What does Retention Algebra for Book Adaptations conclude, in numbers, about “Spoken Compression: Breadth against Repetition, Turn Length and Shannon Naming in a Two-Host Podcast”?

Answer: Chapter 2 shows that a thirty-minute podcast carrying about 10 arguments keeps 0.4820, while stating 20 arguments once each keeps only 0.3117, and that the best uninterrupted turn of one host lasts about 41.5 seconds.

Scope of the numbers: Its parameters are illustrative; all calculations are generated by its programs; it is not legal advice and not personalised business or investment advice. Marked boxes separate published evidence, runnable Python calculations (22 programs), classical sources and the author's hypotheses.

中文版

Data table

The numbers below are printed verbatim by the chapter's Python program and can be reproduced by running it. standard-library Python onlyprinted values that match the answer: 0.3117, 0.4820, 10, 20

ideasshare of bookretained weightexposures (top, last)
20.05000.26042018
40.10000.3801108
60.15000.445775
80.20000.475263
100.25000.482052
120.30000.476351
140.35000.462451
160.40000.440841
180.45000.402231
200.50000.311711
best breadth 10 ideas, retained weight 0.4820
stating all 40 once would need 60.0 minutes
Show the full program
# Chapter 2. p21: breadth against repetition in a 30-minute two-host show.
# The monograph holds 40 ideas with importance w_i proportional to
# 1 / i^0.8. A listener cannot rewind, so an idea stated once is kept with
# probability q = 0.40; each restatement (an example, a question from the
# second host, a recap) is another independent chance, so k exposures
# keep it with 1 - (1 - q)^k. Stating an idea costs 1.5 minutes, each
# restatement 0.75. For every breadth n (top n ideas) the spare minutes
# go greedily to the restatement with the largest weighted gain; the
# objective is separable and concave in k, so greedy is exact here.
# Times, q and the importance curve illustrative.
import math

N_IDEAS, MINUTES, FIRST, AGAIN, Q = 40, 30.0, 1.5, 0.75, 0.40
W = [1.0 / (i + 1) ** 0.8 for i in range(N_IDEAS)]
TOTAL = math.fsum(W)


def keep(k):
    return 1.0 - (1.0 - Q) ** k


def best_for(n):
    spare = MINUTES - n * FIRST
    if spare < 0:
        return None
    k = [1] * n
    for _ in range(int(spare / AGAIN + 1e-9)):
        j = max(range(n), key=lambda i: W[i] * (keep(k[i] + 1)
                                                - keep(k[i])))
        k[j] += 1
    kept = math.fsum(W[i] * keep(k[i]) for i in range(n))
    return kept / TOTAL, k


print("ideas  share of book  retained weight  exposures (top, last)")
best = (0.0, 0)
for n in (2, 4, 6, 8, 10, 12, 14, 16, 18, 20):
    r = best_for(n)
    if r is None:
        continue
    frac, k = r
    best = max(best, (frac, n))
    print("{:5d}  {:13.4f}  {:15.4f}  {:9d} {:9d}".format(
        n, n / N_IDEAS, frac, k[0], k[-1]))
print("best breadth {} ideas, retained weight {:.4f}".format(
    best[1], best[0]))
print("stating all 40 once would need {:.1f} minutes".format(
    N_IDEAS * FIRST))

Single source and citation

Fan, Y. (2026). Content Fission Algebra: Compiling Hard-Core Monographs into Two-Host Podcasts, Study Slides and Quiz Banks (Version 1.0.0), Chapter 2: Spoken Compression: Breadth against Repetition, Turn Length and Shannon Naming in a Two-Host Podcast. 5D Library. DOI pending

DOI pending · monograph page goes live after publication · 5DL-B-151 · v1.0.0 · CC-BY-NC-ND-4.0

BibTeX
@book{fan2026151,
  author    = {Fan, Yafu},
  title     = {Content Fission Algebra: Compiling Hard-Core Monographs into Two-Host Podcasts, Study Slides and Quiz Banks},
  year      = {2026},
  publisher = {5D Library},
  edition   = {1.0.0},
  note      = {DOI pending}
}

Other chapter answers in this book