search

CNMAT / Music-and-Computing

Materials built for MUS158A, MUS158B (B is only io area of patchers)
Other
6 stars 3 forks source link

points.buffer x/y should be phase/value #54

Open ramagottfried opened 7 years ago

ramagottfried commented 6 years ago

@kulpajj what do you think about phase/value instead of x/y? does that make more sense?

equilet commented 6 years ago

I know I'm not being asked, but I personally like it.

ramagottfried commented 6 years ago

which one? (I think I know which one you mean but just want to make sure)

equilet commented 6 years ago

I vote for both /phase and /value instead of /x and /y

kulpajj commented 6 years ago

/time and /value?

meh. I dunno.

On Mon, Oct 16, 2017 at 11:03 AM Jon Kulpa kulpajj@gmail.com wrote:

The confusing thing about /phase is that in @normal_x 0, it IS /x, not /phase.

On Mon, Oct 16, 2017 at 11:02 AM Jeffrey Lubow notifications@github.com wrote:

I vote for both /phase and /value instead of /x and /y

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/CNMAT/Music-and-Computing/issues/54#issuecomment-336974136, or mute the thread https://github.com/notifications/unsubscribe-auth/AWhttXvS8nTphhEbHOePq90iRw4SNQzRks5ss5olgaJpZM4LjRgC .

ramagottfried commented 6 years ago

it's not time! : ) the timing depends on how fast the phase moves

kulpajj commented 6 years ago

I know I know I know. =)

Its /x and /y

On Mon, Oct 16, 2017 at 11:07 AM rama gottfried notifications@github.com wrote:

it's not time! : ) the timing depends on how fast the phase moves

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/CNMAT/Music-and-Computing/issues/54#issuecomment-336976701, or mute the thread https://github.com/notifications/unsubscribe-auth/AWhttcbgm1KmO9GvoY6_7HQzEW3L5Cihks5ss5tvgaJpZM4LjRgC .

kulpajj commented 6 years ago

The confusing thing about /phase is that in @normal_x 0, it IS /x, not /phase.

On Mon, Oct 16, 2017 at 11:02 AM Jeffrey Lubow notifications@github.com wrote:

I vote for both /phase and /value instead of /x and /y

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/CNMAT/Music-and-Computing/issues/54#issuecomment-336974136, or mute the thread https://github.com/notifications/unsubscribe-auth/AWhttXvS8nTphhEbHOePq90iRw4SNQzRks5ss5olgaJpZM4LjRgC .

ramagottfried commented 6 years ago

true... hmm. @equilet any thoughts on the @normal_x 0 issue? it's true that it's not "phase" in the traditional sense then.

kulpajj commented 6 years ago

I don't think anyone is confused by /x and /y

Its just a matter of getting to know the tools and the @normal_x modes.

On Mon, Oct 16, 2017 at 11:11 AM, rama gottfried notifications@github.com wrote:

true... hmm. @equilet https://github.com/equilet any thoughts on the @normal_x 0 issue? it's true that it's not "phase" in the traditional sense then.

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/CNMAT/Music-and-Computing/issues/54#issuecomment-336978213, or mute the thread https://github.com/notifications/unsubscribe-auth/AWhttR3kLXpef6NEShWTs3f3XPzFX7hRks5ss5xIgaJpZM4LjRgC .

equilet commented 6 years ago

What is 'normal x' in 'normal x mode'? I'm confused. ;)

ramagottfried commented 6 years ago

hmm, I sense some sarcasm there : )

by default @normal_x is on (1) meaning that the /x values are normalized 0-1, if @normal_x is off (0) then the values are not normalized, so you would look them up (or interpolate) between them by the original value.

just in case : )

equilet commented 6 years ago

I see. No, there was no sarcasm intended. I'll take some time to review this with @kulpajj and get back into the thread. Sorry if there was confusion introduced, I'm just trying to establish if the two use cases are different enough to come up with new terminology for the object, namespace, or both.

ramagottfried commented 6 years ago

it would be cool to have a name that makes it clearer that there is one value that is used as input for the lookup (/x) and anther that is the value that is output (/y).

thinking of it as points on an x/y plot makes some sense, but we don't at the moment have a way to lookup in the /y axis.

also x/y makes it seem like the points could be really 2D (i.e. jumping around in either axis), but actually the /x value should always be monotonically increasing so again it's not really truly plot.

thinking more about @normal_x 0, I think that is still ok to call it /phase since phase is really more about a point in cycle, or a transition between points. Normalization just helps with dealing with phase.

happy to hear your guys thoughts

kulpajj commented 6 years ago

/position /value?

On Mon, Oct 16, 2017 at 11:46 AM, rama gottfried notifications@github.com wrote:

it would be cool to have a name that makes it clearer that there is one value that is used as input for the lookup (/x) and anther that is the value that is output (/y).

thinking of it as points on an x/y plot makes some sense, but we don't at the moment have a way to lookup in the /y axis.

also x/y makes it seem like the points could be really 2D (i.e. jumping around in either axis), but actually the /x value should always be monotonically increasing so again it's not really truly plot.

thinking more about @normal_x 0, I think that is still ok to call it /phase since phase is really more about a point in cycle, or a transition between points. Normalization just helps with dealing with phase.

happy to hear your guys thoughts

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/CNMAT/Music-and-Computing/issues/54#issuecomment-336991849, or mute the thread https://github.com/notifications/unsubscribe-auth/AWhttVBNb9onPEsOrfNFK2mrvAQVL8n8ks5ss6SYgaJpZM4LjRgC .

kulpajj commented 6 years ago

I'm still partial to /x and /y. It's very easy to understand.

/phase and /value is ok maybe, but I think saying /phase is an index in a list isn't really true. It's /lookup-x-position, so it's /x.

On Mon, Oct 16, 2017 at 11:48 AM, Jon Kulpa kulpajj@gmail.com wrote:

/position /value?

On Mon, Oct 16, 2017 at 11:46 AM, rama gottfried <notifications@github.com

wrote:

it would be cool to have a name that makes it clearer that there is one value that is used as input for the lookup (/x) and anther that is the value that is output (/y).

thinking of it as points on an x/y plot makes some sense, but we don't at the moment have a way to lookup in the /y axis.

also x/y makes it seem like the points could be really 2D (i.e. jumping around in either axis), but actually the /x value should always be monotonically increasing so again it's not really truly plot.

thinking more about @normal_x 0, I think that is still ok to call it /phase since phase is really more about a point in cycle, or a transition between points. Normalization just helps with dealing with phase.

happy to hear your guys thoughts

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/CNMAT/Music-and-Computing/issues/54#issuecomment-336991849, or mute the thread https://github.com/notifications/unsubscribe-auth/AWhttVBNb9onPEsOrfNFK2mrvAQVL8n8ks5ss6SYgaJpZM4LjRgC .

ramagottfried commented 6 years ago

hmm, /position and /value is pretty good... @equilet any thoughts? /position is a lot of letters to type though... hmm, let's keep thinking about it... I have to run.

I like /x and /y too for the simplicity, but I think it's misleading for the reasons above...

what about:

/t and /value ?

t is often used as a variable in equations for interpolation which is what we're doing. https://en.wikipedia.org/wiki/Linear_interpolation#Programming_language_support

I guess x/y is like f(x) = y which is also what we're doing: https://en.wikipedia.org/wiki/Linear_interpolation#Linear_interpolation_between_two_known_points

ramagottfried commented 6 years ago

but there's the problem of needing to specify x as increasing values (or I forget now, maybe I sort them internally).. ok have to run!

kulpajj commented 6 years ago

So /y is value. Easy. /x is a somewhat arbitrary value, and what we care more about is its [[ index ]] is a lookup for the y value we want. Harder to communicate as a /name! =-) It seems like with this in mind, y could be /value and x could be /index. But that's pretty misleading when you take a step back and think it doesn't really capture the real goings-on. So what about /key and /value?

On Mon, Oct 16, 2017 at 11:58 AM rama gottfried notifications@github.com wrote:

but there's the problem of needing to specify x as increasing values (or I forget now, maybe I sort them internally).. ok have to run!

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/CNMAT/Music-and-Computing/issues/54#issuecomment-336996100, or mute the thread https://github.com/notifications/unsubscribe-auth/AWhttVYofVHMYgNawUiEC8bWTkfA3TOBks5ss6dJgaJpZM4LjRgC .

kulpajj commented 6 years ago

an alternative is staying with /x and /y but explicitly teaching what /x is doing as a lookup value that gets a list-position, i.e. index, that corresponds to a y value we want. And that could be a whole tab in an m158 module that would clarify a lot for students.

On Thu, Oct 19, 2017 at 8:26 AM, Jon Kulpa kulpajj@gmail.com wrote:

So /y is value. Easy. /x is a somewhat arbitrary value, and what we care more about is its [[ index ]] is a lookup for the y value we want. Harder to communicate as a /name! =-) It seems like with this in mind, y could be /value and x could be /index. But that's pretty misleading when you take a step back and think it doesn't really capture the real goings-on. So what about /key and /value?

On Mon, Oct 16, 2017 at 11:58 AM rama gottfried notifications@github.com wrote:

but there's the problem of needing to specify x as increasing values (or I forget now, maybe I sort them internally).. ok have to run!

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/CNMAT/Music-and-Computing/issues/54#issuecomment-336996100, or mute the thread https://github.com/notifications/unsubscribe-auth/AWhttVYofVHMYgNawUiEC8bWTkfA3TOBks5ss6dJgaJpZM4LjRgC .