The small finitistic dimensions of commutative rings, III

This paper establishes a characterization of the small finitistic dimension of a commutative ring $R$ in terms of the vanishing of Ext groups for finitely generated ideals, proving that fPD$(R)\leq d$ if and only if the vanishing of $Ext_R^i(R/I,R)$ for $i=0,\dots,d$ implies its vanishing for all $i\geq 0$, and applies this result to derive the inequality fPD$(R)\leq \text{FP-}Id_RR$ and analyze various classes of rings such as $(n,d)$-rings and DW-rings.

The Gist

Here is an explanation of the paper "The small finitistic dimensions of commutative rings, III" by Xiaolei Zhang, translated into everyday language with creative analogies.

The Big Picture: Measuring the "Complexity" of a Ring

Imagine a Ring (in mathematics) as a giant, complex factory. This factory produces various products (called modules). Some products are simple and easy to build; others are incredibly intricate, requiring a massive chain of steps to assemble.

Mathematicians want to measure how "complicated" this factory is. They do this by looking at the Projective Dimension: essentially, counting the maximum number of steps required to build a product from scratch using the factory's basic building blocks.

• Global Dimension: The maximum number of steps needed for any product in the factory.

  • The Problem: In many modern, complex factories (non-Noetherian rings), some products are so weird that they require an infinite number of steps. This makes the "Global Dimension" infinite, which isn't very helpful for comparison. It's like saying a factory is "infinitely complex" just because it has one broken machine that never stops.

• The Solution (Finitistic Dimensions): To fix this, mathematicians invented the Finitistic Dimension. This only looks at products that can be finished in a finite number of steps. It asks: "What is the hardest finishable product we can make?"

The Specific Focus: The "Small" Finitistic Dimension

In this paper, the author focuses on the Small Finitistic Dimension (fPD).

• The "Little" version: Only looks at products made from a finite list of raw materials.
• The "Small" version (This paper): Looks at any product that has a finite assembly plan, even if the raw materials list is huge.

Think of it like a construction project.

• Little Finitistic: Can we build a house using only 10 specific bricks?
• Small Finitistic: Can we build a house using any amount of bricks, as long as the blueprint (the resolution) isn't infinitely long?

The Core Discovery: The "Silence" Test

For a long time, mathematicians had a rule to check if a factory's complexity was low (say, less than $d$ steps). The rule was:

"If a product requires 0 steps, 1 step, 2 steps... up to $d$ steps to prove it's 'empty' or 'zero,' then it is actually the whole factory (trivial)."

The Problem: This rule only checked the first $d$ steps. What if the product looks simple for the first 10 steps, but then explodes into chaos at step 100? The old rule didn't tell us what happens later.

The Paper's Breakthrough (The Main Theorem):
Zhang proves a powerful new rule:

"If a product looks 'silent' (zero) for the first $d$ steps, it will stay silent forever."

The Analogy:
Imagine you are testing a new car engine.

• Old Rule: If the engine doesn't make a noise in the first 5 seconds, we assume it's a perfect engine. But maybe it explodes at second 6? We didn't know.
• New Rule (This Paper): If the engine is silent for the first 5 seconds, it will never make a sound. The silence is permanent.

This allows mathematicians to determine the complexity of the entire factory just by checking a short, finite window of time. If the "noise" (mathematical complexity) stops early, it stops forever.

Why Does This Matter? (The Applications)

The paper uses this new "Silence Test" to solve several puzzles about different types of factories (rings):

1. The Self-Injective Dimension (The "Self-Healing" Metric)
Mathematicians have another way to measure complexity called the "Self-FP-injective dimension." It's like measuring how well the factory can repair its own broken machines.

• The Result: The paper proves that the "Small Finitistic Dimension" (how hard it is to build things) is always less than or equal to the "Self-Healing Dimension."
• Translation: If a factory is good at fixing itself, it can't be too hard to build things in it. You can't have a factory that is incredibly hard to build in, but incredibly easy to fix.

2. The "DW-Rings" (The "Perfectly Balanced" Factories)
There is a special class of rings called DW-rings. These are factories where every product is "well-behaved" in a specific mathematical sense.

• The Result: The paper shows that a factory is a DW-ring if and only if its complexity is very low (at most 1 step).
• Translation: If a factory is perfectly balanced, it's very simple to build things there. Conversely, if it's simple to build things, it's perfectly balanced.

3. The "Prüfer" vs. "Strong Prüfer" Confusion
There are two types of "Prüfer rings" (a specific type of well-behaved factory).

• Strong Prüfer: A very strict, perfect factory.
• Prüfer: A slightly looser, but still good factory.
• The Question: Is every "Prüfer" factory actually a "Strong Prüfer" factory?
• The Answer: No. The author builds a specific counter-example (a mathematical "monster" factory) that is a Prüfer ring but not a Strong Prüfer ring. It's like a car that drives perfectly on the highway but has a broken steering wheel in the garage. It works, but it's not "strong" enough.

Summary

This paper is like a detective story in the world of abstract algebra.

1. The Mystery: How do we measure the complexity of a mathematical ring when some parts are infinitely complex?
2. The Clue: We found that if a part looks simple for a short time, it stays simple forever.
3. The Solution: We used this clue to prove that "how hard it is to build things" is always limited by "how well the ring fixes itself."
4. The Twist: We discovered that some rings that look perfect on the surface actually have hidden flaws, proving that not all "good" rings are "strong" rings.

This work helps mathematicians organize the chaotic landscape of non-Noetherian rings, giving them better tools to classify and understand these complex structures.

Technical Summary

Here is a detailed technical summary of the paper "The small finitistic dimensions of commutative rings, III" by Xiaolei Zhang.

1. Problem Statement and Context

The paper addresses the homological complexity of commutative rings, specifically focusing on the small finitistic dimension, denoted as $fPD(R)$.

• Background: In ring theory, the global dimension often fails to be finite for non-regular Noetherian local rings or non-Noetherian rings. To address this, Bass introduced the "little finitistic dimension" ($fpD(R)$), which is the supremum of projective dimensions of finitely generated modules with finite projective resolutions. However, over non-Noetherian rings, syzygies of finitely generated modules are not necessarily finitely generated.
• The Gap: To resolve this, Glaz introduced the small finitistic dimension ($fPD(R)$), defined as the supremum of projective dimensions of all $R$-modules that possess a finite projective resolution (where the resolution consists of finitely generated projective modules).
• The Specific Question: Previous work by the author and others characterized $fPD(R)$ using special semi-regular ideals, tilting theories, and Koszul cohomology. However, a specific relationship remained unclear: What is the behavior of $\text{Ext}^i_R(R/I, R)$ for large $i$ when $fPD(R) \le d$? Specifically, does the vanishing of Ext groups for $i=0, \dots, d$ imply their vanishing for all $i \ge 0$? Furthermore, what is the relationship between $fPD(R)$ and the self-FP-injective dimension ($FP\text{-}id_R R$)?

2. Methodology

The paper employs advanced homological algebra techniques, specifically:

• Koszul Complexes and Homology: The author utilizes the Koszul complex $K_\bullet(x)$ and its cohomology $H^p(x, M)$ (Koszul cohomology) to analyze the grade of ideals.
• Koszul Duality: A key tool is the isomorphism between Koszul homology and cohomology (Lemma 2.2), which relates $H_p(x, M)$ to $H_{n-p}(x, M)$.
• Characterization via Ext Vanishing: The core methodology involves establishing a logical equivalence between the finiteness of the small finitistic dimension and the vanishing behavior of Ext groups for finitely generated ideals.
• Application to Specific Ring Classes: The results are applied to various classes of rings, including $(n, d)$-rings, DW-rings, and Pr"ufer-type rings, often using counterexamples (such as idealization of modules) to test boundaries.

3. Key Contributions and Results

A. New Characterization of $fPD(R)$

The central theoretical contribution is Theorem 2.5, which provides a necessary and sufficient condition for a ring to have a small finitistic dimension $\le d$.

• Theorem: $fPD(R) \le d$ if and only if for any finitely generated ideal $I$ of $R$, if $\text{Ext}^i_R(R/I, R) = 0$ for each $i = 0, \dots, d$, then $\text{Ext}^i_R(R/I, R) = 0$ for all $i \ge 0$.
• Significance: This resolves the open question regarding the behavior of Ext groups for large indices. It establishes that the vanishing of the first $d+1$ Ext groups for a finitely generated ideal forces the vanishing of all higher Ext groups, provided the small finitistic dimension is bounded by $d$.
• Corollary 2.6: Provides a characterization for rings with infinite small finitistic dimension ($fPD(R) = \infty$), stating that this occurs if and only if for any large integer $i$, there exists a finitely generated ideal $I$ such that $\text{Ext}^i_R(R/I, R) \neq 0$.

B. Relationship with Self-FP-Injective Dimension

The paper generalizes a known result from Noetherian rings to arbitrary commutative rings.

• Theorem 3.1: For any commutative ring $R$, $fPD(R) \le FP\text{-}id_R R$.
  • Here, $FP\text{-}id_R R$ is the self-FP-injective dimension (the least $n$ such that $\text{Ext}^{n+1}_R(N, R) = 0$ for all finitely presented modules $N$).
• Corollary 3.2: Consequently, $fPD(R) \le id_R R$ (the self-injective dimension).
• Significance: This places an upper bound on the small finitistic dimension using the injective properties of the ring itself, extending previous results that were limited to Noetherian contexts.

C. Applications to Specific Ring Classes

The author applies these general theorems to refine the understanding of several ring classes:

1. $(n, d)$-rings and Weak $(n, d)$-rings:

   • The paper proves that every weak $(n, d)$-ring (in the sense of Zhou) and every $(n, d)$-ring has $fPD(R) \le d$.
   • It also proves this for weak $(1, d)$-rings (in the sense of Mahdou).
   • Open Question: It remains an open question whether every weak $(n, d)$-ring (in the sense of Mahdou) satisfies $fPD(R) \le d$.

2. DW-rings:

   • A ring $R$ is a DW-ring if every finitely generated ideal is a GV-ideal (or equivalently, the only GV-torsion module is 0).
   • Proposition 3.9: A ring $R$ is a DW-ring if and only if $R$ itself is a "strong w-module" (a module where $\text{Ext}^i_R(T, R) = 0$ for all GV-torsion modules $T$ and $i \ge 2$).
   • Corollary 3.11: If $FP\text{-}id_R R \le 1$, then $R$ is a DW-ring.
   • Counterexample (Remark 3.12): The converse is false. The ring $R = k[[x, y]]/(x^2, xy, y^2)$ is a DW-ring (since $fPD(R)=0$) but has infinite self-FP-injective dimension. This highlights that $fPD(R)$ and $FP\text{-}id_R R$ can differ significantly.

3. Pr"ufer Rings vs. Strong Pr"ufer Rings:

   • The paper distinguishes between Pr"ufer rings (finitely generated regular ideals are projective) and Strong Pr"ufer rings (finitely generated semi-regular ideals are projective).
   • Corollary 3.14: Every Strong Pr"ufer ring is a Pr"ufer DW-ring.
   • Example 3.15: The author constructs a counterexample (using idealization $R = D(+)Q/D$) to show that a Pr"ufer DW-ring is not necessarily a Strong Pr"ufer ring. This answers a natural question in the negative.

4. Significance

• Theoretical Advancement: The paper closes a gap in the homological theory of non-Noetherian rings by providing a complete characterization of the small finitistic dimension via the vanishing of Ext groups. This moves beyond the "syzygy" limitations of the little finitistic dimension.
• Unification: It unifies the study of finitistic dimensions with FP-injective dimensions, showing that the latter serves as a robust upper bound for the former in the general commutative setting.
• Refinement of Ring Classes: By applying these results to DW-rings and Pr"ufer rings, the paper clarifies the homological distinctions between these classes, particularly demonstrating that the property of being a DW-ring is strictly weaker than being a Strong Pr"ufer ring.
• Future Directions: The work identifies specific open problems, such as the characterization of weak $(n, d)$-rings in the sense of Mahdou, guiding future research in non-Noetherian homological algebra.