Start with a Website, Not a Mobile App • Hacker News Board

Re: Start with a Website, Not a Mobile App

#131

Post by reificator » Fri, Jan 04, 2019, 5:56 PM UTC

jklinger410 wrote: post #56

reificator wrote: post #40

Earlier quoted context omitted.

> >you're going to be hard pressed to get people to install your app just to try it out > In fact, the statistics prove[1] that this is false. Most app installs are to try it out, and promptly delete it. Most users will delete your app shortly after installing it. I'm really not following how this disproves what was being said. They were saying it's difficult to get people to even install for a test run, and you're s…

>They were saying it's difficult to get people to even install for a test run How is it difficult to get someone to install something for a test run when like 90% of app installs are test runs? So you're saying it's difficult to get someone to install your app at all, and that has nothing to do with your product or your marketing, but it has something to do with the technology? And the PWA experience of pinning thing…

> How is it difficult to get someone to install something for a test run when like 90% of app installs are test runs?

As I said in my previous post I'm clearly missing something here. Why are those mutually exclusive at all?

You're saying that most installations are test runs. That implies people are leery about leaving an app they don't want installed. Why does that mean that it's somehow easy to get users to install in the first place? It sounds like they don't want a bunch of apps installed.

> So you're saying it's difficult to get someone to install your app at all, and that has nothing to do with your product or your marketing, but it has something to do with the technology?

No? When did I say your product or marketing were unrelated to install rate?

> And the PWA experience of pinning things to your home screen, which we have absolutely no reason to believe is widely done, is a better option?

When did I make that claim?

> I'd like to see you try and support OPs claim rather than argue about my comment.

Why does disagreeing with you mean I support the OP? Whether or not I support PWAs, native applications, both, neither, or even "native" wrappers around a PWA is irrelevant.

My stance is that your statistics don't disprove the claim made, and to then claim that the OP is arguing from a point of ignorance is not okay.

Re: Start with a Website, Not a Mobile App

#132

Post by scarface74 » Fri, Jan 04, 2019, 5:56 PM UTC

jdietrich wrote: post #121

scarface74 wrote: post #97

Earlier quoted context omitted.

Given the fact that most people are already maxed out on apps on their device with just things like facebook and youtube, you're going to be hard pressed to get people to install your app just to try it out. https://www.apple.com/newsroom/2019/01/app-store-caps-record...

The list of top grossing apps paints a rather different and rather unflattering picture of the app store economy. https://www.apple.com/itunes/charts/top-grossing-apps/

I agree completely. What’s surprising is that Spotify is still a top grossing app even though they haven’t allowed new in app subscriptions for a year or two.

I guess those are legacy subscriptions.

But on the other hand, that also goes against the narrative that PWA’s will replace native apps anytime soon. Most of the apps on the list:

1. Require some type of DRM (streaming media)

2. Take advantage of in app consumables like games where the immediacy of being able to capture the whales through in app purchases make sense.

3. Need the performance of native apps.

Re: Start with a Website, Not a Mobile App

#133

Post by Ajedi32 » Fri, Jan 04, 2019, 5:56 PM UTC

timdavila wrote: post #54

plexicle wrote: post #50

Earlier quoted context omitted.

Chrome on Android will show a banner to add it to your home screen.

True, the situation on Android is much better than iOS. Also after you do this, you get an icon in your app drawer and not just the homescreen.

It also shows up as an app in the system menu, with natively controlled permissions, notifications, battery restrictions, etc.

Re: Start with a Website, Not a Mobile App

#134

Post by reificator » Fri, Jan 04, 2019, 5:59 PM UTC

nytopop wrote: post #120

jklinger410 wrote: post #56

Earlier quoted context omitted.

>They were saying it's difficult to get people to even install for a test run How is it difficult to get someone to install something for a test run when like 90% of app installs are test runs? So you're saying it's difficult to get someone to install your app at all, and that has nothing to do with your product or your marketing, but it has something to do with the technology? And the PWA experience of pinning thing…

> How is it difficult to get someone to install something for a test run when like 90% of app installs are test runs? The link you shared does not provide any statistics about the amount of installs relative to didn't-installs. It only deals with the set of people who have already installed, which says nothing to prove or disprove the point in contention (that increasing the set of people who have installed at all is…

Ah, if I had seen you post this before I did, I wouldn't have bothered to post my own. Thank you for summing it up so succinctly and respectfully.

Re: Start with a Website, Not a Mobile App

#135

Post by johnday » Fri, Jan 04, 2019, 5:59 PM UTC

wvenable wrote: post #119

johnday wrote: post #114

Earlier quoted context omitted.

It depends on your speed, concurrent user capacity, and uptime requirements (among other things). I run two production servers off two separate $5/mo VPSs, one getting a few thousand hits a day. No issues whatsoever.

What happens if something catastrophic occurs?

What do you mean by catastrophic? It's safe from the obvious things:

* DB corruption / destruction: The database is backed up.

* FS corruption / destruction: The production code is backed up with version control (obviously) and can be restored at the drop of a hat.

* The server just randomly falls off the internet: It would take about 2 hours to get a new server up to the same point.

What kind of problems could arise, that wouldn't equally affect any other kind of server provision?

Re: Start with a Website, Not a Mobile App

#136

Post by rimliu » Fri, Jan 04, 2019, 6:02 PM UTC

icebraining wrote: post #70

0xCMP wrote: post #7

There is one side of me that wishes more things were Mobile Apps. On the other hand, it's hard to argue that worse is better here and that any founder should focus on Web unless it's not possible to do it on the web (yet). I'd much prefer a native app (or actually feels native; Well done RN is good enough for me), but the cost of creating and maintaining one is often too high when you don't know what will work. As a…

Is it a matter of UI? I find it interesting that the "web app with native UI" model never took on. There were a few tries (e.g. XUL), and now Expo seems to be a version of that for RN (though focused on testing), yet for various - seemingly circumstantial - reasons it never became a fully fledged alternative to browser apps and installed apps.

Because it solves no problems. Unless you consider “we want web tech everyhere” to be a valid problem. “Cross-platform” development may look like a problem worth solving, but the dolution is a lie. If you use native UI you will have to deal with a native tech to a degree. Is (cost + pain)/benefits worth it—that’s up to you. But seeing how RN stuggled with the thing that’s trivial in native—navigation stack—I’d stay away. Also these bridged technologies will always lag behind and can be really painful after significant updates to UIKit.

Re: Start with a Website, Not a Mobile App

#137

Post by ecp9 » Fri, Jan 04, 2019, 6:03 PM UTC

I don't even bother installing apps anymore, I don't have time to research whether or not they are going to abuse my privacy or have some horrendous TOS. I got my core 10 apps and haven't bought a new one in a year.

Re: Start with a Website, Not a Mobile App

#138

Post by tracker1 » Fri, Jan 04, 2019, 6:03 PM UTC

BucketSort wrote: post #118

shams93 wrote: post #2

Given the fact that most people are already maxed out on apps on their device with just things like facebook and youtube, you're going to be hard pressed to get people to install your app just to try it out. With a progressive web app if they like your app they can add it to their device home screen without installing via the app store and without using up space on their device. Also don't need to pay Apple $100 a ye…

Can you provide data to substantiate your claims? I'd truly like to know if this is true, because my intuition and experience tell me the opposite. People like downloading and trying out new apps. I also disagree with the premise of this post. Many people exclusively use their phones and would never utilize a web version of an app.

I think it depends... it's probably bias to more technical users vs moms, dads and kids generally. I know I'm far less inclined to use a native app for something new. I also hate when I can't do updates from my desktop. I use linux, windows and mac regularly.

Of course via cordova, it's easy enough to create a thin wrapper and add a few bits for notifications and better capture integration (if you need photos/video, etc). Since I'm mostly doing react these days, I'd consider react-native a natural next step.

Re: Start with a Website, Not a Mobile App

#139

Post by tracker1 » Fri, Jan 04, 2019, 6:04 PM UTC

scarface74 wrote: post #3

shams93 wrote: post #2

Given the fact that most people are already maxed out on apps on their device with just things like facebook and youtube, you're going to be hard pressed to get people to install your app just to try it out. With a progressive web app if they like your app they can add it to their device home screen without installing via the app store and without using up space on their device. Also don't need to pay Apple $100 a ye…

On iOS you won’t get notifications or be able to run anything in the background. But if you are trying to start a business and the $100 a year you would have to pay to distribute your app will make or break you, you have bigger issues.

With cordova you can create a thin wrapper that can add those features you need, starting with a PWA and extending.

Re: Start with a Website, Not a Mobile App

#140

Post by phs2501 » Fri, Jan 04, 2019, 6:04 PM UTC

endorphone wrote: post #34

shams93 wrote: post #2

Given the fact that most people are already maxed out on apps on their device with just things like facebook and youtube, you're going to be hard pressed to get people to install your app just to try it out. With a progressive web app if they like your app they can add it to their device home screen without installing via the app store and without using up space on their device. Also don't need to pay Apple $100 a ye…

We've been hearing the same pitch for literally years. Heck, I've made some of them myself: "This is it, people! Native apps have been rendered unnecessary!" No PWA competes with a native experience. Not performance wise, not usability wise, and ultimately for the developer not even development wise. It may still make a lot of sense, and there are a lot of arguments for web apps, but the enthusiasm in this discussion…

> Indeed, right now we're seeing a big uptick in Instant Apps on Android -- go to a webpage and it actually loads a native app -- and I fully expect the same to appear on iOS.

I'd never heard of this but I can guarantee I will immediately quit and never install an app that gets forced upon me via this method. This is like the terrible "Try our app!" web popups, only 10x worse.

I realize like most people on HN I'm hardly the norm when it comes to mobile users, but UGH.