I’m curious as to how the graphical aspects of the Mark II are going to work. By now we’ve all seen the images, and watched the videos. There’s some very cool examples of Mycroft helping with cooking and setting timers.
How are those kinds of features going to work with Picroft units (assuming there will still be the possibility of Picrofts after the Mark II is in production)? How are those features going to work with desktop installations?
In a perfect world, we could add a display to our Picrofts with a custom printed enclosure and use the existing graphics there, or have desktop widgets for our PC based installations.
Is there an API available (planned to be available) for those kinds of interactions, or will each graphical display need to be part of the skill development?
I guess I’m just curious as to how that’s going to work and what will be available to modders.
What I want to know, is will we be able to load our own content… For example, I’d love to have a custom face - like say, Hatsune Miku whilst I interact with Mycroft!
It should be interesting to see what people come up with… And if no one makes a Miku skill, well it’ll be an excuse for me to spend a weekend learning what’s involved.
Does this mean that each individual skill will responsible for animations as well as graphical elements, and will need to integrate their graphics with the standard Mycroft system graphics and animations?
Good question, my assumption - just an assumption at this stage - is that we will have some standard libraries to use as well (like thinking, smiling etc) so you don’t have to do all the graphical work.
Can you give any details on how the implementation will be done in the future / if its already planned or worked on? I´d also like to know if we will be able to address any generic screen, not just the one available on the mark 2. For me it would be great if I could extend the graphical interface to work with the touchscreen on my raspberry pi. If possible even make use of the touch features to interact with skills via some touch UI.
my 2 cents, dsplay should be like the audio service, a generic wrapper that talks to any number of backends, one of which would be the mark2
i personally would use remi as a backend, specially for the touchscreen use case
other options would be a simple opencv, open in browser, html rendering, default image viewer…
we could add a backend named “kitchen”, which is simply a old ass tablet stuck on android 4 that is always open in a webpage served by mycroft (remi ), render picture in this webpage and boom… display pics anywhere easily with almost no setup!
im literally waiting for you guys to finish this up!
