There is a principle of user interface design which says you think you know how people will use your product but they will probably use it in ways you didn't expect. You have to figure out how to pave the cow trails (make it easier for them to use it the way they want to).
We're just 4 engineers and 2 designers for now so we have to focus on things that really count. One of the engineers is dedicated to an iOS app and it's pretty far along.
There are a few things that are core to Cake's success and it feels like focusing on getting them right is crucial. One of those is topics. When you started this conversation, you used it the way we expected: you discovered the most relevant topic for your post and didn't try to add topics where it doesn't belong to get more eyeballs. You didn't create a new topic at random that no one is following.
It feels to me that topics will be critical to scaling with good signal to noise. It's my expectation that if the torrent comes, people will select more and more fine-grained topics. And our role in helping users select the right topics feels critical.
As for following people, we are getting requests for that and I expected us to. One problem social graph companies face is if you follow someone, you get all of their interests. On Facebook that means you have to read about your uncle's political views when you follow him to keep up with the cousins. YouTube creators say they can't post anything in their channels except the topic their subscribers expect of them.
We're not working on following people at the moment because it's a big technical challenge and we have an opportunity to do it really well, which will take a lot of thought. My guess is what you really want is to not miss posts from interesting people within the topics you follow, but not their views in topics you don't care about. Am I right?
Jeff Bezos has this wonderful philosophy of focusing on what will never change. In his case it's selection, low price and fast delivery. In our case I think it's signal to noise and content discovery, no?
To that end, we introduced panel conversations and I can tell the world does not understand them until they experience them, but then they love them because more signal, less noise. And they give a voice to people who would otherwise never post because trolls. Here's one about Google+:
It feels that, like topics, panels are very important to us and we need to keep refining to make them work really well. Both those feel more important to me than a public API.
As for financing, eventually we will have to run ads, but with a big difference. For one thing, interest-based advertising is the most effective and we are organized around interests. We don't need surveillance capitalism like Facebook does to figure out where advertisers should place their ads. Also, our ads can be high fidelity in the stream like on Instagram. And I hope that we can do like YouTube Red and let you subscribe to see no ads. I'm a YouTube Red subscriber and I love it.
Let us know if you have more questions.