add RFC notes for June 24, 2019

Author: joeyaiello

notes/20190610 - RFC Discussion Notes.txt renamed to notes/20190610 - RFC Discussion Notes.md

@@ -0,0 +1,38 @@
+# RFC Meeting - June 24, 2019
+
+## Attendees
+
+No quorum today...
+
+* Jim Truher
+* Dongbo Wang
+* Joey Aiello
+
+## Notes
+
+* PSGet 3.0 RFC has been updated, need to review with Steve when he's back
+* Closed PR #138 on the basis that it misunderstood the current process
+* Cursorily reviewed and briefly discussed PR 202 for side-by-side `pwsh` support
+    * Initial reaction is that this is explicitly a non-goal
+    * We are primarily trying to follow a fix-forward approach to `pwsh`
+    * Complexity of managing multiple namespaces in package managers is very high
+        * Also discouraged by distros
+* Process on aging PRs out?
+    * `Waiting on Author` for some period of time?
+    * Open question: do we give it a number and reject it, or do we just close the PR?
+        * The former provides more visibility to others who may want to pick it up
+        * PR after would move it to 2-Draft-Accepted with the existing number
+    * Probably won't use automation to remove `Waiting on Author`
+        * Only the author's response would change it
+* Still haven't codified the way that experimental features reduce the burden to RFCs
+* Jim: "code speaks louder than words"
+    * Jim did ulimit/umask implementation before he submitted the PR
+    * At a bare minimum, input/output examples are necessary
+    * Dongbo: "if you're really going to implement it, you need to think about all the details up front"
+    * We don't want to think of the implementation plan, but having a design is critical
+    * One great example is Test-Connection changes (#172)
+        * Only 121 lines, and it gives everything you need to know about input, output, parameter sets, etc.
+* We may need to be more direct about possible breakage that RFCs cause
+* Reviewed login shell RFC (PR #186)
+    * Pending performance measurements of Rob's new method, we will take his method of `pwsh -l` / `pwsh -LoginShell` instead of using `pwsh-login`
+    * Open question: on a container, does Docker implicitly start the pwsh entrypoint from sh?