Hacker News Reader

Comment by doomrobo | Hacker News Reader

Personal experience as a web dev: the mobile Safari PWA experience (and even regular browser experience) is terrible. The browser is riddled with bugs and lacks support for features that have been around for a while. I apologzie for not having examples bc it was a few months ago that I was struggling with this, but hopefully someone else can fill some in. I honestly believe Apple's stranglehold on the mobile browsing experience has been one of the most disempowering things to mobile users in the last decade.

View on HN

Why is the answer to this question always “Safari sucks”?

Are there no good PWAs that run on Android? Are there none that, though perhaps smaller is scope, run well in Safari?

People have been advocating for PWAs for a very long time now but it’s incredibly hard to get examples of good ones. There is always some complaint about Safari instead.

Telegram is probably the best example of a PWA. It's not as good as the mobile app, especially for gestures, but it's better than a lot of native apps.

Telegram really is an outlier among messengers. Each version is developed by like one person, and they are all feature-rich, feature-complete, fast and smooth.

Apple's stronghold affects PWA feasibility on Android too. If I have to write a PWA for Android and a native app for iOS to cover gaps in Safari, why choose a PWA at all?

Apple may be ~50% of the market in the US but it’s ~20% of the market worldwide. On the desktop anyone on any mainstream OS can use Chrome or FF.

Yet someone the total lack of PWA success is 100% Safari’s fault.

Look at all those users I listed above. How are there no common/well known PWAs if they’re so great? People use websites like the Google Suite and Figma and such. They use web apps built as desktop apps in Electron. They use native apps.

Almost no one uses PWAs. Even if Figma or Gmail or whatever is available to be saved as a PWA it doesn’t seem to be commonly done.

There must be some other problem than Safari. It can’t be the big issue.

> ~20% of the market worldwide

But a much higher percentage of the profitable market, in terms of app income.

There are many markets where it is hardly a presence out of those 20% and not every company is a global corporation.

Google, to some degree, worked with Twitter to make twitter.com a 'flagship example' of what PWAs could be. If that's the best Google could come up with, the platform stands zero chance.

I prefer the Twitter PWA to the native app. What do you think is wrong with it?

Filled with empty loading pages? Not smooth at all compared to the native app? And having zero animations doesn't help at all

And that’s for an app that is actually trivial.

Twitter is actively trying to force people to use the app by making web experience miserable.

Everything loads instantly for me, including the app because I always have a browser process running anyway. What browser are you using?

Because Safari sucks.

The reason iOS and Google/Android PWAs suck is because Apple does everything in its power to make them suck:

- Complex and user-unfriendly "add-to-homescreen" functionality

- Limited/trailing functionality for native APIs

- The obnoxious url/browser bar persistent at the bottom of the page, with horrendous and uncustomizable hiding/showing of said bar.

People advocate for PWAs, but I never use them because as an end-user they're awful to use, because developers can't create them to the quality of native applications because of these limitations. Of course the complaints go to Safari in that case.

They absolutely could have PWAs be of higher quality, but they don't, for obvious reasons.

I get why they suck on iOS.

Why aren’t they on Android? Windows? Linux? MacOS?

Existing PWAs should be better known, especially in HN-like/tech circles.

But that’s not what I see.

I can't entirely speak for Android, but I assume that Google has similar nonsense (as evidenced by sideloading being a pain to the average user)

For Windows, Linux, and MacOS, there's no real need for them, or not enough need for them to exist. There are no policies or guidelines to follow, there is no 30% tax.

The browsers people use on desktop devices provide everything that developers need and can reasonably expect of the platform. There aren't any usability differences, there are barely any performance differences, and there are barely any feature differences.

In fact, I think many desktop end-users would prefer the app to be web-based because it feels so native.

In addition, it's what another commenter said - if it can't be on iOS, then what's even the point of having one?

Because many countries hardly have an iOS presence and there are tons of companies that only care about the local market.

You can hide the URL bar with the right meta tag, no?

If the user adds the PWA to their homescreen through the user-unfriendly process, I believe so.

But users won't (and don't) go through that.

>There is always some complaint about Safari instead.

when one app is used by roughly half of all mobile users, that's not something you can ignore. so complaining about that one app when these conversations come up does not seem out of place. essentially, the conversation is something like:

- PWAs are great!

-- How well do they run on Safari?

- PWAs are great on other platforms.

-- So about half of mobile users? Doesn't seem that great to me

Safari is used by half of US users. Worldwide it’s closer to 20%. And the desktop is totally open.

I’m not saying I expect PWAs to be huge, but it seems like they should be more notable PWAs if people really wanted them so bad.

meh, semantics. the point is that there's a large enough audience that it doesn't add up to making sense for dev time.

There's a bunch that run very well on Android: Twitter, Instagram, Uber, Lyft, even TikTok all have very good PWAs.

TikTok aggressively funnels you into the native app for various business reasons, but the experience is well polished if they'd let you use it all the time.

Safari (really that iOS won't let you pick a different browser) has definitely been holding back the growth of PWAs though. Without decent install to home screen support (apple doesn't implement install banners for web), no push notifications, many of the native API integrations missing, it's hard to go too far down the web path before the business folks want the native app for growth angles. It's a lot more about making money than the better UX that can be had with native.

> many of the native API integrations missing

Those "native API integrations" are Chrome-only non-standards that neither Safari nor Firefox are ever going to implement.

Personal experience shipping a big PWA: Safari's PWA experience is not just bad, it's all but undiscoverable:

Open the Share sheet, scroll down, tap "Add to Home Screen," fill in the information, tap Add.

There are simply no users that do this, and it is not an alternative to having a mobile app in the App Store.

> There are simply no users that do this…

100%. PWAs are not a priority for Apple (and shouldn't be for developers) because users don't search for apps in web search engines, they search for them in the App Store.

App store discoverability is very poor. Web discoverability is much easier to manage. It's something you have more control over and can buy easier through marketing via any channel you want. App store pages just don't convert well.

I agree about the installation process (though the user shouldn’t need to fill in any info). The actual user experience when using home screen PWAs can be made to be OK, but there really needs to be a meta tag or whatever to allow an ‘Add to Home Screen’ banner to appear (similar to the way you can link to the App Store if you have a native app).

This was an interesting recent development if it ends up being true, specifically Apple opening up iOS browser engine restrictions:

https://9to5mac.com/2022/12/13/apple-mulls-opening-browser-e...

Personal experience as a web user: I agree.

I'm not even talking about bugs per se. Just the experience. I find myself constantly battling misinterpreted gestures. I try to zoom out on a page. Safari thinks I want to switch to the Tab View. I don't. Please just make the page smaller so I can see it all.

Or, I try to grab a UI slider that's positioned near the left edge of the viewport. Safari thinks I want to navigate back to the previous page. Nope. Apparently if form controls are too close to the screen's edge, they're read-only at that point?

So many gestures are overloaded now that it's a stressful thing sometimes to get it to do what I want. Being pixel-perfect with my fat dumb finger is no fun.

Is it because it's actually riddle with bugs or just that most devs built on top of features only available on Chrome?

Reply on news.ycombinator.com