TE
科技回声
首页24小时热榜最新最佳问答展示工作
GitHubTwitter
首页

科技回声

基于 Next.js 构建的科技新闻平台,提供全球科技新闻和讨论内容。

GitHubTwitter

首页

首页最新最佳问答展示工作

资源链接

HackerNews API原版 HackerNewsNext.js

© 2025 科技回声. 版权所有。

Research as a Stochastic Decision Process

2 点作者 Luc超过 1 年前

1 comment

082349872349872超过 1 年前
Someone (I had thought it was MJD, but uncle GOOG is not pulling it up on his blog) once told a story about how software estimation was a bit like transfinite recursion on ordinal numbers: sometimes you have an estimate in ℕ, and it&#x27;s possible to say &quot;I&#x27;ll be done in 5 days, give or take&quot;, but sometimes your estimate is above omega, and the best you can do is tell someone who asks &quot;Right now, I don&#x27;t know how long it&#x27;ll take, but come back in 5 days (give or take) and then I can tell you&quot;.<p>(and it wasn&#x27;t JDH, either: closest I get there is <a href="https:&#x2F;&#x2F;jdh.hamkins.org&#x2F;transfinite-nim&#x2F;" rel="nofollow">https:&#x2F;&#x2F;jdh.hamkins.org&#x2F;transfinite-nim&#x2F;</a> )