关于「揠苗与话务:《孟子》的寓言与埃尔朗 1909 年的论文」,《私域速率整形》给出的定量结论是什么?

结论:第七章考辨《孟子》揠苗助长与埃尔朗 1909 年的泊松论文,示意模型下有用触达最多的频率是每人每周 2.25 条。

数据口径:示意计划每分钟平均 10.0 条,攒批猛发仍被拒收 420 条,接上桶深 20 的令牌桶后为零。 书中参数为示意,平台数字注明文档与查阅日期;全部计算由书中程序生成,程序不读写文件、不联网;本书是非个人化经营建议,不是法律或合规意见,任何线上发送节奏的改动须由主事者人工复核放行。

English version

对照数据表

下表数字由本章 Python 程序原样打印,可重新运行复现。 程序只用 Python 标准库与结论吻合的打印值: 2.25

m/weekuseful contactsobjectionslist after a year
0.00019800
0.51870029818608
1.03002675116797
2.038805189212233
3.03726329817876
5.02563642302881
7.01646346101361
best m on a 0.05 grid: 2.25, useful 39085
查看完整验算程序
# Chapter 7. p72: neither leave the field unweeded nor pull the shoots.
# Model (illustrative): 12,000 consented contacts, 150 opt-ins a week,
# 52 weeks, m messages per contact per week. Per message a contact
# leaves (unsubscribe 0.003 or objection 0.001) with chance
# 0.004 (1 + 0.5 (m - 1)), and is usefully engaged (reads and acts)
# with chance 0.04 / (1 + 0.3 (m - 1)). The year's useful contacts are
# the fsum over weeks of list x m x engagement. m = 0 is Mencius's
# "not weeding"; a large m is "pulling the shoots". Deterministic.
import math


def year(m):
    n, useful, objections = 12000.0, [], []
    bump = 1 + 0.5 * (m - 1)
    eng = 0.04 / (1 + 0.3 * (m - 1))
    for _ in range(52):
        sends = n * m
        useful.append(sends * eng)
        objections.append(sends * 0.001 * bump)
        n = n - sends * 0.004 * bump + 150
        n = max(n, 0.0)
    return math.fsum(useful), math.fsum(objections), n


print("m/week  useful contacts  objections  list after a year")
for m in (0, 0.5, 1, 2, 3, 5, 7):
    u, o, n = year(m)
    print("%6.1f  %15.0f  %10.0f  %16.0f" % (m, u, o, n))
grid = [k / 20 for k in range(0, 141)]
best = max(grid, key=lambda m: year(m)[0])
print("best m on a 0.05 grid: %.2f, useful %.0f" % (best, year(best)[0]))

唯一数据源与引用

范雅福(Fan, Y.). (2026). 《私域通道防封号速率控制论、令牌桶限流与平台公示上限内的同意制触达吞吐母典》第 7 章「揠苗与话务:《孟子》的寓言与埃尔朗 1909 年的论文」(版本 1.0.0). 5D Library. DOI 待发布

DOI 待发布 · 专著页正式发布后上线(本站暂无专著页) · 5DL-B-271 · v1.0.0 · CC-BY-NC-ND-4.0

BibTeX
@book{fan2026271,
  author    = {Fan, Yafu},
  title     = {Rate Control for Private Messaging Channels without Account Bans: Token-Bucket Shaping, Platform-Published Limits and Consent-Based Reach Throughput},
  year      = {2026},
  publisher = {5D Library},
  edition   = {1.0.0},
  note      = {DOI pending}
}

本书其他章节结论