Bokeh Team,
Great work on the new release. Among other things, the documentation and front page look a lot smoother which is a big improvement for users!
I am curious what you guys would suggest concerning embedding bokeh plots in a larger webapp. I’ve been building a data display system inside of a flask app, which works pretty well. I’m pretty used to working in flask and am happy with embedding the plots using the components method. I am in the process of building out a larger backbone interface and would strongly prefer to keep flask as the backend because its what I’m used to. How much am I going to miss out as the tornado server wrapper develops in bokeh? At the moment, I am happy to write my own javascript, but at some point it would be nice to use the bokeh server features as well. An example is dynamically loading large datasets. I suppose I could use both flask and tornado but that would get ugly pretty fast?
One idea I had is to embed audio as part of the bokeh js to make interactive sonfication of the data from the plots. You can check it out looking at ocean “spectrogram” under height plot: swellspect.com. A flask endpoint passes the audio data as base64 so can have “playByteArray(bouy_music[‘frame_wfs’][music_frame_idx]);” directly in the javascript.
Id be happy to make a open source demonstration of this with vocal communication sounds to demonstrate bokeh sonifcation.
Jeff