When this was written I was an Apple fanboy.
These days, as a dev, I really don't trust Apple at all. I would never even think of making a product for any of the Apple platforms. As a freelancer I've stopped taking any iOS/macOS projects too. It's not the 30% but Apple's attitude of controlling their turf like the mafia under pretended moral superiority.
As a user I'm as cynical as I've ever been. I use macOS and Windows on a daily basis, and for dev work I generally prefer macOS over Windows. But I'm always running at least a major macOS version behind and never buy any 1st gen Apple product. I simply expect their stuff to fail in one way or another. I've been bitten way too many times to trust them. Nvidiagate, Radeongate, Yosemite, iPad 3, iPhone 4, etc, the list is very long.
Just weeks ago, my wife's Macbook Air (a 4 year old machine) started having issues with the keyboard and trackpad disconnecting. Apple's authorized repair services* say they want to replace almost all the parts (pretty much excluding the screen) and the repair cost is almost as much as buying a new Air.
* Apple Mexico doesn't really do repairs here like in the US so you're forced to go through one of those services.
Or right now, the Apple TV has been suffering issues with Atmos for the past couple of months. There's a thread in Apple's forums with 14 pages of users complaining which keeps growing even though posts are deleted by the mods constantly.
I'm personally sympathetic to Graham's points in this 2009 piece. He was prescient about the "bummer" it is to be stuck in a monopolist's monoculture.
But Apple's growth in market value since – from $175B at the end of November 2009, to $2.73T today – at least raises the possibility that Apple wasn't making a "mistake", from the perspective of Apple shareholders, managers, & employees.
Maybe once you've reached a certain powerful 'commanding heights' via other strategy & technology, the optimal enterprise strategy is truly to bully developers, & take their lunch money for yourself.
Still a drag on human progress & welfare, sure.
> An organization that wins by exercising power starts to lose the ability to win by doing better work.
Yup. That's more or less what's happening or Oracle right now. They'll profitability soldier on for a while due to the massive amount of legacy servers running Oracle databases at companies all over the world. But, every $BIG_CORP out there has on their roadmap a long-term plan to either drastically reduce their spend on Oracle products, or more often, eliminate it entirely.
Well, as much as the business model is untenable, it lasted 13 years and counting. [0]
The things that have changed aren't strictly upgrades or downgrades. Apple's approval process is much quicker than it was in 2009, but instead of sitting on a bug fix for four weeks, now they reject you with no explanation while 20 other apps doing the same thing Apple is angry about make millions on the store.
>They get away with maltreating developers, in the short term, because they make such great hardware. I just bought a new 27" iMac a couple days ago. It's fabulous. The screen's too shiny, and the disk is surprisingly loud, but it's so beautiful that you can't make yourself care.
I feel the exact same way with the 13" iPad Pro I bought last year. Amazing hardware, but it feels like computing in a straitjacket. Even jailbreaking it only goes so far to fix the underlying problem, which is that playing by Apple's rules is a genuine roadblock to a lot of pro app developers who are accustomed to selling direct-to-consumer.
However, I don't think Apple actually cares. The iPad Pro exists for exactly one kind of customer, and one kind of customer only: professional artists. As far as they're concerned, people who want proper developer tools, the full Adobe or Autodesk suites, or what have you can just give up the touchscreen and cellular modem and buy a Mac.
>With Apple that seems less the case. When you look at the famous 1984 ad now, it's easier to imagine Apple as the dictator on the screen than the woman with the hammer.
...Did Paul Graham consult with Tim Sweeney on #FreeFortnite?
[0] Also, the submission title should probably have a (2009) in it...
Related:
Apple's Mistake (2009) - https://news.ycombinator.com/item?id=7344783 - March 2014 (56 comments)
Apple's Mistake (2009) - https://news.ycombinator.com/item?id=6329991 - Sept 2013 (8 comments)
Apple's Mistake - https://news.ycombinator.com/item?id=1081514 - Jan 2010 (25 comments)
Apple's Mistake - https://news.ycombinator.com/item?id=950751 - Nov 2009 (269 comments)
So I built a very basic prototype app to explore an idea I had using Flutter (imagine something like a time boxing system where you track how frequently you get distracted). Spent a couple days roughing it out, and then I was able to load it onto my own Android device to start testing. Literally just plugged in my device to my PC, confirm the prompt on the phone, and then `flutter run -d <device_id>`. I also wanted to send it to my brother (who has an iPhone) so he could start playing with it too.
So I had to pay Apple $100 to open a developer account. Then sign a bunch of contracts. Then register an app ID and generate certificates. Then bundle the app and submit it to appstoreconnect. Then wait for a bunch of automated checks to pass and tell me that the app ID is missing entitlements. Then figure out what that even means, attempt to fix the entitlements, and run through the submission again. Then set up testing groups and invite my brother to test the app. Then I have to "submit" the app for approval for testing. And a couple days later finally get approval from Apple so that I can start testing it. And that's the short version; there were a bunch of nonsensical Xcode quirks to deal with.
For context, I've been an iOS developer for a decade, and it still took me the better part of a weekend to get through it all, even though I know exactly what to expect. And this is so one fucking person can try an app for me.
I hate Apple. I don't know how else to put it. I don't think they're evil, I just think they don't give a shit about their developers. Did you know their bug tracker is private? I'll find some weird behavior with, let's say UITableView or something, and spend all day trying to figure out what subtle thing I'm doing wrong using their comically poor documentation. And then somewhere buried in Stack Overflow I'll find someone saying "oh yeah that's a bug in the framework, I submitted a radar 2 years ago". So there's some bug that they know about, and I'm not even allowed to see that. It feels hostile on their part, and it makes me feel hostile on my part. I think Paul Graham was right.
Focusing on iPhone users and and not iPhone developers probably was/is the right thing to do.
iOS isn't a developer's platform and if it were, developers probably wouldn't be interested in it very much, since what developers want more than any API, tool or system is users.
Not that they couldn't focus on users better, or couldn't support developers better w/o impacting users negatively. They could and should, IMO.
Apple has pretty much fixed the issue this post is concerned with (very slow app review process)... but it took several years -- checking the internet, it looks like ~2016 was the general turning point, so around 7 years after this post. It shows just how little the developer experience matters to developers if you have users.
As @Apocryphon comments in a subthread, there's a commodification of app developers.
With iOS, Apple has crossed the threshold of who needs who. Now app developers need Apple way more than Apple needs them.
In many cases, one app to do thing X is mostly as good as another app for the same X. There's no bargaining power to be had unless you're a huge name.
As a developer, you're not leaving the App Store unless you're leaving mobile app development entirely. It'd be commercial suicide (for most) to do that.
Developers have pretty much solved the launch-fast-and-iterate problem with server-side updates — they don't need to submit a new build to Apple for every update. I'd argue that Apple's app store policies have mostly resulted in higher quality, more secure apps than we would have otherwise (and Android apps are evidence of that), which resulted in more trust and faster adoption from users.
I think it's easy to read this post, then look at Apple's market cap and conclude that Paul got this one wrong.
I don't think he did.
As an entrepreneur, when I held the iPhone 4s in my hand ~10 years ago, I saw only possibilities. In 2022 when I hold my iPhone 13 in my hand, I see a known quantity with all the use-cases permanently ossified. I don't dare to dream about what is possible. Only Apple can truly move this platform forward. I'll be a passive observer.
This is not how I feel about the Mac, and I think Paul's post accounts for the difference. I really feel if I have a great idea and execute I can meaningfully impact the trajectory of the platform.
There is no doubt the iPhone and the iPad will continue to be commercially successful products. But now, 14 years after the launch of the App Store, that feels like such a poor measurement of their net-new impact on society. The magic is gone.
Developers really don't have a lot of leverage in anything we do, as much as we like to think so.
Many of us work in such a way that we get summoned every morning at 9 to line up and answer "what did you do yesterday that justified 10% of your paycheck and what will you do today that justifies 10% of your paycheck?
Turnover among executives is considered a crisis. Turnover among developers amounts to whining about our disloyalty and companies screaming at the government to increase immigration.
We are a commodity. A valuable one and hard to obtain one to be sure, but that is it.
Apple understood that and won and developers primarily build for Apple now.
> They treat iPhone apps the way they treat the music they sell through iTunes. Apple is the channel; they own the user; if you want to reach users, you do it on their terms.
Yep, fast forward to now and it seems that they got it right. Nobody left. Not the users, not the developers. It turns out we really are like musicians
While I agree with a lot of the sentiment in this blog post, it certainly didn't age well. I don't see "developer love" making a lick of difference for how successful the App Store has been (which is, phenomenally). At this point the only thing I see reigning in the power of the App Store (and Play Store, a point pg also got wrong in this post, but I think that was understandable in 2009) is regulation.
Favorite passage:
How would Apple like it if when they discovered a serious bug in OS X, instead of releasing a software update immediately, they had to submit their code to an intermediary who sat on it for a month and then rejected it because it contained an icon they didn't like?
> Their reputation with programmers used to be great.
Maybe in a vague secondhand sense? Mac developers have pretty much always been a tiny, tiny niche, and I've heard plenty of complaints from them.
If you can make something really nice for developers to use, they may use it - if there are some users to sell to (or, it's orthogonal to the released product).
This is partially why Electon has spread so far I think - developers like working on Macs (because it's basically "linux with a nice gui" for all practical purposes) meaning that if Microsoft wants developers to use VSCode - and not get used to non-Microsoft products - they need to get to where the developers are.
But in those case the devs are the users.
A beautiful SDK won't save you, I've heard stores about how good the BeOS API was, or how wonderful development for various other failed devices/technologies was.
But if you're trying to keep a dying platform alive, you need to reduce friction, and making a good SDK is a way. And Macs were dying for many years ...
I've dealt with the IRS and with the Apple App Store. The former has far better customer service and is much more transparent.
This post was perfectly on point.
I think it can be argued that Apple has never been developer friendly, but these days it is painfully difficult to ignore.
And I think this is a huge mistake, I would not be surprised if Microsoft end-up eating their lunch again.
As other commenters have already mentioned, it doesn’t really matter how “broken” the App Store is as long as Apple owns the user relationship. Developers follow the user base, which is dependent on Apple’s hardware and overall ecosystem. Although Paul provides a few ad-hoc conversations with frustrated developers, it doesn’t seem like any of the issues mentioned impact the supply of developers for both Apple and its App Store in a meaningful way.
Paul suggests that Apple is “evil”, as if the company is immoral and intentionally aiming to harm a group of people (in this case, the App Store developers). I don’t see any “evil” intent here, and think that Apple intended to develop a feedback loop to incentivize App Store developers to improve quality control.
It’d be difficult to obtain, but what would be interesting to see is a graph of the following data points:
- The frequency of moderate to high-severity client-side bugs released for the same app to Google's Play Store and Apple's App Store.
- If there were ever times where Apple has tightened or loosened their approval process, the before and after on the frequency of client-side issues.
Although app developers have no control over the app approval process, one way to mitigate turnaround time risks in fixing issues is implementing as much logic as possible server-side (of course, there are definitely times where this isn’t feasible).
Given that pg is a VC, it's interesting to take these comments into consideration given that in 2009 - Apple's market cap was ~$100B and today it's ~$3,000B ($3T).
Getting a 30x return for a company at scale is a phenomenal outcome for any VC. Seems like Apple did something right.
The part about lost goodwill sounds very true. I used to love Cocoa, and did my best to write quality apps for it, despite Mac's small market share.
But App Store ham-fisted rules made me distrust and resent Apple. I've stopped using Apple-only languages and frameworks as much as I could, and as a user I'm looking to leave the platform entirely (which unfortunately after a over a decade of using, I'm locked into and resent that too).
I get it, in 2009 smartphones weren't devices that everyone owned. But I still find the myopic obsession with "techie" approval as a harbinger of success pretty narrow-minded. People didn't like the iPhone in 2009 because programmers liked it; people liked it because it felt like (and was) a revolutionary product!
> They're so attracted to the iPhone that they can't leave. But they're looking for a way out.
I know hindsight is 20/20 and all, but there's a much broader point that could have been made here about "what happens in a world in which entire software distribution chains are at the mercy of one or two megacorps?" Now that would have been a remarkably prescient gripe, and a much more accurate guess at the "mistake" Apple was making (if you can call the current backlash a "mistake"?)
Article is from 2009. Nevertheless interesting to go back in time and see what has changed. Or hasn't.
Sidebar: am I seeing this wrong, or is PG's website really showing a bad SSL cert? Looks like it is serving some cert for Yahoo domains?
It also isn't auto-upgrading to HTTPS, which is a smaller issue, but just kind of surprising for someone with PG's tech stature.
Apple gets lots of hate but one thing that's rarely mentioned is how the Play Store prevents you from monetizing a successful app. They do this by making it impossible to turn a free app into a paid app
A normal app goes through these stages: 1- Free. If it's good, it will receive good reviews, which will make it more popular 2- Paid
The first stage is an investment on the part of the developer. He invests his time and effort. But many developers only find out that it's impossible to move to (2) once it's too late (reviews can't be ported to new apps)
> Now a lot of programmers have started to see Apple as evil.
Well, not until 2016 when Apple began to offer ads in the AppStore -- as though that was a good thing for indy developers.
And a couple years later, when Spotify keeps getting pushed down the list: https://www.nytimes.com/interactive/2019/09/09/technology/ap...
This is why we're going to develop for Spotify's API first.
File along with Michael Dell's "shareholder value" remark.
I don't think Paul realizes how much it doesn't matter that Apple realizes how much it matters that it's broken. Or rather, that it doesn't matter that it's broken.
>>> Programmers don't use launch-fast-and-iterate out of laziness. They use it because it yields the best results. By obstructing that process, Apple is making them do bad work, and programmers hate that as much as Apple would.
I dunno, getting you app approved on the (jailbreak) cydia store, via like big boss back in the days was equally hard as the official appstore. This both frustrated and satisfied me; actual people reviewing the work and giving cryptic feedback haha. Is/was(?) the bigboss process inherited from the apple appstore?
It's gotten much worse, but why should the corp care? Shareholders are happy and the sheeple are mostly happy!!
Apple app store revenue from 2017-2020..
https://www.statista.com/statistics/296226/annual-apple-app-...
woody_harrelson_drying_his_tears_with_money.jpg
Interesting. Ben Thompson’s Aggregation Theory is more predictive of Apple’s domination of mobile applications than this. Perhaps this is just porn for developers because it makes us feel powerful. The truth, though, is that the developers go where the demand goes.
+1 for expressing the need for a “50% of the size of the MacBook Air“ machine. Oh, and with e-paper display, please.
Sorry. I should have known better than to try to participate in this thread. My apologies.
Carry on.
Reading the comments here look like the beginning of a tipping point.
Another genius
Absolutely right, aaaaand... it doesn't matter. People will do whatever is necessary to get their app in the user's hands. But Apple doesn't have Steve Jobs anymore and that matters a lot.
Anyhow, the question driving everything now is "Security?". How do you deliver non-dangerous software to innocent people? What is the (un)fashionably libertarian answer on this anyhow? Let them eat blackmail?
You all still buy their goofy aluminium laptops and publish on their store so I guess Apple wins.
(2009)
undefined
undefined
undefined
Yawn
This article doesn't hold up so well since Apple has a market cap > $1T. Please tell us how they're gonna fail now, Paul. This is like Ted Dziuba trying to convince the world in '09 that NOSQL was useless and would die soon. Pure genius
> Now a lot of programmers have started to see Apple as evil.
Not me (caveat: I worked at Apple). Clumsy is the word that comes to mind with regard to their handling of the AppStore. I see little or no evil intent
> Their fundamental problem is that they don't understand software.
That's a bold statement.
I'll be more nuanced. My impression is that Steve Jobs treated 3rd party software on "his devices" as a necessary inconvenience. He preferred the software he had a hand in, the software developed in house.
To me that is enough to explain Apple's approach to the AppStore.