Live data from Hacker News

Why we won’t be supporting Sign in with Apple

blog.anylist.com

311–320 of 485 posts

Re: Why we won’t be supporting Sign in with Apple

#311
post #10

Earlier quoted context omitted.

Well, that, and also if they don't remove FB login, then the would be required to also implement Sign in with Apple. They didn't want to implement Sign in with Apple, so they had to remove FB login.

Which is a pretty anti-competitive move on Apple's part. They're essentially using their control of the iOS ecosystem to benefit unrelated products. I seem to remember that those kinds of actions didn't work out well for Microsoft :)

So, forcing them to give users a choice is somehow equivalent to... forcing vendors to not give customers a competitors browser?

Ok sure. That seems totally the same. Not at all as ridiculous as the invented "monopoly of iOS" that people use to justify the claim Apple is abusing a monopoly position.

Next thing you'll complain Coca Cola is abusing its monopoly position on Coke.

Re: Why we won’t be supporting Sign in with Apple

#313

Earlier quoted context omitted.

The check on this is that you are now putting an inconvenience on that apps users. Maybe it’s okay in some isolated circumstance however, I imagine this won’t sit well if it becomes a reoccurring theme for iOS users, as they will come to view Sign In With Apple as unstable/unsafe (like Google and it’s graveyard). It would quickly render Sign In With Apple useless if the trust that it will continue to work whenever us…

> The check on this is that you are now putting an inconvenience on that apps users. That hasn't stopped Apple from remove apps or preventing updates -- clearly inconveniences on users -- for whatever reasons they want. It has happened and it will happen again. There is no check on this unless your app's audience is Netflix/Spotify/Facebook huge . If you're just an average developer you can be killed off at any time.

I’m not exactly sticking up for Apple as much as I’m trying to demonstrate practical thresholds that would realistically limit this behavior. I don’t think Apple wants 100,000 upset customers let alone millions.

Hardware requirements for software is a decades old concept, and it’s true they deprecate and obsolete supported software and hardware on for older platforms, but it’s rare I’ve seen Apple taken a user hostile approach here within supported lifetimes though it has happened yes it is rare.

Their developer experience on the other hand does not see the same care and attention a lot (most even) of the time. That’s because, and I believe this strongly, Apple never wants 3rd party software to have platform influential power over them again, like Microsoft and Adobe did for decades. It’s sad but not unsurprising that their platforms can be very developer antagonistic if you don’t take their happy path (and sometimes even then). To them though, it doesn’t matter until it affects a large quadrant of the Apple consumer base and in some occasions yes not even then, but largely it’s the consumer who has the biggest voting block with Apple in terms of pressure on the platform, as I’ve watched it play it they never had a history particularly after the iPhone came out of having the best developer relations relative to say, Microsoft, who provides a very positive experience in comparison

It’s just not in their DNA because of the fear of having too powerful of vendors putting pressure on the platform that they otherwise control outright. When you look at their policies in this context they make a heck of a lot more sense (even if you don’t agree. I certainly do not always)

Re: Why we won’t be supporting Sign in with Apple

#314

Earlier quoted context omitted.

Having had this same problem multiple times, I actually made it a point to save a “login” for those sites that when I autofill it reminds me which auth service I’ve used. Username: Log in with FB Password:

Neat trick. Some websites don't even bother giving you any choice for traditional username/password input though.

In that case, I just click the icon for my pw manager (1Password in this case), and the modal shows me the same message.

Re: Why we won’t be supporting Sign in with Apple

#315
post #171

Earlier quoted context omitted.

Time to create a new service to unify all your SSO accounts! One single SSO!

OpenID would have done it, but Facebook and Google neutered it in favour of OAuth so they could cement themselves as primary players.

It's been some time last I checked, but isn't OpenID Connect provided anymore by Goog/Fb? Why wouldn't that be a reasonable choice if you wanted a protocol that, from the dev side of things, allowed you to uniformly target external auth providers, or your own?

Re: Why we won’t be supporting Sign in with Apple

#316

As they point out at the very bottom, all their arguments apply to all third-party sign-ons, so they're removing Facebook as well. So there's nothing specifically against Apple, despite the title seeming to imply it -- just that they're taking the move right now because of Apple's new policy coming into effect. I've got to say, I really wish there were a way to know whether I already used Facebook, Google, or Apple t…

"all their arguments apply to all third-party sign-ons" No they don't. Other sign-on options don't obfuscate the email address. They are likely removing FB login as otherwise their next app update will be rejected by Apple for supporting third party login but not Apple login.

Additionally many other 3rd party login systems have been around much longer than Sign in with Apple, which was another strike against Apple for the author.

Re: Why we won’t be supporting Sign in with Apple

#317

Earlier quoted context omitted.

I'm going to be fascinated to see what this does for conversions. My company built the Neil Young Archives, when doing so we initially launched with Social log ins and at one point Neil decided Facebook and Google were evil and wanted to remove the access. According to our logs a full 2/3s of all users were registering with a social account and we were having great success getting folks to log into a free service (We…

This escenario seems an clear candidate for A/B testing since I would be conflicted by wanting to provide privacy but understanding it might impact user signups. Seeing some actual numbers would help me in making that decision.

A/B testing auth methods is tricky. It's fine for "can we get better sign up rates" but it wreaks havoc on "can my previous users still sign in". At best you can AU/NA test - show different pre-auth treatments in different, very distant geographies that nevertheless have roughly similar user characteristics.

Re: Why we won’t be supporting Sign in with Apple

#318

Worth noting that AnyList automatically subscribed me to a marketing list without double opt-in or any kind of consent, which is exactly the kind of behaviour that makes me not want apps to have my real email address.

The article says "When you provide us with your email address, it is never sold, shared, or used to invade your privacy." So, one of you is lying. I don't have the means to determine who, but I don't see what your motivation for lying would be, but can see what theirs may be.

And to extend that, if they are a spammy company, that would be exactly why they would be complaining about SIWAI.

Re: Why we won’t be supporting Sign in with Apple

#319

Earlier quoted context omitted.

Having had this same problem multiple times, I actually made it a point to save a “login” for those sites that when I autofill it reminds me which auth service I’ve used. Username: Log in with FB Password:

Neat trick. Some websites don't even bother giving you any choice for traditional username/password input though.

My router doesn't accept a username at all, which actually makes a bit of sense since there's only one 'account' on it. Unfortunately, this means that neither the browser nor LastPass will save the password. My previous router allowed me to change both the username and the password, then pre-populated the username field with whatever I'd entered (but didn't disable the field), which I always thought was odd...

Re: Why we won’t be supporting Sign in with Apple

#320
post #167
post #101

Earlier quoted context omitted.

The article also implies that if anyone can guess your email address, they can send you/share with you a list. I wonder what anti-spam measures AnyList implements?

I assume they rely on the fact that sharing a random list with a random person a thousand times is pointless.

What if the list contains/is itself advertising, and the random person is everyone on a huge list of active email accounts?
Post reply on HN