Open GoogleCodeExporter opened 9 years ago
(De-assigning Mike for the time being until he has time for this; he's
currently working on the 'share exploration' functionality.)
Mike, just to give context about where this came from: we're about to finish
building the redesigned learner view, so Amit was doing an audit of the various
interactions and trying to make them look really nice!
Btw, just to record a thought I had about "play target sequence" -- I think
that, at some point, we should actually remove this from the music widget, and
make it into a separate noninteractive "play audio" widget, since this
functionality really belongs in the 'content' part of the state and isn't
really part of the interaction itself.
Also, +Xinyu for info, since she's currently looking into creating some
music-based explorations.
Original comment by s...@seanlip.org
on 31 Jan 2015 at 4:17
Good point, thanks for de-assigning.
Re: the non-interactive "play audio" widget, I realized that this might be
inconvenient for the authors, since you will then need to create the target
audio sequence twice: once when you're configuring the music widget and
determining what the correct answer is, and once when you're adding the
noninteractive play audio widget to the state content.
Original comment by amitdeut...@google.com
on 2 Feb 2015 at 4:21
I'd love to get Michael's/Xinyu's thoughts on this too, but I think splitting
off the "play expected audio" part would make the interaction more flexible, if
anything. If we do that then it opens up possibilities like "play the bass line
of this audio clip" or "identify the first two notes of this tune". Right now
it's a bit like having a TextInput interaction that can be used only for
questions of the form "Type in the string X".
Original comment by s...@google.com
on 2 Feb 2015 at 4:38
I definitely like the idea of splitting the play functionality off into a
seperate widget, as it would surely increase the possibilities for learner
interactions. I am concerned that, in this context, a "Play audio" widget might
not necessarily work with the music widget becuase there isn't an audio file
being played directly but actually a series of midi notes triggering audio
files. Surely there are ways around this, just thought it was worth mentioning.
Ultimately, I think the resulting functionality and extensibility increase
would justify the refactoring it would take to make this "play nice".
Original comment by wagnerdm...@gmail.com
on 2 Feb 2015 at 5:18
All good points. Also, there might be music interactions beyond just 'recreate
the following target music sequence', so separating the music playing into a
separate noninteractive widget would free up the interactive music widget to be
more flexible.
Original comment by amitdeut...@google.com
on 2 Feb 2015 at 5:25
I agree with Sean and Mike. There are many cases that require both a
demonstration and an answer with the music widget (and often they will contain
different notes). It seems like a good idea to put the 'demonstration' part in
the 'content' part of the state rather than the 'answer' part.
A question: by "play audio" widget, are we referring to something that plays an
audio clip, or something like the current music widget, which can play a
sequence of notes that are displayed on a score to the learner (and highlight
parts that are being played etc)? (Or maybe both?)
Original comment by wxy.xi...@gmail.com
on 2 Feb 2015 at 5:26
OK, it sounds like the consensus is towards splitting.
To Xinyu's last question -- that's a good distinction. I think we are really
describing two RTE extensions. One of them simply plays an audio file (and this
seems to be a common request: see
https://code.google.com/p/oppia/issues/detail?id=159 ). The other plays a
sequence of MIDI notes, perhaps combined with an associated interactive score
display.
Original comment by s...@seanlip.org
on 2 Feb 2015 at 6:36
I've filed issue 568; this and issue 159 are the issues for tracking the
subsidiary RTE extensions. So, the scope of this issue (566) is now for
improving the design of the current music interaction.
Thanks very much for the quick feedback!
Original comment by s...@seanlip.org
on 2 Feb 2015 at 8:14
Original issue reported on code.google.com by
amitdeut...@google.com
on 31 Jan 2015 at 3:13Attachments: