President Obama, who I support, has encouraged even more primitive, bubble-in educational malpractice than President Bush did. But, in his State of the Union Address, even the president said that we should "teach with creativity and passion," and "stop teaching to the test."

I drafted a post asking why President Obama could believe that he can continue to impose test-driven accountability and get something different than non-stop test prep and curriculum narrowing. Why would the new Obama words lead to anything other than the damage that is being done by Secretary of Education Arne Duncan's version of NCLB-type accountability on steroids?

The next day, Secretary Duncan participated in the "data summit," sponsored by the Data Quality Campaign which is funded, in part, by the Gates Foundation. I speculated that this preach-to-the-choir event might explain why the Obama Administration presents such an illogical educational policy. Bill Gates likes standardized testing, so another generation of poor students of color must be sacrificed to teacher-proof, rote instruction.

Even the author of Gates-funded report under discussion, "Data for Action 2011," acknowledged, "all this data won't amount to anything if we don't change the culture. Aimee Guidera said, "We have to transform the way we think about data from a hammer that's going to hurt teachers to a flashlight that's going to help."

The report provided some zingers for my draft post. The post ridiculed the Obama administration for giving billions of dollars of gold-plated hammers to districts, along with a request that they stop using those tools to beat down teachers. It asked for a scenario where true-believers in standardized testing would stop treating teachers like nails to be pounded into compliance.

Then, I read the latest post by one of my favorite education writers, Dana Goldstein. Goldstein wrote, "One thing I admire about Bill Gates is that he isn't afraid to update his thinking when presented with new information." Gates writes an annual letter on his philanthropic activities, and this year's letter, "marks a significant shift in Gates' thinking, and shows he has learned a lot from dialogue with classroom teachers. The letter argues strongly in favor of teacher peer review, a more holistic, classroom-observation based evaluation strategy that is popular with teachers and their unions."

And sure enough, Bill Gates has just explained why his preferred policies, as well as those of the Obama administration, could be the beginning of the end of the educational malpractice encouraged by data-driven accountability. Gates endorsed the Tampa peer review system where the administration and unions work together to to train 2% of their teachers to work with principals to evaluate and improve instruction. He wrote that "the first round of evaluation revealed that many teachers need help engaging the students to prompt critical thinking and problem solving. The district started to organize its professional development around these findings."

Gates concluded:

After seeing how valuable peer evaluation is, I think it should be part of every public school personnel system. Dedicating 2 percent of teachers to do this work is a large investment. It can mean raising the average class size by 2 percent or spending 2 percent more money. ... Without this investment I don't think an evaluation system will get enough credibility with the teachers.

I am assuming that Gates would also embrace "the Grand Bargain" where results from value-added models are interpreted by teams of peer reviewers, not by management alone. The last people who should be allowed to interpret the results of these experimental models should be administrators. They have an inherent conflict of interest in deciding whether a teacher's failure to meet his growth target is due to the educators' ineffectiveness or due to administration's policies that create intense concentrations of the most challenging students in the toughest schools, and often deny neighborhood schools the power to enforce their academic, attendance, or behavioral standards.

Without peer review, "multiple measures" of teachers' performance are likely to mean that teachers just have to jump through multiple hoops. Similarly, teachers and our unions must fight to the end against the use of statistical models to DRIVE the evaluation process. We can not allow a statistical model to indict teachers as ineffective, so they then have to prove themselves as not guilty of being a bad teacher.

With peer review, however, test score growth estimates could be used to complement or supplement human evaluations, and as they provide a check and balance. Peer review could lead to the all-important change from data-DRIVEN to data-INFORMED accountability.

If I had a magic wand, I would still un-invent value-added models, reverse the test-driven malpractice of the last generation, and replace "the culture of accountability" with a culture of peer review. It is possible, however, that President Obama and Bill Gates are agreeing to the next best thing. We could agree to disagree about the reasons why a generation of data-driven "reform" has failed. Then we could use peer review as a step toward a truce in our educational civil war, stop teaching to the test, and get back to the goal of teaching with passion and creativity.