jalencas

It may seem frustrating at first. But when tv networks (and not just them, but also newspapers, big brands, podcast networks…) switch from having 'channels' to having 'apps', you'll see the genius behind it, and why it's worth the developer time. That's what I'm looking forward to.

Discoverability is terrible. For this specific app, it doesn't matter as much, since the majority of our users will already know our service, so we can tell them how to look for the app (which is in itself bad UX). For potential new users, they need to learn about the app somehow, so we need to make an extra marketing effort.

For apps not tied to an existing web property (I'm planning to do some games on my own, for example), you need an effective marketing channel, even if it's just word of mouth. No one is going to just come up with some search terms and find your app.

These kinds of apps would be appropriate for the platform, but you have to take into account the input from the user. Training material would need to be navigated with a remote (or a game controller) and for a video conf. app, you'd need a separate camera (so far, the USB port is not accessible, so it would need to be a bluetooth/wifi camera, most likely an iPhone)

Our case was the other way around, Apple TV first, and we are currently working on the iOS app. We have a previous iOS app, but it only knows about alerts. A full dashboard monitoring app was on our roadmap, further down the line, but with the release of the Apple TV (and having the luck of being selected for a developer kit) we decided to tackle the tvOS version first.

This decision was made easier due to the fact that both platforms can run the same code, so most of the API interaction and view rendering code will be reused. It was mostly a greenfield project, because even though we leveraged the knowledge from our alerts iOS app, we took the opportunity of writing everything in Swift, which functional expressiveness makes it slightly more useful when it comes to dealing with APIs.

Hi, I'm the main author of the app in the linked article. (Finally had to create an HN account after lurking for years.)

I'll be glad to answer questions regarding development, audience and similar aspects, but I'll do that tomorrow as it's just past midnight in Europe/UK (where we are based). Also feel free to ask about our monitoring service, I'm sure my colleagues will chime in where my knowledge ends.