naming conventions

[brokkr]

• we refer to the things in update_q as 'subdata' bu they are instances of 'subupdate'
• output functions should be named in clear association to their error level/logger. also we should institute a top-down naming conventions so that place comes before specificic purpose (i.e. fail_dl over dl_fail)

[brokkr]

Point 2 was cleared in the output overhaul. As for the rest... There are quirks and non-consitencies. Reanming stuff at the the 11th hour before finalizing just seems to be a bad idea. If I come back to it in five months time what naming schemes am I likely to remember: The one that has persisted for years or the one made up for consistency's sake? Subdata makes sense. It's a packet of data about a sub. And subupdate makes sense because it's parallel to subupgrade. Unless I come across worse offenders I'm thinking wontfix.