if ( not _Connect-device matches "*batch*" ) and _connect-batch <> "yes" then tt_Dashboard.con_self = tt_Dashboard.con_self + 1.
So "self" is counted for non-batch self-service connections. It (mostly) equates to local interactive connections.
Somehow I suspect that you are going to suggest that that might be sub-optimal 
It took longer than I expected (that pesky day job keeps getting in the way...) but ProTop 3.141 is now available to download!
Pop on over to https://wss.com/progress-openedge-monitoring-with-protop/#download or your favorite dashboard and grab a fresh copy of the ProTop client.
Highlights that I am willing to confess to include:
Updated db startup parameter discovery to accomodate the purging of _Startup and _Statbase with oe12.
Fixed a problem with etime exceeding the size of an integer in lib/zippy.p
Fixed a problem with non-default -basetable and -baseindex resulting in incorrect table and index stats in releases prior to oe11.7
Added logRd to logWr ratio to dashboard, reordered the left most column a bit.
Slightly reordered the right most dashboard column (connections).
Added a new data collector for Pro2. The "2" command will display data about Pro2 and adding "Pro2Activity" to pt3agent.cfg will send that data to the portal for trending purposes.
Added LOB activity (LOB data is not available prior to OE12)
Re-mapped some keys. Type ? to see the new assignments on the help screen.
Fixed user table and index activity issues when non-default base and range are in use on the "U" screen.
Added new fields for user locks, lock HWM, TRX activity, -Bp usage and more to "U" (user info). Also re-aligned screen.
Removed unnecessary internal temp-table indexes to improve performance.
Fixed an issue with "users of" tables or indexes inappropriately remembering the history of tables and indexes being examined.
Fixed an issue with non-existent "use an existing custId" program flow at install time.
Fixed some problems with OS Info due to off-brand shells spewing non-standard output for "ulimit -a".
Cleaned up some issues with user names and "flags" on the uio screen
Fixed a problem with settings via the "%" command needing two iterations before they would "stick".
Fixed a Windows localization issue with options to WMIC when getting "df" data.
The "unknown" data comes from the osInfo data collector. You might need to add that to etc/pt3agent.cfg
AI, BI and APWs are combined now into a single "async writers" section.
Looks like it dates back to pt3.3n.
It is fixed now.
That's kind of dumb. Who the heck coded that?
That is an interesting idea. A more complex rules engine for the alert processing is something that I have been thinking about for a while. On the other hand there is a lot to be said for keeping it simple.
Regarding the specific alert: there is a lkTblPct field which I use for a similar purpose. Rather than the HWM it reflects the locks in use at that moment. In some ways that is better -- the HWM sticks around until you restart and if something went bump in the night is going to keep alerting once you hit the threshold. Whereas lkTblPct will only alert if something is actively causing a problem.
Nope. You're right -- that's a bit more difficult to calculate and it seems much less likely.
But I'll add it to the "think about it" list 
The flakiness is being addressed in portal v4.
The "***" is indicating that there is an unusual relationship between table reads and the indexes for that table.
Speaking very broadly, usually you expect to see a single index that has more or less the same amount of activity as the table. If we see index reads at more than 1.5x table reads that is strange enough that we put a "***" there to draw your attention to it.
Portal v4 has a lot of enhancements to stuff like this. Real Soon Now.