-# Limits the number of processes spawned
-# This does not depend on Danga::Socket or any other external
-# scheduling mechanism, you just need to call start and finish
-# appropriately
+# Like most Perl modules in public-inbox, this is internal and
+# NOT subject to any stability guarantees! It is only documented
+# for other hackers.
+#
+# This is used to limit the number of processes spawned by the
+# PSGI server, so it acts like a semaphore and queues up extra
+# commands to be run if currently at the limit. Multiple "limiters"
+# may be configured which give inboxes different channels to
+# operate in. This can be useful to ensure smaller inboxes can
+# be cloned while cloning of large inboxes is maxed out.
+#
+# This does not depend on PublicInbox::DS or any other external
+# scheduling mechanism, you just need to call start() and finish()
+# appropriately. However, public-inbox-httpd (which uses PublicInbox::DS)
+# will be able to schedule this based on readability of stdout from
+# the spawned process. See GitHTTPBackend.pm and SolverGit.pm for
+# usage examples. It does not depend on any form of threading.
+#
+# This is useful for scheduling CGI execution of both long-lived
+# git-http-backend(1) process (for "git clone") as well as short-lived
+# processes such as git-apply(1).
+