6 ms·
This approach is a gotcha: If you want to distribute Mozilla Firefox or Google Chrome for free, this isn't possible under the old policy, and the new policy wou
by kevingadd 3y ago
This approach is a gotcha: If you want to distribute Mozilla Firefox or Google Chrome for free, this isn't possible under the old policy, and the new policy would mean 'open source web browsers cost half a euro per update'. It's nonsense.
- natch 3y agoBut the old policy allows distribution of these apps for free and developers have the option of sticking with the old policy — so I don’t see the problem.
- greycol 3y agoApples argument that they're complying is about as brilliant as "I don't need a liquor license because I'm not selling beer I'm renting glassess for the time it takes to drink a beer". The only way to get into 3rd party stores is to pay (or be at risk of having to pay) Apple which is not the case if you only distribute through the Apple store, this clearly disadvantages 3rd party stores in comparrison to their own store.
- kaba0 3y agoIn general, getting into a 3rd party store is not a good thing, and it doesn’t benefit end-users. I only ever want to see 2 app stores on my phone, the AppStore, and an FDroid alternative full of free, open-source apps. Anything else is good to be disincentivized.
- natch 3y agoThis makes it sound like complying is supposed to be a good thing. That’s an unfounded assumption imho.
- turquoisevar 3y agoThis sounds as a “want my cake and eat it too” argument. Apple’s default construction for getting payments for the usage of their IP is via commission on revenue. This is by definition a differential pricing strategy because not everyone has to pay for the usage of Apple’s IP. Many didn’t like the commission structure and made all sorts of arguments against it. It would just be for payment processing, it would just be for distribution, etc. Apple has always maintained it was primarily for the use of their IP, all the rest is thrown in as a bonus. They have structured it as such in the developer agreement and US courts have wholesale accepted it as such. One alternative that has been floated around a lot by people that accept that Apple wants payment for their IP but didn’t like the commission structure regardless was to split off the fee for IP into its own thing. Now they do that very thing in the form of the CTF (at a more competitive rate than Epic does for Unreal mind you). In addition there’s a separate commission for App Store services and a separate commission for payment processing. Now the new complaint is that this payment for their IP for first installs on EU iPhones per 12 months in excess of 1M installs in the EU isn’t good either. So what is the desired outcome? Use Apple’s IP for free? I don’t see how this disadvantages third party stores. Does it disadvantage Steam when Epic comes knocking on my door for their share of the pie? Epic charges me 5% of all my revenue above a million, Apple charges me €0.50 of all my EU installs on iOS above 1M installs in the EU. Apple’s fee is directly tied to my usage of their IP, I pay them €0.50 in 12 monthly installments for each installation that goes over 1M, but they don’t touch anything else I make off of that install. Epic wants 5% over every dollar I make over $1M worldwide. Don’t get me wrong, personally I was content with my 30% and the 15% is a steal for what I get out of it. But if commission of revenue is the big bad, then the only logical thing for licensing IP is an upfront cost for usage whether you earn money with it or not. This is how it was with consoles. Thousands upfront for the right to publish and using the IP + thousands for every build to be certified + commission over revenue. Indies later on got a reduced rate in the hundreds, until you got big.
- ClumsyPilot 3y ago> So what is the desired outcome? Use Apple’s IP for free? Can I publish an application that does not use Apple IP, let’s say something written in ARM assembler? If not, then this argument about Apple IP is in bad faith, and not the real reason for anything.
- jasomill 3y agoMaybe so, but I have no doubt Apple lawyers could make a plausible argument — in good faith, even — that you might need a license to link against required iOS libraries.
- turquoisevar 3y ago> Can I publish an application that does not use Apple IP, let’s say something written in ARM assembler? iOS apps can be written in assembly. I’ve done so myself out of curiosity, and it seems others have done so as well[0]. Of course, I wouldn’t recommend it. It’s excruciating. That said, isn’t this just a bad-faith argument wrapped in an attempted gotcha? Where your premise is that if you don’t have a choice, then you shouldn’t be charged? If Apple would say: “Hey, you can use Swift for free because it’s open source, and you’re welcome to benefit from our contributions to it, and while you’re at it, you’re allowed to use all the basic low-level frameworks that communicate with the OS for free as well. But anything more than that, and we want you to pay us for it.” Would you then be fully on board with them charging for using the remaining IP they own? Are IP and ownership rights fluid based on your own moral framework, depending on if one has a choice or not? Should I just blow past the toll booth of the only bridge here that allows me access to the rest of the country? How about the only ISP that provides service here? Should they just allow me to use their infrastructure so I can write you this comment? The other day I saw a nice TV at the local store. It’s pretty much the only place I can get this TV. Do I have your blessing to take it because I don’t have much of a choice? I make apps for a living, I’m literally the only place on earth where you can get my app. My app saves users a lot of money. Should people just pirate my app and force me on a diet of hardtack? Where does this fluidity start and where does it end? I thought, for better or worse (and we can spend hours debating this part alone), we decided that when someone owns something, and especially when they create something, they get to decide who, if anyone, gets to use it and at what’s price if any. 0: https://github.com/richardjrossiii/iOSAppInAssembly https://github.com/richardjrossiii/iOSAppInAssembly
- eptcyka 3y agoDo they have a provision that one will be able to JIT code in their blessed AppStore in EU now? If so, I've misread the announcement, as I didn't get that.
- MrDarcy 3y agoDoesn't seem like nonsense to me. Mozilla is a registered nonprofit so they're fine. Google could donate Chrome to a non-profit if they wish, or not and pay. Seems reasonable.
- crazygringo 3y agoNo -- as far as I can tell, alternative browser engines are going to be allowed in the old terms within the EU. Unless you can point to something that says the opposite? Apple points out that the new terms are required for third-party app stores and third-party payments. Not for browser engines. I'm expecting both Google and Mozilla to make their browsers, with their engine, available in Apple's regular store in the EU. (And even in the new agreement, it's not half a euro per update. It's per year that contains an install or update. If you update 12 times a year, you're not paying 12 times. You're paying once.)
- nox101 3y agohttps://developer.apple.com/support/alternative-browser-engines/ https://developer.apple.com/support/alternative-browser-engi...
- throwaway2037 3y agoThis link alone is worth a whole separate HN discussion. Thank you to share. The requirements list is very interesting. For example: > Meet the following functional requirements to ensure your app is using a web browser engine that provides a baseline of web functionality: Pass a minimum percentage of tests available from industry standard test suites: 90% from Web Platform Tests and 80% from Test262 And: > Program security requirements: You must do the following: Use memory-safe programming languages, or features that improve memory safety within other languages, within the Alternative Web Browser Engine at a minimum for all code that processes web content; Is WebKit written in C++? That language certainly isn't "memory-safe" (I am not dumping on C++ here!). I assume that Apple uses fuzz testing and static analysis to find memory safety issues. So... does WebKit pass this rule? Hmm... And, are there any browsers written in memory-safe languages? I assume they would be too slow (Java, C#, etc.) I'm not sure if Rust is considered memory safe here. (Can you have null pointer exceptions in Rust -- dereference a null pointer?)
- saagarjha 3y agoI believe Apple considers themselves compliant because they use technologies to improve the safety of their C++…?