From: Vaughan Johnson <vaughan@au...> - 2010-10-29 22:44:42
Thanks, Roger. Overall, I think we need a more sophisticated system for
enabling/disabling. For example, I think BassBoost should be disabled
for non-Wave track.
That's obviously a *lot* of work. And I think should be designed in
But as you say, loosening it up is the easiest way to get the positive
side you mention. Am cc-ing audacity-quality, because there may be
objection to it there.
On 10/24/2010 7:20 AM, Roger Dannenberg wrote:
> This discussion was growing so long it was hard to find the inserted
> comments/replies, so I left most of it out (let me know if you'd prefer
> to keep the whole discussion intact).
> A simple thing to do for now would be to change the default effect flag
> that requires a Wave track selection to just require any track
> selection. Most effects are no-ops on non-wave tracks, but effects that
> stretch time will operate on all tracks. The down side is that you will
> be able to select an effect like BassBoost... on a non-Wave track, and
> no change will result (but that shouldn't be too surprising -- what
> would you expect and why would you try it?) The positive side is that we
> make almost no change to existing flags code, and Change Pitch...,
> Speed..., and Tempo... will work on Label and Note tracks.
> The SelectAllOnNone logic would still automatically select everything if
> there is NO selection of ANYTHING.
> On 10/22/2010 3:13 PM, Vaughan Johnson wrote:
>>> [Al, I think]:
>>>> This is especially important because of the "select all on none"
>>>> feature. I think it's a really bad feature, but because it's there,
>>>> there's a big difference between doing nothing to the selection and
>>>> having nothing selected.
>> Not sure I agree it's a bad feature, but it was definitely not designed
>> in, and was hacked in post-facto, so we're still finding repercussions.
>> That's true, too, of sync-lock as it's expanding into other track types.
>> It was not in the original design (nor even were label tracks, iirc),
>> and its original intention was strictly for syncing wave and label
>> tracks, so it's still causing unforeseen repercussions.
>> All that, to me, speaks to doing as little re-architecting of it all as
>> possible at this point, in order to not introduce more uncertainty and
>> hackage. Top priority is getting 2.0 completed and released. So let's
>> discuss what minimal changes can be made to get this to some sensible
>> (probably incomplete) state.