Now he tells us..

The architect of NCLB is having second thoughts - additional evidence that a national consensus on education may be developing.

(Yet more evidence: a recent national survey of educators ranked 21st century skills - such as the ability to collaborate, innovate and create - as the most important priority for the new administration.)

In an article in The National Review (from which I quote liberally), Michael Petrilli, the co-author of "No Child Left Behind: A Primer" writes that he's "reluctantly come to the conclusion that NCLB as enacted is fundamentally flawed and probably beyond repair."

Petrilli now concedes that requiring all states to reach proficiency by 2014 while allowing them to define proficiency was spurring a "race to the bottom".

He also admits to other problems that took him longer to recognize, such as "the conversion of schools into test-prep factories," and the fact that school-choice laws are meaningless when in most of our big cities, there are too few good schools to go around.

He even admits that a change of focus is needed from teacher "quality" to teacher effectiveness - which is probably better handled at the state level.

However, Petrilli remains a supporter of the ideas underlying the law. Foremost: that virtually all children have the capacity to achieve a reasonable level of proficiency in reading and math by the time they turn 18 — and that it’s the education system’s job to make sure they do.

Petrilli still believes that improving education is a national imperative, and that the federal government can and should play a constructive role - which would begin with "a more realistic assessment of what the federal government can reasonably hope to achieve".

Certainly, the federal government has a role in ensuring that high poverty schools receive equitable resources.  It would also serve a useful function if it collected and published reliable and comparable data on the performance of the nation’s schools.