[mythtv-users] [RFC] Proposed settings rework

Allan Stirling dibblah.allan.stirling at googlemail.com
Wed Nov 4 00:21:21 UTC 2009


Robert McNamara wrote:
> <wise words>
> 
<snip>

I would like to comment further on this, since it is easy to 
misunderstand the intent with these changes.

The primary goal here is not just freeing up the 12 or so 
pages of confusing settings that the end-user is presented 
with for each "configuration" section.

Each setting has behind it thousands of lines of code - A 
majority of which is run by only a few individuals. If a 
setting is no longer exposed in the "default" 
user-interface, that code still exists, but is used by even 
fewer people. It is also not necessarily as well maintained 
as the rest of the commonly used code.

The repercussion of this is that with the Myth codebase as 
it is currently, it's essentially a crapshoot if a developer 
can reproduce an end-user's issues. Add to that that it's 
impossible to fully test a change given the time available.

More settings means more code complexity. More code 
complexity means more bugs.

All of this adds up to the fact that just punting the 
settings off to an "advanced" section, or (shudder) a 
database-only setting is not actually helping bring Myth 
forward in either usability or reliability metrics.

Cheers,

Allan.


More information about the mythtv-users mailing list