new evidence, innovations, and feedback.
I agree, because we'll need to throw away all these current useless fads (see #4 and #5 above, for examples).