Here's a (potentially silly) question: must a local homomorphism of noetherian local rings $A\to B$ which is finite type also make $B$ a finite $A$-module? It seems like either there should be a well-known proof or a well-known counterexample, but my google skills (and my head) are coming up empty right now.
Hi Let $F$ be a field and $K$ an extension of $F$. the element $a$ in $K$ is algebraic over $F$ if and only if $F (a)$ is a finite extension of $K(a)$ I think this is true??
@KReiser I am not an expert, but can it be argued as follows? Suppose the local rings are $(A, \mathfrak{m}), (B, \mathfrak{n})$. If $B$ is a finite type $A$-algebra, $B/\mathfrak{n}$ is a finite type $A/\mathfrak{m}$-algebra. By Nullstellensatz field extensions which are finite type are finite, hence $A/\mathfrak{m}$ is a finitely generated $B/\mathfrak{n}$-module. By Nakayama's lemma, this basis gives rise to a generating set of $B$ as an $A$-module, not?
Ah, no, ignore me. I need to assume $B$ is a finitely generated $A$-module in the first place for Nakayama.
Haha, there are actually two issues with what I am saying. (1) I am confusing $B/\mathfrak{n}$ with $B/\mathfrak{m}B$; there's no reason $\mathfrak{m}B = \mathfrak{n}$, unless the original map of local rings was unramified. (2) The form of Nakayama I want to apply is $M$ is an $A$-module and $M/\mathfrak{m}M$ is a finitely generated $A/\mathfrak{m}$-vector space, and to get that $M$ is a finitely generated $A$-module out of there. I think this is true if $A$ is Artinian
@KReiser @loch Consider $k[t^2]_{(t^2 -1)} \subset k[t]_{(t-1)}$ where $k$ is algebraically closed. This is finite type, because anything in the codomain can be written as $f(t)/\prod(t - c_i)$ where none of the $c_i$'s are $1$. Multiply top and bottom by $t + c_i$ whenever $c_i \neq -1$ to get something of the form $P(t)/(Q(t^2) \cdot (t + 1)^n)$ where $t^2-1$ does not divide $Q(t^2)$. So the latter can be obtained from the former by adjoining $t$ and $1/(t + 1)$
2
Doesn't look like the latter is finite over the former though. I think $t + 1$ in the latter ring is not integral over the former
Finite extensions are integral, so that can't happen
After my observation on the unramified thing I was trying $k[t^2]_{(t^2)} \subset k[t]_{(t)}$ for a long time but that's actually finite :)
Turns out the answer to this problem is yes, any local ring homomorphism of noetherian local rings of finite type actually makes the target a finite module over the source.
Let our rings be $(A,\mathfrak{m})$ and $(B,\mathfrak{n})$ with $f:A\to B$. Then $B\cong A[x_1,\cdots,x_n]/(f_1,\cdots,f_m)$, and base-changing by $A\to A/\mathfrak{m}$, we see that $B/\mathfrak{m}B\cong (A/\mathfrak{m})[x_1,\cdots,x_n]/(\overline{f_i})$. But the quotient of a local ring is local, so this ring has one maximal ideal. By Noether normalization, this means it's a finite module over $A/\mathfrak{m}$, and by fpqc descent, this means $B$ is module-finite over $A$.
@KReiser This is cool! Can you explain why $B/\mathfrak{m}B$ is a finite $A/\mathfrak{m}$-module implies $B$ is a finite $A$-module? I wanted to do something like this earlier using Nakayama but thought I needed $\mathfrak{m}$ to be nilpotent in $A$ for this (aka $A$ is Artinian). I don't know fqpc descent though :)
Oh crap did I verify faithfully flat? Uhh gotta check that
@BalarkaSen Anyways your example is a localization of $k[t^2,t^{-2}]\to k[t,t^{-1}]$, which is a finite flat map of rings (as long as characteristic isn't 2), I think.
@BalarkaSen Descent is basically the converse of base-change: if you have a base change square (arrows point down and right) with the bottom map nice enough and the left hand map with some property, then the right hand map must have that property sometimes too.
The reason it's the converse is that usually we make a conclusion about the left arrow from knowing something about the right arrow.
Hm, I don't quite see why it should be finite though, because I think integral closure of $k[t^2]_{(t^2-1)}$ in $k(t)$ is essentially germs of functions vanishing at both the preimages of $1$ by $t \mapsto t^2$, so $t+1$ is there in particular
I mean, I think it's finite because we have a basis: $\{1,t\}$.
Sorry let me explain better: $\{1,t\}$ is a basis of $k[t,t^{-1}]$ as a $k[t^2,t^{-2}]$-module, so it's finite free and thus flat. These things are preserved by localization, which is what you've done, right?
Let me see if I follow. How are you realizing $k[t]_{(t-1)}$ as a localization of $k[t, t^{-1}]$?
Ok, just localize at $(t-1)$
Ok, so you're arguing it's flat. That seems right to me. I am still not convinced that $k[t^2]_{(t^2-1)} \to k[t]_{(t-1)}$ is finite though; there must be something wrong
@BalarkaSen I've convinced myself you're correct. We can write the map as a composition $k[t^2]_{(t^2-1)}\to S^{-1}k[t] \to k[t]_{(t-1)}$ where $S$ is the multiplicatively closed set of polynomials in $t^2$ not vanishing at $t=\pm1$. Then this first map gives the middle term as a finite free module over the first with basis $\{1,t\}$, and so it's finite type and flat. The second map comes from inverting $t+1$, which is flat and finite-type, so we win. Thanks again for the help!
Any tips on how not to make simple mistakes or catch them quickly? I just spent half an hour making sure all the calculations were correct and thinking the problem had to do with how the modulus was taken. Turned out I just swapped the value of two variables.
Let $R$ be an integral domain with $1$ that is multiplicatively generated by the set $G$ and $Q = \text{Frac}(R)$ its field of fractions.
Suppose that there exists a group homomorphism $d : (Q, \cdot) \twoheadrightarrow (R, +)$ from the multiplicative structure of the field onto the additive str...
A countable set is more or less like a sequence. If your set is countable, you can list off the elements --- $x_1, x_2$, and so on --- so that for any element of your set, it gets listed off in finite time. (It won't take forever to see it, it might just take a long time.)
An uncountable set is a set that's so large you can't list off the elements.
It's sort of famously hard to see that uncountable sets exist; the argument is a little tricky.
@Knight It depends on exactly how the "mass balance" works. Most scales actually measure force, not mass. Force times time is momentum, and the added $6.7$ kg at $9.81\text{ m}/\text{s}^2$ times $90\text{ ms}$ would provide $6.7\cdot9.81\cdot0.09/0.35=16.9\ \text{m}/\text{s}$. This means to me that the ball bounced back at $6.9\text{ m}$/$s$
@robjohn Sir but I’m getting a different answer. $$\text{Initial momentum of ball}= 3.5 \\ \text{Final momentum of ball}= p \\ \text{Time elapsed}= 90 \times 10^{-3} \\ \text{Force applied = Force experienced} = 6.7 \times 9.8 $$
$$F =\frac{p- 3.5}{90 \times 10^{-3}}\\ p = 9.4094$$ and therefore the final velocity is $9.4304/0.35= 26.875$
I think the chat is now kinda normal. Because you guys start to use Lol ,XD...etc. I was think everyone here are sensitive and serious. Also you will get flagged easily 😂😂😂.
A choice of a local orientation of $\Bbb R^n$ at the origin is equivalent to choosing a vector space orientation of $T_0\Bbb R^n\simeq \Bbb R^n$ which is in turn equivalent to choosing a basis $(\mathbf{e}_1, \mathbf{e}_2, \cdots, \mathbf{e}_n)$ and remarking that every basis $\mathbf{b}$ of $\Bb...
Lmao just that they make the same long exact sequence (Gysin)
i have no idea why they are the same maps; i know how to interpret the weird map in Gysin as "cup product with Euler class" and i know how to compute it separately from specseq as induced from the transgression
i never figured out why they match
it just makes sense because most long exact sequences irl are natural
so it should be the same maps lol - bullshit reason
Why are math people so serious and get offended so easily lol. I got flagged lmao and couldn't talk for 30 min. BTW I am math people myself with asperger's syndrome.
Necessary background: prime $\Omega$ extended to all of $\Bbb{Q}^{\times}$.
The twin prime conjecture is that $\Omega^{-1}(2) \cap (K - 1) \cap \Bbb{Z}$ is an infinite set where $K = \{ x^2 : x \in \Bbb{Q}^{\times}\}$ is the subgroup.
This is true since if $\Omega(x) = 2$ then $x = pq$ and if $...
@StupidKid oeis.org/A003418 "An assertion equivalent to the Riemann hypothesis is: | log(a(n)) - n | < sqrt(n) * log(n)^2. - Lekraj Beedassy, Aug 27 2006." (due to L. Schoenfeld)
I have been bugging my head for a while now, does anyone know how to show that 3 is a primitive root modulo 7^2 (i wonder how one can do that efficiently)
@Knight thanks for the welcome! indeed i've been extremely busy with life and inefficient with my time management
@Knight confinement is doing wonders to my mental and physical health /s + we have so much more work for college, sitting in front of a computer for a longer time is causing me back problems again
@TobiasKildetoft i got one idea, there's this theorem that if $s$ is a primitive root mod $p$ then either $s$ or $s+p$ is a primitive root mod $p^2$, maybe if i show that $s+p$ is not a primitive root it may turn out to be easier
@Knight what about you? what did you learn during this time? :)
@Knight yes, I don't like the atmosphere surrounding courses, it stresses you out and doesn't give you enough time to digest the material unlike if you were instead a self learner
Let $X_n$ be separable spaces for each $n$.
Then, $X=\prod_nX_n$ is separable. (n is a natural number)
Note: I am considering the product topology.
My attempt:
Let $D_n$ denote the countable dense subset of $X_n$.
For each $n$, choose $\textbf{x}_n\in X$ such that $x_n\in D_n$ .
Put $D=\bigcup_...
Is there some characteristics exist for one to critique this simple new way of solving qudratic equation without looking or comparing to the old way? Is there any challenging features on those problem exist to solving those quadratic equation even for mathmatician given only pen and paper?
Elabor...
what does it mean to critique a way of solving a problem? mathematically speaking, it works and that's about it. is your question of didactical nature?
Hi, does that formula able to work for the quaduatic formula of the form ax^2+bx+c=0, if not, is there any challenging feature to solve these equation by hand except factoring assuming both root are real also rational.
Hi, anybody can make good suggestion will be greatly appreciate!