On an article published in the racsam

Nguyễn Sum · 2025

Read On an article published in the racsam by Nguyễn Sum (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2502-11119.txt.

Introduction

Nguyen Sum writes a short note disputing a 2024 journal article on the hit problem for the polynomial algebra at odd primes. That paper appeared in an academy journal of mathematics and gave an application of its results. His verdict is blunt: some of the results, he argues, are not new, and others are false. The note sets out comments to support both charges. It is a corrective piece rather than a new study, meant to put the record straight on that article's claims.

Catalog record

CREATORNguyễn Sum
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE12 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2502-11119

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

On an article published in the racsam Nguyễn Sum Abstract . In this Note, we provide the comments on the paper [ 3 ] [ A note on the hit problem for the polynomial algebra in the case of odd primes and its application ] which was published in the RACSAM [Rev. Real. Acad. Cienc. Exactas. Fis. Nat. Ser. A-Mat. 118, 22 (2024)]. We will show that some of the results in this paper are not new and others are false. 1 1 footnotetext: 2000 Mathematics Subject Classification . Primary 55S10; 55T15. 2 2 footnotetext: Keywords and phrases: Steenrod operation, Peterson hit problem, polynomial algebra. 3 3 footnotetext: The main contents of this Note have been submitted to Mathematical Reviews of AMS and to the managing Editors of the RACSAM. 1. Introduction Let p be an odd prime number and \mathscr{A}_{p} be the mod- p Steenrod algebra. Denote by P_{h}=\mathbb{F}_{p}[t_{1},t_{2},\ldots,t_{h}] the polynomial algebra over the field \mathbb{F}_{p} with p elements in h generators t_{1},t_{2},\ldots,t_{h} , each of degree 2. This algebra is regarded as an \mathscr{A}_{p} module. Let GL(h,\mathbb{F}_{p}) be the general linear group of rank h over \mathbb{F}_{p} which acts regularly on P_{h} . In the paper [ 3 ] , the author presented some results on \mathscr{A}_{p} -generators of P_{h} and applied them to study the third mod- p Singer algebraic transfer which was defined in Crossley [ 2 ] . The main results of the paper are theorems numbered 2.1 and from 2.4 to 2.8. However, some of the results in the paper [ 3 ] are not new and others are false. We use the notations as in [ 3 ] . 1. It is well known that P_{1} is an \mathscr{A}_{p} submodule of P_{2} . A monomial basis for \mathbb{F}_{p}\otimes_{\mathscr{A}_{p}}P_{1} is a subset of the one for \mathbb{F}_{p}\otimes_{\mathscr{A}_{p}}P_{2} , hence Theorem 2.1 is only a special case of Theorem 1.1 on Page 172 of the paper Crossley [ 1 ] . It is not new. 2. On Page 5, the author presented an example which substantially overlaps with the one in Crossley [ 1 , Page 175] . However, the author did not cite the paper [ 1 ] of M. D. Crossley. 3. There are serious mistakes in the proof of Theorem 2.4, so I think that this theorem is false. 4. The part “ g^{p} is hit” of Theorem 2.5 is not new. It is only the unstability of \mathscr{A}_{p} module P_{h} which is listed in Tannay-Oner [ 4 , Prop. 1.4 on Page 16] . 5. The statement of Theorem 2.6 is not correct. Moreover, the result of this theorem is trivial. 6. Theorem 2.7 is stated without proof. However, its statement is also not correct. I think that the author did not prove this theorem. Hence, its accuracy is not determined. 7. The proof of Theorem 2.8 is refused with the following reasons: - The computation of [(\mathbb{F}_{p}\otimes_{\mathscr{A}_{p}}H^{*}(V;\mathbb{F}_{p}))_{n^{(s)}}]^{GL(3,\mathbb{F}_{p})} is based on a basis of the vector space (\mathbb{F}_{p}\otimes_{\mathscr{A}_{p}}H^{*}(V;\mathbb{F}_{p}))_{n^{(s)}} . However, one doesn’t know any basis for this space yet. - This proof is based on Theorems 2.4, 2.6, 2.7 but the accuracy of these results are not determined. - There are infinite many odd prime numbers but computer programs based on algorithm of MAGMA can only check the space (\mathbb{F}_{p}\otimes_{\mathscr{A}_{p}}H^{*}(V;\mathbb{F}_{p}))_{n^{(s)}} for a finite number of prime p . Thus, the results of Theorem 2.8 and Corollary 2.11 have not yet been accurately determined. Below is the detailed comments for the above items. 2. The detailed comments 1. In [ 1 , Page 172] , Crossley stated Theorem 1.1 as follows: Theorem 1.1. As far as it goes, Table 1 gives a monomial basis for M^{*}(2) . Table 1 Degree, n Basis for M^{n}(2) \leq 2(p-2) , n even \{{\color[rgb]{1,0,0}x^{i}y^{n/2-i}}\,|\,0\leq i\leq n/2\} 2(((i+1)p+j+1)p^{s}-2) \{x^{(k+1)p^{s}-1}y^{n/2-(k+1)p^{s}+1}\,|\,\min(i+1,j)\leq k\leq p-1\} 0\leq i,\,j\leq p-1,\,s\geq 0 \cup\{{\color[rgb]{1,0,0}x^{(k+1)p^{s+1}-1}y^{n/2-(k+1)p^{s+1}+1}}\,|\,0\leq k\leq i\} 2(((i+1)p^{r}+j+1)p^{s}-2) \{x^{(k+1)p^{s+1}-1}y^{n/2-(k+1)p^{s+1}+1}\,|\,1\leq k\leq p-1\} 1\leq i,\,j+1\leq p-1, \cup\{x^{(j+1)p^{s}-1}y^{(i+1)p^{r+s}-1},\,{\color[rgb]{1,0,0}x^{(i+1)p^{r+s}-1}y^{(j+1)p^{s}-1}}\} r\geq 2,\,s\geq 0 2((p^{2}+ip+j+1)p^{s}-2) \{x^{(k+1)p^{s+1}-1}y^{n/2-(k+1)p^{s+1}+1}\,|\,i\leq k\leq j\} 1\leq i\leq j\leq p-2,\,s\geq 0 Statement 1.2. In the degrees, n , not dealt by Table 1, M^{*}(2) is 0 . Note that x=t_{1},\,y=t_{2} and M^{*}(2)=\mathbb{F}_{p}\otimes_{\mathscr{A}_{p}}P_{2} . A monomial basis of M^{*}(1)=\mathbb{F}_{p}\otimes_{\mathscr{A}_{p}}P_{1} consists of monomials in the basis for M^{*}(2) whose power of y=t_{2} is 0. Hence, from Theorem 1.1 one gets (the red monomials with power 0 for y ). Corollary. Table 2 gives a monomial basis for \mathbb{F}_{p}\otimes_{\mathscr{A}_{p}}P_{1}=M^{*}(1) . Table 2 Degree, n Basis for M^{n}(1) Note \leq 2(p-2) , n even \{x^{n/2}\} The first red monomial with i=n/2 2(i+1)p-2 \{x^{(i+1)p-1}\} The second red monomial with 0\leq i\leq p-1 s=j=0,k=i 2(i+1)p^{r}-2 \{x^{(i+1)p^{r}-1}\} The last red monomial with 1\leq i\leq p-1,\,r\geq 2 s=j=0 Thus, Theorem 2.1 in this paper overlaps with this corollary and Statement 1.2. Hence, it is not new. 2. In [ 1 , Page 175] , Crossley presented an example as follows: “For example, let p=5 and consider the element x^{4}y^{5}+x^{8}y:\mathscr{P}^{1}(x^{4}y^{5}+x^{8}y)=(4x^{8}y^{5}+5x^{4}y^{9})+(8x^{12}y+x^{8}y^{5})=3x^{12}y . Thus x^{12}y is hit by \mathscr{P}^{1} .” On Page 5 of the paper [ 3 ] , the author presented an example as follows: “For instance, with p=5 and the monomial f=3t_{1}^{12}t_{2}\in P_{2}=\mathbb{F}_{5}[t_{1},t_{2}] , we see that f is in the image of \mathscr{A}_{5} since f=\mathscr{P}^{1}(t_{1}^{4}t_{2}^{5}+t_{1}^{8}t_{2}) .” This example substantially overlaps with the one of Crossley. However, the author did not cite the paper [ 1 ] of M. D. Crossley. 3. The proof of Theorem 2.4 is presented on Page 5 as follows: “We use induction on h . Obviously, the theorem is true for h=1 . Suppose that the theorem is true for P_{h-1} . We write f=tf_{1}+f_{2} , in which t=t_{h},\,f_{1}\in P_{h},\,{\color[rgb]{1,0,0}\deg(f_{1})=n-2(p-1)} and f_{2}\in P_{h-1} , \deg(f_{2})=n . Then by the Cartan formula, we get 0=\mathscr{P}^{1}(f)=\mathscr{P}^{1}(tf_{1})+\mathscr{P}^{1}(f_{2})=t\mathscr{P}^{1}(f_{1})+t^{p}f_{1}+\mathscr{P}^{1}(f_{2}). As it is known, every term of \mathscr{P}^{k}(f_{2})\,(k\geqslant 0) involves exactly the same variables as f_{2} does. So, since f_{2} is independent of t , setting t=0 , we get \mathscr{P}^{1}(f_{2})=0 and so t\mathscr{P}^{1}(f_{1})+t^{p}f_{1}=0 . These, together with the inductive hypothesis, yield that {\color[rgb]{1,0,0}\mathscr{P}^{1}(f_{1})=tf_{1}} and f_{2}=\mathscr{P}^{1}(g_{2}) for some g_{2}\in P_{h-1} , \deg(g_{2})=n-2(p-1) . Thus, {\color[rgb]{1,0,0}f=tf_{1}+f_{2}=\mathscr{P}^{1}(f_{1})+\mathscr{P}^{1}(g_{2})=\mathscr{P}^{1}(f_{1}+g_{2})} . This completes the proof.” This proof is false because the author wrote f=tf_{1}+f_{2} with t=t_{h} , hence the relation \deg f_{1}=n-2(p-1) is false because \deg t=2 , n=\deg f and \deg f_{1}=n-2>n-2(p-1) . Since p is odd, the relation t\mathscr{P}^{1}(f_{1})+t^{p}f_{1}=0 implies \mathscr{P}^{1}(f_{1})=-t^{p-1}f_{1}\neq tf_{1} for all f_{1}\neq 0 . Hence, the relations \mathscr{P}^{1}(f_{1})=tf_{1} and f=tf_{1}+f_{2}=\mathscr{P}^{1}(f_{1}+g_{2}) are false. 4. Theorem 2.5 is only an easy consequence of the unstability of \mathscr{A}_{p} module P_{h} and the Cartan formula, however there are many errors in its proof (some of them are corrected in the Correction of [ 3 ] ). Moreover, for any homogeneous element g\in P_{h} , one has g^{p}=\mathscr{P}^{\frac{1}{2}\deg g}(g) (see Proposition 1.4 in the paper Tannay-Oner [ 4 , Page 16] ). Hence, the condition \deg g=2p^{j} is unnecessary. 5. The author stated Theorem 2.6 on Page 6 as follows: Theorem 2.6. For positive integers t , q_{i},\,1\leqslant i\leqslant h , let \varphi:P_{h}\to P_{h} be the linear map defined by \varphi(f)=t_{1}^{q_{1}p^{t+1}}t_{2}^{q_{2}p^{t+1}}\ldots t_{h}^{q_{h}p^{t+1}}f^{p} for all f\in P_{h} . Then if f is in the image of \mathscr{A}_{p} via f=\sum_{0\leq k\leq t-1}\mathscr{P}^{k}(f_{k}) for some f_{k}\in P_{h} , then then \varphi induces a homomorphism (\mathbb{F}_{p}\otimes_{\mathscr{A}_{p}}P_{h})_{n}\to(\mathbb{F}_{p}\otimes_{\mathscr{A}_{p}}P_{h})_{pn+2p^{t+1}\sum_{1\leqslant i\leqslant h}q_{i}} for all n\geq 0 . The statement of this theorem

Read the complete text