This is no where near finished but here's some first bits of startup notifications in #phosh (i'm usually using #epiphany on the #librem5 but I needed something that takes longer to start up so i went for #firefox):

@agx @agx Splash screens aren't very useful as they don't communicate anything to the user other than "something happened". We realized this is bad UX in web design decades ago. If it can't be instant, tell the user how long it will take. #ux #uxdesign

@adam @agx Speaking from experience, communicating that *something* is happening is better than not providing visual feedback to the user's action. This is a good step in the right direction.

@okennedy @agx Yes, that's one step up from "not responding" or frozen. Which is the best user experience of the following:

- Your spouse leaves without saying anything to you?
- Your spouse puts a orange circular icon on the door and then leaves?
- Your spouse says, "I'm going out to get milk" and then leaves?
- Your spouse says, "I'm going out to get milk and I'll be back in 20 minutes.", leaves, and comes back in 20 minutes?

@adam @agx I completely agree that there are more opportunities for improvement here, and if that was your entire argument then I withdraw my objection. Your original post had a very disparaging tone that made it seem like you felt that this was moving things in the wrong direction. (It seems to me like a splash screen would be a prerequisite for a progress bar)

@okennedy @adam exactly! The whole point is to demo that we detect the app launch correctly, come up with some animation and have a rough idea when to close that thing (which is the most complicated bit). Spinners, nicer graphics, etc are yet to come. It's all divide and conquer:separate UI representation from the lower level plumbing so design and plumbing can progress.