+ // LATER: refine this semaphore process
+ // to make a robust signalling mechanism:
+ // app1 requests, app1..n do something (or don't - they may be dead),
+ // app1 realises all apps have done their thing, it then continues with synchronized data.
+ // this probably needs two files - a request file,
+ // and a response file which is acknowledged by the app1 requestor for each app.
+ // eventually, no more responses are received for the request, and the app can then only continue with its store.