2

I will state it again. Given two rings R_1 and R_2 (with or without identity. It's not specified.). If R_1[x] is isomorphic to R_2[y] (No such requirement that the isomorphism sends the constant terms to constant terms), can we deduce that R_1 iso. to R_2?

I feel there might be a counterexample but it's quite hard to find one.

flag
2 
This is the cancellation problem of Zariski. There are many counterexamples. See math.lsa.umich.edu/~hochster/Lip.text.pdf – J.C. Ottem Dec 8 2010 at 14:19
This (interesting) question has received many good answers on math.SE: math.stackexchange.com/questions/13504/…. I have voted to close it here as "no longer relevant". – Pete L. Clark Dec 9 2010 at 1:20

2 Answers

6

This was recently asked and answered at http://math.stackexchange.com/questions/13504/does-rx-cong-sx-imply-r-cong-s As answered there, it is possible to find two non-isomorphic commutative rings whose polynomial rings in one variable are isomorphic. An example is given in http://www.ams.org/journals/proc/1972-034-01/S0002-9939-1972-0294325-3/home.html

link|flag
3

Let X be an affine variety with two non-isomorphic vector bundles V and W that become isomorphic after adding a trivial line bundle to each. Then the coordinate rings of the total spaces of V and W should yield a counterexample. (Though you might have to do some extra work to verify that the total spaces of V and W are non-isomorphic as varieties.)

The counterexample cited by Tobias takes X to be the 2-sphere, V the tangent bundle to X, and W the trivial plane bundle over X.

link|flag

Your Answer

Get an OpenID
or

Not the answer you're looking for? Browse other questions tagged or ask your own question.