Mass movements tend very strongly to require mass communications.<p>Closed, encrypted, closed-access discussions are the precise opposite of mass communications.<p>And from the perspective of operational security, encryption within large groups tends to be of highly limited value in preventing information dissemination. If your adversary can simply join the group or turn existing members, ... or host the service itself directly ... then confidentiality is lost.<p><i>Large-scale subversive movements rely on impunity, not tradecraft.</i> They operate in the open <i>because they can</i>, or possibly <i>because they have no effective fear of doing so</i>. This might be through allies, self-defence capabilities, counterstrike or retaliation capacity, or other factors.<p>Secrecy, tradecraft, encryption, and the like are not entirely useless. They are vital to strategic planning <i>within a small and trusted core leadership</i>, and also over generally short periods of time during active operaton to maximise enemy uncertainty over actual plans and activity. Deep-strike or retaliatory-strike teams may operate under cover or concealment, but typically reveal their position when striking (snipers, SBLMs, wet-ops teams). Cyberwarfare is an exception to this general rule: computer networks themselves elevate uncertainty over position and identity, though routes and peers are typically apparent.<p>As Google+ shut down from September 2018--April 2019, I helped with assessments of migrating users and communities elsewhere. A large part of that was assessing just how many groups existed, how many members, and how much activity. Available data suggested about 5 million groups. Sitemaps revealed over 8 million by shutdown date, growing rapidly even in the final weeks, though also showing many deletions.<p>In particular, G+ Communities could be Open (anyone can join), Closed (moderator must approve), or Private (activity and membership hidden). Both "open" and "closed' communities are "Public".<p>90% of communities were public, 10% private.<p>Of the public communities, roughly equal numbers were Open (55%) and closed (45%). But the mean <i>membership</i> was 91% open, 9% closed. <i>Simply requiring moderator approval reduces membership by ~90%.</i> Those of you designing intake funnels, take note.<p>Private community size wasn't available to me, but I strongly suspect a similar magnitude effect.<p>Extremists grow movements when they can operate overtly. Being pushed underground will impact surveillance relatively modestly, but organising and recruiting bigly, in my belief.<p>Further notes and graphs on G+ community characteristics:<p><a href="https://social.antefriguserat.de/index.php/Migrating_Google%2B_Communities#Google.2B_Community_Characteristics_and_Membership" rel="nofollow">https://social.antefriguserat.de/index.php/Migrating_Google%...</a><p><a href="https://old.reddit.com/r/plexodus/comments/9zx67d/google_communities_membership_analysis_preview/" rel="nofollow">https://old.reddit.com/r/plexodus/comments/9zx67d/google_com...</a><p>G+ Communities membership distribution
<a href="https://joindiaspora.com/posts/ab6a5470f57001368d400218b79d8c8d" rel="nofollow">https://joindiaspora.com/posts/ab6a5470f57001368d400218b79d8...</a><p>G+ Communities: Most Recent post date by month
<a href="https://joindiaspora.com/posts/23acfd90f05d01368b220218b79d8c8d" rel="nofollow">https://joindiaspora.com/posts/23acfd90f05d01368b220218b79d8...</a><p>G+ Communities Activity Analysis: It’s the Posts, Stupid
<a href="https://joindiaspora.com/posts/7fce6710f55001368d3a0218b79d8c8d" rel="nofollow">https://joindiaspora.com/posts/7fce6710f55001368d3a0218b79d8...</a><p>Plot Community size vs. frequency, log-log.
<a href="https://joindiaspora.com/uploads/images/4c728664f0f342e0203f.png" rel="nofollow">https://joindiaspora.com/uploads/images/4c728664f0f342e0203f...</a>