[mythtv] A need for advanced player settings
f-myth-users at media.mit.edu
f-myth-users at media.mit.edu
Tue May 31 21:41:17 UTC 2011
> Date: Tue, 31 May 2011 15:26:56 -0400
> From: Kyle Rose <krose at krose.org>
> Which is precisely why the kernel folks came up with tainting: so they
> could easily identify users took some action that was unsupported. A
> similar system combined with a utility to easily gather information
> about a user's configuration would probably work here.
Yes. I suggested exactly this some months/years ago over some similar
issue of whether Myth should allow some bit of configurability. I may
even have mentioned, for example, the way that VAX/VMS did tuning: it
typically used a table roughly like
NAME MINIMUM DEFAULT CURRENT MAXIMUM UNITS
and hence figuring out what you changed from the defaults was easy. **
If "internal" variables existed, then they could have an additional
flag saying that they were, in fact, internal, and taintedness could
be computed by looking for variables with that flag which were not at
their default values. If that information was only used when emitting
a version string, users would have little inclination to manually hide
the taintedness by (say) changing the default or the flag. (And if
they did hide it and then seek help, I'll bet discovery of that would,
uh, cause those wishing to support them to decline to help further. :)
** It also led to such gems as
TIMEPROMPTWAIT 0 64 64 1000000 MICROFORTNIGHT
which I recall tripping over running SYSGEN one day. Since I know how
many seconds there are in a day and in a year, it took only a couple
seconds to realize that was a value of around a minute... :) [It took
longer to look up the documentation for the variable, in those days of
paper manuals---how long the system would wait for the operator to
type in the correct date if it booted and didn't think the stored date
was reasonable, before it just continued booting anyway.]
More information about the mythtv-dev
mailing list