Hacker News Reader

Comment by IshKebab | Hacker News Reader

All sounds great. I'm quite curious what their objection to Apple's browser choice screen is though:

> The Commission is concerned that Apple's measures, including the design of the web browser choice screen, may be preventing users from truly exercising their choice of services within the Apple ecosystem, in contravention of Article 6(3) of the DMA.

Looks relatively reasonable to me:

https://preview.redd.it/ios-17-4-db1-new-default-browser-pop...

Is it because they always put Safari at the top and then try to randomise in a mix of browsers most people won't have heard of?

View on HN

Having Safari hardcoded to the first position in the list certainly won't fly. The Windows browser choice ballot and the Android search engine choice screens eventually settled on a two tier with the five most popular options in a random order at the top, and a second tier was shown after them also in a random order.

But the other problem is the "not now" button, because it's what the large majority of users will click. Does it by any chance have the effect that Safari is kept as the default, and the question is never asked again? The Windows browser choice ballot and the Android search engine choice screens forced the user to make a choice.

Isn't having the 5 top choices always be the same kind of the same violation?

It entrenches the big guys.

The idea that any of this is or ever was a problem to the end user is laughable honestly, this is political theater. End users want to be able to browse the web, that’s the point of a web browser. Most of them are based on the same underlying technology and implementations. Throwing around political power to force OS teams to spend engineering resources on the illusion of choice is actively harmful for basically zero benefit to the end user.

It's not a problem for the end user to be unable to pick which browser they want to use?

Speak for yourself, I would consider being forced into a single browser choice a huge problem, to the point that I will not buy a device that does it.

The fact that alternate browser engines were banned on iOS was indeed a huge problem, but I think the current legislation goes too far. As the GP said, many users won't care. I think trying to force them to care by presenting a randomized list of a dozen different options at setup time is misguided. I don't see the issue with merely having a default so long as there's an easy way to change it and Apple isn't applying any rules or restrictions to competing browsers that they don't apply to their own.

Indeed, outside Hackernews hardly any user cares about the browser.

However it becomes relevant once a dominating vendor uses the monopoly to limit the capabilities of the browser to benefit their app store etc. and that is hard to notice. For good competition the EU tries to limit that early.

Yeah, and what they’ve done here will result in a Chrome monopoly everywhere. They needed to break Chrome before going after Safari and Apple.

Apple really shouldn't have used a monopoly strategy to block another, more theoretical monopoly from happening.

I agree, I’m not on Apples side in this stuff, and actually think the App Store strategy tax is killing Apple as a worthwhile company. I wish they’d just dump the whole thing. But the monopoly of Chrome is getting to be as bad as the IE6 days, and is worth prioritising in my opinion, because it affects everyone, not just those who buy into Apple products.

Except that Apple monopolistic behavior would have stayed limited to iOS devices, which they make and maintain. If users were staunchly opposed to this “monopoly,” they had the option to switch to any of a huge selection of highly comparable choices.

Google wins in this situation, because now their dominance expands to far more devices.

> they had the option to switch

Let's not mince words; the "option" you are referring to is to throw away your iPhone and buy a different device. The fact that this is the only path of recourse is the obvious reason why Apple is being investigated multinationally.

Users don't have to be opposed to a monopoly for one to exist. Customers loved Bell telephone's monopoly, because Bell gave their users free long-distance calling. That love did nothing to stop what the government did next.

Users "throw away [their phone] and buy a different device" basically every 2 years, anyway. Apple's policies haven't changed significantly, and their market share among mobile devices has stayed between 11 and 25% for 10 years now. Their "monopoly" is limited to what runs on the devices they sell, and its impact on the web is at worst a wash, and at best good for maintaining diversity among web engines.

Were iOS to represent an impactful plurality of mobile devices, I'd agree that there's monopolistic potential, but right now this is just a case where everyone (users, competitors, and government representatives interested in "sticking it to uber corporations") are all incentivized to push for legislation/regulations of this sort.

> Users "throw away [their phone] and buy a different device" basically every 2 years

If the DOJ's accusations of lock-in are true, then it's kinda a moot point. Perhaps even more criminal on Apple's behalf.

> its impact on the web is at worst a wash, and at best good for maintaining diversity among web engines.

It is not. Apple deliberately ignored PWA functionality for a decade until regulators threatened them enough to add it. They have dragged their feet supporting actual third-party browser options[0] and undeniably harm user freedom in the browser market.

"at best", Apple is using an illegal double-standard to prevent Google Chrome from competing with their underpowered browser. It is a pathetic plea to avoid competition, which is inexcusable no matter how powerful Google is. Realizing user harms to prevent a theoretical problem is not an excusable pattern of behavior - the DOJ and EU are absolutely correct in their judgement. They'll likely be right when they implicate Google too - but that's a different discussion, and entirely different harmful business strategy.

[0] https://mozilla.github.io/platform-tilt/

And inside hackernews nobody cares what the outside people care about.

I can see that some people would think that, but it's a design that was converged on after several iterations and seems to have worked for both the regulators and the companies. I think both of the previous examples actually started with just enough entries for a single screen, and were later expanded to include more entries in a second tier.

From the user's perspective, having to make a choice from an unsorted list of hundreds or thousands of items would not work. That's why the regulators seem to accept the idea that not every product has the right to be on those choice lists, and not every listed product needs to be equally prominent. What's important is that the criteria are objective and don't give an unfair advantage to the platform owner.

When Microsoft was made to do a browser choice screen for the EU in 2010 they had a two tier system and it seemed to go well with the EU. On one hand, sure - it is an advantage for the popular browsers. On the other hand the goal is more that the manufacturer not be using their position to force the browser so heavily not necessarily to put every Joe Schmoe Browser in front of the same number of eyes. From a practicality perspective it solves the main problem without becoming a complication in itself.

Safari is not hardcoded to the first position, it is fully randomized.

From all the comments in the Reddit thread and from Apple's documentation it's entirely random.

Thanks, I was going just by the OP. Then I'd have to imagine that the problem is with the "not now" button.

> But the other problem is the "not now" button, because it's what the large majority of users will click

Yeah, because not now, dammit. I am not ready to choose the browser, I have no clue. Geezus, give me an option if you really insist on me to have options but don’t FORCE me to make my mind up when I’m not ready, right there on the spot. How do I even know if this is the best choice I am about to make? Just imagine that every time someone taps “not now” they simply say “I don’t care”. Why do I need to feel the breath of a bureaucrat on my neck when settling up my phone? Just because the bureaucrat had some beef with Apple I’m now forced to make some possibly uneducated decisions. Does your 70 year old mother or 3rd uncle knows which browser is the best and the implications of making the choice are?

Go and sort out more pressing matters. Leave my walled garden alone.

You are entirely free to just choose Safari, in that case.

Sure, but how does a person not fluent in tech know what the good choice is. Think from the perspective of a tech illiterate for a moment. They won’t know and the law will forbid making recommendations, and Safari cannot be put as first choice because the bureaucrat doesn’t like it.

I’m curious. Why isn’t Microsoft forced to present a similar screen for browsers?

Microsoft has been forced to show the Browser Choice screen way before DMA: https://en.wikipedia.org/wiki/BrowserChoice.eu

I presume that was only visible when the region is Europe. But also the obligation expired in 2014. Weird.

> Microsoft's obligation to display the Browser Choice screen to Windows users expired in December 2014.

Criticism I have seen:

- Safari will always still be installed and have its icon on the first page of your home screen by default, even if you choose another browser.

- Choosing another browser doesn't actually select it - it brings you to their app store page, where you still have to tap the (relatively tiny) install button.

- If you already have other browsers installed that are not included in the choice screen (e.g. Firefox Nightly, I believe), then you're still forced to install/pick a different browser.

- you can't uninstall Safari. The DMA clearly requires Apple to allow uninstalling it.

- the Share Sheet for in-app webview privileges Safari (Safari-specific "Add to Reading List", "Add Bookmark" at the top) regardless of default browser.

- apps with in-app webviews can roll out a custom engine, but otherwise it uses WebKit regardless of default browser. It should obviously use the webview facility provided by the default browser, if it offers one.

- confusing to change defaults. There's no central "Default Apps" screen in Settings. You have to go in the Settings app, scroll way down to a given browser, and then you can change the default. So the default browser setting is weirdly treated as an app-specific setting rather than a system setting.

- If Safari is not the default, you can change your default browser within "Settings.app >> Safari". If Safari is the default, you can't change your default browser there at all. Whereas the default browser selection menu will always be shown in the Settings.app submenu of third-party browsers.

The default browser has no technical ability in iOS (currently) to supply an in-app component to replace WKWebView or SFSafariViewController, and frankly as an app developer I shudder at the idea of having to test our use of in-app browsers against every browser, every release.

I think that in-app browsers that are solely used to render the app, and not separately accessible web pages, should be considered an implementation detail of the app. That also means that apps should be free to bring their own browser engine for that purpose (à la Electron). However, where apps provide access to web sites that are also accessible outside the app, the user should be able to configure the browser for that (possibly different from the system default). That should pose no particular problem for app developers, as those external web sites need to work under different browsers anyway.

> as those external web sites need to work under different browsers anyway

You say that but many sites essentially only support Chrome (and Edge by extension). If you install Firefox, or keep Safari, and the external sites just continues to assume everyone downloads Chrome you get nice broken web views. Or if a user changes the default browser (and thus web view engine) after the fact.

> many sites essentially only support Chrome

Yes, but they shouldn’t, and it’s not that difficult, so I have no sympathy. It’s also not that many.

"It’s fixable so it should be fixed, and even if it’s not fixed, it doesn’t matter," is a pretty good summary of how much thought the pro-regulators are bringing to the topic.

I'm not saying sites can't or shouldn't be fixed to support browsers other than Chrome. The fact of the matter is these sites exist and aren't incentivized to support anything but Chrome. When an overwhelming proportion of visits are through Chrome they only support Chrome, just like when those visits were IE6 and sites only supported IE6. Their advertisers love all of Google's web API proposals because they feed fingerprinting mechanisms.

Sites that only support Chrome and have no need to support Safari (WebKit) will continue to support Chrome. Users that want to have a third party browser handle web views in apps or web apps won't have meaningful options as WebKit won't be a requirement so every site will code only to Chrome. The status quo at least requires a vendor support WebKit if for no other reason is to support web views in their iOS apps.

I don’t see why the situation for viewing public websites from within apps should differ from the situation on the desktop. People choose which browser they use on the desktop, and live with any difficulties that may or may not bring. (E.g. I’m using Firefox and have few problems.) So the same would be the case for browsers in mobile apps.

Firefox isn't a transitive dependency on the desktop. If an app uses a WebView[0] control on Windows, Firefox has no effect on its behavior and performance. It uses the Edge engine to power WebViews. On iOS WKWebView currently uses WebKit.

The EU apparently wants to change this behavior on iOS. A WKWebView would be backed by the system's default browser selection if it made claim to support WKWebKit views. So your third party app with no direct connection to Firefox that uses a WKWebKit view now depends on Firefox handling the loaded pages in the view. If they don't work in Firefox or whatever browser the user has selected the third party app is affected by the choice of browser. This also has an affect on web apps on the Home Screen.

The issue with Chrome is it already dominates the web. App vendors will drop Safari WebKit support as soon as WKWebKit no longer mandates the site work in Safari WebKit. Without iOS enforcing support for Safari WebKit by being the only option for the non-trivial number of iOS users of the web, Chrome will be the only supported browser in apps and the wider web.

[0] https://learn.microsoft.com/en-us/microsoft-edge/webview2/?f...

...apps don't use WebViews on Windows (or MacOS really, for that matter) because they are unstable APIs that Microsoft and Apple both break in nonsensical ways. It's why Electron is king, in many respects. The only desktop software that use WebViews afaik are the "native apps" which really ought to be handled with a leaner library anyways.

> App vendors will drop Safari WebKit support as soon as WKWebKit no longer mandates the site work in Safari WebKit.

Some will. Wanna guess why?

> It should obviously use the webview facility provided by the default browser

Android doesn't do that either. You get Android System WebView which is always WebKit/Blink.

Windows didn't provide any browser engines apart from Trident through mshtml.dll either.

It would make iOS development harder - instead of getting a web view working just for WKWebView, you would need to get it working for every WebView out there. Given that they are not full browsers, it's not as simple as following web standards/caniuse. For example, service worker support in in-app browsers is limited.

If this actually gets implemented, every app out there will most likely pin the in-app browser engine to WKWebView OR bundle a binary blob with their own browser engine directly into .ipa. Bundled browser engine in .ipa has privacy implications - the app will be able to fully read secure HTTPS cookies of the sites that user visits - something that's currently protected.

> Android doesn't do that either. You get Android System WebView which is always WebKit/Blink.

Not if the application uses Custom Tabs to start a web browsing activity.

https://developer.chrome.com/docs/android/custom-tabs/

Firefox fully supports this and if you set it as a default browser you will many apps open content in a Firefox powered webview. (Activity)

Apple can do the same. Not trivial but definitely possible.

> [...] if you set [Firefox] as a default browser you will many apps open content in a Firefox powered webview. (Activity)

This is great because the Firefox webview allows to break it out in to a proper Firefox tab in the app. Also, the Firefox webview also uses extensions installed in Firefox app, e.g. uBlock Origin.

Thanks to you and others for pointing these out. I want to make clear that this isn't the EU's list and is just stuff I could observe disregarding my default browser choice. But if there's good reasons, by all means, don't make developers' lives hell.

> apps with in-app webviews can roll out a custom engine, but otherwise it uses WebKit regardless of default browser. It should obviously use the webview facility provided by the default browser, if it offers one.

This has massive potential to break an unbelievable number of things, first-party and third-party alike. WebKit is used in all sorts of places people wouldn’t expect, for example at one point UILabel (the native control responsible for displaying non-scrolling text and labels) used WebKit under the hood for rendering attributed strings.

Users want to truly exercise their choice of browser screen so badly that a list of nitpicks about the ordering of options in the share sheet and the prominence of an "Open" button which works the same way for every other app install flow is thwarting them from doing so.

The DMA is not (directly) about what users want; it's about markets having level playing fields, where all competitors have the same opportunities and thus have to win based on merit, rather than based on dominance in a different market.

Of course, the idea behind that is that that competition will then lead to things that users really want, without having to mandate what that is specifically through central planning.

If users can be lost in meaningful numbers to such comparatively tiny frictions, it seems like the market will never sufficiently be a level playing field.

I agree that it's a hard problem (like fair markets regulation in general - e.g. it's not straightforward how to define a particular market either). Unfortunately I'm also not sure what the better solution would be - at least trying to make them as fair as possible seems to be the somewhat reasonable choice, compared to full central planning or a free-for-all market.

Arguments on the margin make sense, but DMA compliance has huge resource costs and will surely come with eye watering penalties. And yet if small barriers are indeed so insurmountable then the benefits will be limited. So: is it worth it?

I think cookie banners have not meaningfully changed users’ ability to exercise privacy preferences, but they have probably cost the economy many tens of billions of dollars to implement, enforce, and litigate. As far as I can tell, there isn’t much reflection on what that says about effective legislation.

Interesting! When I got presented this list Safari wasn't even "above the fold" which surprised me, so I assumed it was randomised.

When I clicked Safari it also opened an App Store sheet for Safari, which I thought was weird but probably done to keep it consistent with the other options (more like "see more details" before making the choice definitive/installing)

Thanks for sharing.

I'm curious about the sorting of that list. From the image (mirror: https://i.imgur.com/pjo78lS.png):

   Safari
   Vivaldi
   Opera
   Web@Work
   Edge
   Onion Browser
   Seznam.cz
   Brave
   Firefox? (cut in half)

Is it Safari first then randomized names? It doesn't look alphabetical order to me.

Chrome doesn't even show up, Firefox is almost hidden.

I don't know how the list is generated. But as others have pointed out, there seems to be established precedent that hardcoding Safari first would not be compliant. And alphabetizing would create bad incentives (Calling dibs on "Aardvark" as a browser name, BTW).

The problem I see with randomization is that with all the browsers that are skins on Chromium nowadays, this would be likely to steer users into a Chromium based browser.

> Calling dibs on "Aardvark" as a browser name, BTW

It's ok, I'm naming mine after Aachen

AABrowser is mine.

Dibs on !0A Aangstrom.

Green and eco, go for it.

Are those real browsers or just reskins of Safaris inner web components?

For now they're all WebKit. Apple now allows other browser engines to be used but they'll probably take a while to come out, if ever.

Oh, that's good news.

I'm excited for Firefox to appear there using its own engine.

When you look at the list you have to be concerned about the state of the browser market.

Chromium is going to dominate desktop and mobile and entrench pro-advertising measures eg. browser fingerprinting, long term first party cookies etc for years to come. Meta and Google must be thrilled.

Of course it makes a mockery of GDPR and other privacy measures EU has been pushing.

But users are demanding choice in browsers, and they are picking the least private, most pro-advertising one en masse! What’s its biggest benefit? It offers a suite of native-replacement, in-the-standards-process-but-not-finalized APIs so developers can write slow, bloated applications once and run them everywhere! This is what users want!

It is only good for Google. Not for the web.

It might be that Apple enforces the use of webkit.

This screen has already so much dark history by Microsoft on Windows.

“Select your birthdate from this randomized list of possible dates.”

You wouldn’t.

Cannot imagine why this is being downvoted. Presenting users with a randomized list is a dark pattern that discourages use of the associated feature.

Randomization is required by the regulation. Another instance of what a smart person wants out of a feature and what the EU wants being very different things.

It's being downvoted because randomizing something that is by its nature chronological is very obviously different from randomizing the order of something where you want to explicitly avoid giving the impression that one takes priority over another.

Chrome is the most popular browser and Apple’s largest competitor and isn’t on the first page?

Reply on news.ycombinator.com