Comment by ryanwaggoner | Hacker News Reader

Comment by ryanwaggoner | Hacker News Reader

Parent Full thread

Not that you need my vote, but I agree with all of this. In particular:

- Go true native, not hybrid

- Use ads for testing ideas

- Don't forget about email!

I'm also intrigued by your approach of not going market-by-market like a lot of local-focused marketplaces do. Your approach seems to make sense, and obviously worked for you. I particularly like that you had a different onboarding flow for users in new markets, that's smart.

But still, I'm not sure that this approach would make sense for a lot of local-focused apps? I think the fear would be that it would be relatively easy to get 100k users but they'd be spread out so thin that they'd find little utility in the app and stop using it. So you'd have 100k users in your database, but a tiny fraction of that in terms of actual active users. But maybe that doesn't even matter, since you wouldn't have had all those inactive users anyway?

Could you speak to that? I'd love to hear how many of those 100k users are DAU or MAU, but understand if you can't share that :)

Congrats!

Sure, of course! So this is one thing that might be unique to Winnie — local data is one of our value props (and the best one) but it's not the only one. We discovered pretty early on that even in the local communities people were asking not-local questions like "what's the best diaper?". This told us that there was some demand to be part of a parenting community, even if it wasn't giving you particularly local insights. This is one feature of Winnie that works anywhere in the world, and we do have an active cohort of users who just use us to talk and get advice from other parents.

We also did something that I forgot to mention in the article that helped us grow nationally before we had a lot of proprietary data. One of the nice things about starting a company in 2017 is that there are tons of great resources available to you. Free or cheap services that solve what used to be really hard problems are readily available. One such service is Foursquare. When we launched Winnie, if you opened the app in an area where we didn't have data, instead of showing you nothing we instead showed you results from Foursquare. This was admittedly not the best experience, but it gave people affordance to still find places and write reviews.

Refusing to go market-by-market also forced us to build a bunch of proprietary and very cool infrastructure that collects data at scale. One early system we built could actually figure out which restaurants had changing tables and highchairs, nationally and instantaneously, at a VERY low cost. I can't say how we did that but you'd be surprised at what's possible if you have the will and ambition :)

Go true native, not hybrid

Users don't care. Most of them can't tell.

The OP wasn't talking about users, but rather best practices on how to get the App Store makers to feature you. I've heard the App Store editorial team specifically give the "go native" advice.

It's definitely not required, it just helps. Our app got featured in one of the App Store's daily stories despite it being written in React Native.

I kind of doubt that whoever does the feature pics looks at the binaries to figure out if it is native or not. As long as the end result is not easily distinguishable from native, I'm sure you are just as likely to be featured.

Yea if you play around you can really make an app look like native. Even in desktop it's difficult to figure out web apps wrapped under electron when they are done right..

> Even in desktop it's difficult to figure out web apps wrapped under electron when they are done right..

Name one.

VSCode is close if you squint or don't know where to look for the evidence of Electron.

Its UI paradigm looks out of place on macOS–it very much looks like a website in an app wrapper.

For you, but not for most people I’d guess. Or most people don’t care enough and that’s why VSCode is hugely popular.

"Its UI paradigm looks out of place on macOS "- Which part?

Sorry, didn't see the reply. You want a web app as close to native? Look no further - https://www.nylas.com/download, source https://github.com/nylas/nylas-mail.

It's close, but not quite there. I don't see vibrancy or features from standard text controls.

Telegram desktop

Telegram is a native app: https://github.com/overtake/TelegramSwift

I'm an Android developer and a few times in 2017 I confused a Corodva app with a native app. With Chrome 63 and customized over-scroll behavior Twitter PWA is really impressive, no reason to install native Twitter app. 2018 might be breakthrough year for hybrid and PWA apps.

I'd say there's no point just framing your site in a native container, but it's still viable to use webviews where it makes sense. Particularly if you're just delivering loads of static content.

And you're the reason I don't use very many phone apps.

Developer apathy becomes user apathy. If you make great apps, your customers will be enthusiastic about using them. If you make mediocre apps, you will attract mediocre users, or no users at all.

You are most likely using some hybrid apps and you don't even know it. When done right, you pretty much can't tell.

And when done wrong...

Which can be said of native apps as well

For a very long time, and it may still be true, both Uber and Instagram were hybrids. While you may not use them a lot of other people certainly aren't apathetic about those app. Facebook was a hybrid app for years although it's a native app now.

Mediocre apps are annoying, I agree, but that rarely has anything to do with the underlying technology.

> - Go true native, not hybrid

What about for games? Unity is extremely tempting, especially when the alternative is to write the game twice in two different languages...

I don't think there's native for games. You're going to use a game engine and almost none of the OS UX conventions, so go with Unity. I don't think there are any developers that don't.

I'd say Unity is native. Arguably so are JavaScript runtimes like react-native or nativescript. She's probably just referring to the poor experience of hybrid web apps.

I'd honestly just lean in on PWAs if you're a content/SEO focussed business.

Exactly, Unity and Unreal makes a native and highly optimized graphics engine for every platform they support. The content you create lives on top of a native app.

Games are the exception to this rule. Apple and Google have no problem featuring games built in Unity as long as they are performant and polished.

Isn't this true for all apps? If an app is performant and polished, while also following UI/UX guidelines, it has just as good of a chance to be featured if is React vs Native? Is there anything that goes against this? Or is the idea its much harder to follow UI/UX guides while using a hybrid system?

I suspect that the reviewers see LOTS of low-quality apps written on portable frameworks every day. These are accessible and cheap options for doing quick development, and unfortunately that means that lots of junk gets churned out on them, and some of it even looks pretty decent since there are so many free UI frameworks. There are some platforms that are almost as easy and cheap to use as wordpress.

A company that does a respectable job of developing a nice-looking and performant native-application is going to stand very far apart from this crowd. The same company may have done just as well in terms of performance and UI conformity with React Native but will they stand out as well to tired and jaded reviewers? I think that is what the OP is getting at. No one can predict who will be featured, but true native is probably one thing that helps.

Reply on news.ycombinator.com