Secret iPad | Hacker News Reader

Secret iPad

thinkfractional.blog

537 pointsfraXis82 comments

Even now, hearing about this app I think "Wow! That's so novel and cool! (magical, even)". I've said it before, but I dearly miss the creativity people had in the first few years of iOS. We have much better devices now; we should have better apps! I should be so desensitized by how interesting the new apps are that a skeuomorphic guitar tuner sounds banal and obvious! Instead today's apps are boring. They've been reduced to nothing but feeds to scroll through, clients for online services, and crappy games.

What in the world happened?

Edit: Some people are mentioning the quality productivity apps available, particularly on the iPad Pro. I didn't mean to suggest there aren't quality apps, I meant to suggest there aren't interesting apps that turn these devices into wondrous multi-tools. Apps that use the sensors in novel ways to interface with the real world; that make your phone or tablet more of a device and less of a screen.

I also want to say I didn't mean to imply that there are no quality games on mobile; it's just hard for them to survive (Apple Arcade is an interesting workaround for this), and also games in general just usually don't have the kind of novelty I'm talking about here.

It turns out you can make more money repeatedly selling $0.99 token packs for simple dot puzzle game powerups than one-time $1.99 or even $2.99 whole-app purchases. There are also more people who play silly puzzle games in waiting rooms around the world than people who play guitar.

There is still tons of innovation in the App Store. It’s just on Skinner Box mechanics now.

Yes, I think you've hit on the crux of it. Content streams, accessory apps to larger services, and games with micro-transactions all have one thing in common: they make way more money than you can typically hope to make off of a standalone, novel app that costs $1-$3.

This was the argument of why an Apple controlled app store was so important. But they seem to be actively pushing people towards those high profit, user hostile, apps.

There is a cohort of games that is 10x worse than anything you'd find on the web. Unskippable 30s blocks of ads every few minutes, that just promote more crappy games.

Apple introduced Apple Arcade because of the horrible state of games on the App Store.

I'm curious how well devs are doing in AA. The potential market is huge but you're also getting a tiny slice of $5/month. I see some of the same games on consoles for $20 each and up.

Apple is paying what amounts to an advance. Game developers are guaranteed a minimum.

https://gameanalytics.com/blog/apple-arcade-mobile-game-deve...

In their latest report, Apple announced that they want to help developers get set up on their service by paying for some of their costs, which should take some pressure off when building your game. How much they’ll contribute is still yet to be revealed, so maybe don’t hold your breath.

it's also a completely hostile and unnavigable app store. the search is bad, the charts are bad, the recommendations are bad, it's obscured on purpose.

The charts are really bad...

There are some fantastic iPad apps. Procreate and GoodNotes are two of my favorites. They are also insanely inexpensive.

I can't imagine making something as good as Procreate and then only charging $10 for it. If I had an idea for something complex on the iPad I would have serious reservations about developing it. You can't charge reasonable prices for software unless you do it in $0.99 chunks.

Procreate is the killer app of the Apple pencil and maybe even the iPad Pro. The pricing is indeed strange. There are people paying $999 for a tablet, $129 for the pencil, just to run a $10 app.

This is, of course, completely unsustainable. Savage Interactive has only been able to sustain this particular price point because it's presumably closely correlated to new iPad purchases. It's the most high-profile iPad visual art app, perennially promoted on the App Store, and so you can model it more or less as a tiny percentage of quarterly iPad revenue. That is a market position, in its vertical, with space for approximately one.

I don't have a solution for this problem. There has to be something in between a one-time $10 purchase and a $120 annual subscription, but Apple hasn't shown any interest in finding it.

Paid upgrades have worked well for other platforms but Apple stubbornly refuses to support this, instead trying to drive users into subscriptions they don't want.

I think it's too late now. Apple has taught an entire generation of users that mobile software is not worth paying real money for and effectively killed the ecosystem of apps they need to promote the iPad as a general purpose computing device.

Agenda notes app has elegantly solved this. Pay once and permanently unlock all current features. If a new release has a feature you want, you can purchase again.

It's more work for the developer but they did a great write up on how they built this capability: https://medium.com/@drewmccormack/a-cash-cow-is-on-the-agend...

Hmmh, on its own its not enough, I think. Just read through the post, and there is no trial of expensive features in their setup, just a set of free features to begin with. But it seems that Apple allows these days to trial features, as long as your app remains usable and useful without them.

Well, not apple to be fair. Developers are free to set their prices to whatever; it just so happens some developers set their prices to $0.99 and set that as the expectation

It's a race to the bottom. There's always somebody willing to go lower and the app store doesn't give you much room to stand out. Before I quit I had users complaining my app was overpriced at $1.

Unlike Wintel, Apple absorbs the lion’s share of profits in the iOS ecosystem, and seem to want apps to be free complements to their hardware.

Perhaps Apple Arcade for apps?

When you put it like that, yeah, mobile app pricing is messed up.

Compare that to desktop programs like Adobe Creative Cloud, which costs more than $500 a year. Or console games, where the console costs about $300 (Switch Lite) and a single game costs almost $100.

(All priced CAD, rounded.)

It all depends on perspective. From my point of view, Adobe's pricing is definitely messed up and I know a few people who are still using pre-CC versions because they don't want to pay for a subscription, especially if they only use Photoshop occasionaly.

On a similar note, I pay on average 5USD per (PC) game by tracking sales and would only consider paying more than 10USD for new games that I want to play really badly (e.g. DOOM Eternal). Console game pricing always seemed way too high to me, I'd spend what I saved on not building a PC in just a few AAA titles.

For an even more extreme perspective, I haven't paid for software of any sort in over 20 years, and I don't pirate either. I don't play games and if I want to edit a photo I figure out how to do it in gimp.

I've been busy the last two decades too. I've built websites, used CAD to design and machine dozens of physical objects, designed circuit boards, wrote and compiled software, made music, simulated chemical reactions, created and edited text documents, converted files from all sorts of formats to other formats, deployed databases with millions of rows, run two small businesses, and sent and received thousands of communications over myriad channels. All using free and open source software.

If there's something I want to do with a computer and I can't find a free solution that does it I write new code to accomplish my task (but honestly, this almost never happens these days).

I swear procreate was much more expensive than $10 when I bought it. I can’t find evidence of it being reduced, maybe my memory is faulty.

I’m pretty sure I paid $20 for it. Ridiculously good value either way.

> What in the world happened?

Business incentives changed.

The early iPhone/iPad app ecosystem focused on showcasing creativity. The UI was new and unique, and apps were created to take advantage of the gyroscope for games, or the audio input for apps like the guitar tuner thing in this article.

But that doesn't put food on the table, especially if the land baron is taking a 30% cut for every sale.

Why bother creating complex apps that push device capabilities to their limits, when what the majority of the market wants to do with their phones is tap away at a Candy Crush/Clash of Clans clone, or scroll mindlessly through one of 3 social media apps?

It simply became more lucrative to create clones of products where market demand had been validated, and stuff it full of ads and in-app purchases.

A lack of constraints mixed with a lack of real benefits.

Late-stage iPhoneOS was magical because of the constraints people were working with.

If you tell a developer, "Hey! Here's hardware with effectively no possible limits," they'll never write anything of value for it. A text editor and a way to do something you're already doing 5% faster.

If you tell a developer, "Hey! Here's hardware that's extremely slow and limited by today's standards, make something magical," they'll move the sun and stars to get you something that'll wow you.

Early-stage iOS was magical because of the new frontier offered; the constraints had mostly disappeared by then, but the ground was new and it presented what was effectively a new way to interface with the machine.

True, constraints promote creativity, the form of a sonnet or the rules of a fictional world or 8 bit graphics.

But in this particular case, they used the new strengths of the device that enabled things not possible before (removing constraints), yet thought their idea might be too much for it (hence the prototype), and their later idea was too much for it (needed a bigger screen). So their ideas didn't benefit from the constraints.

An upcoming test of your hypothesis is the PS5 100x SSD. Will it promote or stifle creativity?

The sense of the "magical" is the newness. Miracles, when commonplace, become commonplace. (Though I am still in awe at billion/sec pocket electro-computers... OTOH no one else is, it's just a cheap phone).

There really is no excuse for the lack of "real" software on the iPad (using "real" as in desktop quality). Only a small amount can be blamed on the system itself, appealing to a very niche audience (like XCode).

Arbitrary Javascript code written on the iPad can run, as evidenced by Play.js. Arbitrary Python code written on the iPad can run as evidenced by Pythonista. Arbitrary Swift code can run as evidenced by Swift Playgrounds. There's a Github app, there's a file explorer app, you can connect USB drives and keyboard and as of very recently, a mouse.

So where is VS Code for the iPad? Where is desktop-class software for the iPad? Why is this [1] article from 8 years ago still the best-in-class way to write code on an iPad? Is there still some kind of missing support I'm not seeing?

[1] https://yieldthought.com/post/12239282034/swapped-my-macbook...

> Arbitrary Swift code can run as evidenced by Swift Playgrounds.

Arbitrary Swift code which can only run because Apple grants themselves private entitlements that let it work.

There are a ton of great desktop quality apps on the iPad, I think the thing is that prosumer users don't know about them, and also people have been conditioned to think that any iOS app over $5 is expensive.

As one example, the music ecosystem is really good. There's Garageband and several great 3rd party DAWs, and a ton of apps which provide audio units that you can plug into, such as Model D for synths, or Stark for guitar amp modeling.

For photo/video, there are a few really great options: Affinity Photo is comparable to Photoshop, and LumaFusion is comparable to Final Cut or Premiere.

Agreed on the music ecosystem. The only thing that gets to me is the lack of keyboard shortcuts and being able to bind any midi cc to anything. You can't use midi to start and stop recording in Garageband for instance, which is something all desktop DAWs let you configure.

LumaFusion is good but I really wish iOS supported more video codecs than two delivery codecs (h264 and h265). I mean, they own ProRes. Cmon just add support for it in iOS.

I believe this is simply because people doing serious coding work always want to have a proper permanently attached keyboard. So they would use a MacBook Air with ARM CPU, but not a tablet.

It's definitely not about the hardware, given that the iPad has. mouse/trackpad support now. It's about the software.

Software is written for hardware. So long as most people don't use iPads with a mouse, few iPad apps will bother optimizing UX for it.

If that was the case no one would write code on a Surface, a laptop with an external keyboard, or even a desktop tower.

I do consider my desktop towers keyboard "permanent".

I don't swap keyboards, I always use that one.

Of course, by that standard, someone with an iPad with a keyboard attached (the ones that are part of a case/cover) could just as easily consider their keyboard "permanent".

Okay but how many people run iPads with keyboard attachments yet don't have a desktop or laptop that would be better for the purpose anyways? I'd wager few.

If that market were larger it would be catered to.

Sounds like a chicken and egg problem. No one does it because the software they need doesn’t exist. The software they need doesn’t exist because no one uses the iPad that way.

VS Code on desktop can run arbitrary processes, that’s how it can, say, have in integrated terminal. You cannot even remotely do that on iPad. Executing Python and JavaScript in a pre-compiled runtime environment[0], like Pythonista and Play.js do, is an entirely different approach and results in completely different products, ones like... Play.js and Pythonista.

The reason that VS Code is not on iPad is 100%, top to bottom that the framework and App Store restrictions prevent running arbitrary processes.

[0]: VS Code can run any binary on your system, iPad apps can only run binaries that they are compiled with. This distinction is why almost no existing development tools can be ported to iPad, almost all development tools are built on interacting with many binaries that the tool was not necessarily compiled with.

So “real software” is just development software?

No. Name desktop class software you want that isn’t available on an iPad and it’s just as valid. Name a profession who can’t do their work on an iPad and it’s just as valid. I just called our development software because this is a software development forum full of software developers.

Hell, why aren’t there more desktop class video games on the iPad? Not because it can’t run them, that’s for sure. The hardware is more than powerful enough.

Have you looked at the state of sound-processing and music-making apps? Drawing and painting apps? Animation apps? Those areas look somehow mature to me.

Yes, no good IDE because the target audience is wielding MBPs anyway.

There are no desktop class games because a touch screen and the control doesn’t translate well and people refuse to pay for premium games.

Gamepads have been available for the iPad since at least 2013.

We have over three decades of evidence in the console market that a game can’t be successful if it requires a non bundled accessory.

Guitar Hero [1] and Rock Band [2] were hugely successful [3] and required bigger and more expensive accessories than a simple gamepad.

[1] https://en.wikipedia.org/wiki/Guitar_Hero

[2] https://en.wikipedia.org/wiki/Rock_Band

[3] Guitar Hero III was the first game to reach $1 billion in sales https://arstechnica.com/gaming/2009/01/guitar-hero-iii-first...

The iPad Pro doesn't come with a pen or a keyboard but that doesn't prevent them from flying off the shelves anyway.

How many successful apps require a pen or keyboard?

Microsoft Office and Google Drive would be the big ones. No one's paying $799 to watch Netflix when a $250 iPad will give them the exact same experience.

And notice that neither Microsoft, Google, nor Adobe sell “iPad apps”. They sell a subscription that allows you to use their apps across all platforms.

I've never paid for the app versions of Office or Drive and have used them for years. You can still use them across multiple platforms free of charge.

Because many of us would rather game on a PC, or a console

There is a multitude of reasons for this. Beyond some limitations which still limit the ipad, it is the chick and an egg problem. You won't have real software on the iPad until there are users (buyers) for it and you won't have users until the software exists. This is especially important for corporate usage - at my work place I haven't heard of any iPads being provided to the employees, which means, no one can use an iPad and buy/develop software for one.

There are and were limitations preventing desktop class software on the iPad, some have lifted but it also takes quite a long time to take up on that, see also above.

- Keyboard support. For a lot of tasks, you need a keyboard. It took a long time until Apple sold keyboards for the iPad, and it still was quirky. There is a row of keys missing and even when coupling bluetooth keyboards, the extra keys didn't work. What Apple didn't really get was: a lot of people need an Esc key for their work, many would like funktion keys. It is still a struggle to run as simple as an ssh session on the iPad. Also, so far I haven't found a truely good SSH client on the iPad.

- Mouse support - really has been added only with 13.4, we will see how this develops.

Restrictions by Apple:

- For a long time, there was no real file exchange between apps, no joint file spaces. This has been partially fixed, still feels cumbersome and very much depends on the apps to get right.

- There is a ton of stuff that could be done on the iPad, but is banned outright. Termux is great on Android, not allowed on the iPad. The iPad would be great for development, if not mostly disallowed.

- And yes, for a lot of things, the screen is just too small :).

- The app store. Great and bad at the same time. The issues are probably best seen on macOS, because on macOS the app store is not mandatory. A lost of applications only exists outside the app store. While I liked the convenience of the app store and the fact that I don't have to deal with license keys and was free to use the app I bought across all my devices, the perpetual updates for free are condemning most longterm development. Paid upgrades were the life blood of the software industry for decades. They mean a regular revenue stream from power users like subscriptions, but didn't force the casual user to constantly pay, only when that user wanted the features of the newer versions. Subscriptions currently seem to be the only way, how a actively developed application can survive in that universe, but limits the user base.

- In app purchases. In principle a good idea, it lead to a reality where apps are flooding the app store which constantly try to extort money from the user. That games are allowed to charge in total more than lets say 2x what an AAA title would cost, is scandaleous. A model where an app could offer a free trial and then is purchased normally would be much more supportive of quality app development.

Many of this can be worked around, but unfortunately that is rather the exception. There are still great apps around which are a joy to use, but the whole ecosystem is drowned by all the bad stuff.

> Also, so far I haven't found a truely good SSH client on the iPad.

It might be worth checking out Prompt - supports some nice edge cases too, like SSH agent forwarding.

> - There is a ton of stuff that could be done on the iPad, but is banned outright. Termux is great on Android, not allowed on the iPad. The iPad would be great for development, if not mostly disallowed.

Check out iSH (free, available as a testflight, but also open source on GitHub) - https://github.com/tbodt/ish It's a full user mode x86 emulator, letting you run alpine Linux in a nice terminal, and install packages from it. I don't know to what extent iSH would address your desire for something like Termux.

I am reasonably happy with blink, that it supports mosh is a big bonus as iPadOS likes to kill background tasks, but the screen refresh is not fast (the output is html based) and it doesn't deal well with split windowing.

I am aware of iSH, but as you say, its requires "using" the testflight system and while that might be ok for enthusiasts, it is nothing for the large audience. Obviously , as testflight is limited to 10k users.

I would guess that there are more great apps published daily than ever before. The issue is probably discovery. Even if you build something great, you are competing for attention, and people who can pay for ads get the most attention.

Honestly, yes. I go through the App Store nearly every single day looking for something interesting and every single day it’s “have you tried Bear yet?” Or Day One, or Calm or Clash of Clans or some cookie cutter “Crossy Road” style nonsense.

Judging by what’s promoted on the App Store, there hasn’t been any new software written for iOS in years. They’ve just been republishing the same iOS Essentials article twice a day since 2015.

There's no money to be made in the app store, for the most part. OTOH, there's plenty to be made in monthly service revenue (and 30% of it isn't taken by apple in most cases).

The money thing is more a power law thing. There is DEFINITELY money to be made in the app store (I worked on an app that made a billion in revenue in seven months from a product with only twelve engineers directly working on it) but it’s almost all going to the winners in a specific category.

>>> I worked on an app that made a billion in revenue in seven months from a product with only twelve engineers directly working on it

uh? Can you indicate what? How many other apps have similar penetration?

He's mentioned working on Pokemon Go before, so I'd guess that.

Can you share the space your app was in out of curiosity?

Pokemon GO, it’s a game.

Yeah, that's a much better way of putting it.

You may be looking at things with rose colored glasses. The #1 app in the early days was something called Fart App.

Music production software is currently booming in the IOS world, there are professional producers using the iPad to make music e.g. Madlib announced that he made all beats for the album Bandana on his iPad

If you like this kind of behind the scenes, i highly recommend Ken Kocienda's book "Creative Selection" about the development of some parts of the first iPhone.

Totally agree; it's really fun to hear how things like autocorrect evolved over the development of the iPhone, and really how the designers and engineers thought about their work at the time. It's also nuts to think that so much of the iPhone software and experience was done inside of 18 months.

Thanks for reminding me, this is the perfect time to finally read that book.

It was a great read!

I wasn't there but I heard from the folks at Netflix who were there that they had a very similar experience getting Netflix to work on a pre-release iPad. Secret dark rooms, bulky cases so you couldn't tell the exact dimensions, cagey Apple reps with NDAs, very little time with the device, etc.

Here’s another post about a similar situation with an app which was on the first iPad’s sent to reporters:

    https://home.theodoregray.com/blog/2014/7/31/how-the-elements-came-to-be

    Discussion: https://news.ycombinator.com/item?id=22775624

So I guess the term on the NDA is 10 years?

I love these behind the scenes accounts of stuff.

I’m surprised they weren’t told about it at all in advance. Is that standard for Apple nowadays?

The author writes well. Slightly off-topic compared to the iPad piece but the story about his time in a glider in the Yorkshire dales is a cracking read:

https://thinkfractional.blog/a-day-to-remember/

Thanks for mentioning this. What a fantastic read.

Nice read, interesting how you can have apps already be successful before you even release them.

I still use GuitarTookit! Skeuomorphic design has fallen out of favor but I’m glad they haven’t changed it

Thank you for sharing it. It made for a captivating read.

Now _that_ was a fun read. Very well written!

Reply on news.ycombinator.com