@flyboarder that looks interesting, but inputs have an internal state no matter what you do

it's just part of the browser

so you have two (non mutually exclusive) options:

- update your state based on a model of how an input works

- force the input internal state to represent your state any time you expect it will get out of sync

@thedavidmeister I'm not sure what you mean, does the pattern above not work for your inputs?

@flyboarder you haven't used any inputs in the pattern?

@flyboarder i wasn't having any issues with divs or regular attributes as you can treat them as "write only"

the problem i'm having is that inputs are read/write by nature, on their value attribute specifically

@thedavidmeister I am using the value as the child in the example but that could be an input value attribute instead, the difference is the cell graph

@flyboarder would you mind tweaking it a little bit to use an input value?

i just want to be 100% sure to compare apples to apples

in my mind the issue isn't anywhere in the cell graph except the very final do-watch that triggers the do! callback

@flyboarder also i want to take your example and drop it into the test that i already wrote and see what happens 🙂

i don't want the tests to pass/fail just because i didn't quite understand what your solution is doing

@thedavidmeister just heading to bed, I'll update the example in the morning with inputs

@flyboarder no worries, i'm just glad to be getting some help/attention 😉

again you can use the hoplon.sorting cells from brew to make this smaller, i just expanded each step

@alandipert would javelin be open to spec macro improvements ?

I often run into uninformative javelin errors

@flyboarder definitely