Joe Hewitt on the iPad | Hacker News Reader

Joe Hewitt on the iPad

joehewitt.com

161 pointsnirmal66 comments

I almost didn't upvote this in deference for those who are (understandably) sick of the iPad discussion. But it represents my viewpoint quite succinctly.

I don't support how closed the Apple platform is, but I think the iPhone OS is great for a whole range of computing. On the positive side its created a market for developers, without pushing out indie developers (like consoles do).

I certainly have sympathy for the idea of teaching kids to program on these devices, and how hard Apple will no doubt make that. While its going to be hard to get any kind of interpreter on the device, I'm sure someone will succeed eventually - and then we could well have a generation of kids who grow up programming because of the device.

> While its going to be hard to get any kind of interpreter on the device, I'm sure someone will succeed eventually - and then we could well have a generation of kids who grow up programming because of the device.

Yes, someone will eventually succeed. There will be a jailbreak for this device which will turn it into a more interesting platform for students and technical users.

The complaint is that it should have been this way from the start, and we're betting on a flaw in Apple's design in order to get the most of their product.

I was thinking more of getting something past the app store approval process. I think if it was obvious it couldn't be used to circumvent the store itself it might have a chance. Jailbreaking will happen sooner rather than later I'm sure.

Why would a jailbreak be needed? What am I missing?

An iPad developer could compile up copies of some of life's goodies like bash, emacs, Perl, Ruby, Python, some lisp variants, Apache, etc.

Then the developer write a little Mac-application which let the user apply for the Apple Developer program. When that is granted, the application installs the developer tools and downloads the iPad projects for emacs, etc.

Every user which use scripting applications will be his/her own developer. Without jailbreaks.

I'm pretty sure you have to pay $100/year to sign up to the iPhone dev program and I imagine the iPad dev program will be the same. So your idea works fine except for the part where everybody who wants to install your package will first have to pay Apple $100. I'm not sure how many people will be willing to do that just to start learning about programming.

>>I'm not sure how many people will be willing to [pay a tax of $100/year] just to start learning about programming.

Huh, learning? That is like saying people won't buy a car so they can learn to drive... :-)

A user could have an iPad to write and/or run suitable scripting programs. Like emacs. [Edit: Or other programs written with Perl/Python/Ruby/Lisp etc.]

If that is worth a tax of 8-9$/month will vary between users.

> If that is worth a tax of 8-9$/month will vary between users.

And between teenagers who may or may not have access to dad's credit card, or between countries who may or may not have access to that program ;)

This also means that a user won't be able to install some software with a single click ... he'll have to first go through the registration process, with a valid credit-card, and some time on his hands.

I really don't know why people try to downplay this. If this model gets popular (and it already is to some extent because of the iPhone) ... devs and many regular users simply get proper fucked.

>> This also means that a user won't be able to install some software with a single click

>> ... many regular users simply get proper fucked.

My point was to suggest automation of the process. I didn't imply that this would be painless, just less painful than jailbreaks.

If it becomes common, even Steve Jobs will have to give in and have a simple checkbox for allowing scripting apps. (Like different sets of installable applications in Debian/Ubuntu).

If I recall you can't use any scripted languages on the ipod as part of the developers license. Has that changed?

You can't distribute any scripting language interpreter (and thus no apps written in a scripting language) via the app store. However there is no reason why you cannot install a scripting language and use it on your own phone if you have the dev tools.

Shouldn't be to hard to make an HTML5 based programming tutorial or programming environment that exposes javascript via eval and saves the resulting snippets in the local storage...

If we leave aside the downsides, we are left with two major things here:

- Simplified computer that the majority can handle fully on their own without someone's help. They can choose what apps they want, get rid of ones they don't want, etc. It does take the stress & overhead away from installing stuff on your machine. That means people will install stuff.

- Assumed location awareness, connectivity, camera (eventually), mic, headphones, accelerometer, ambient light sensor along with the UI is a genuinely new paradigm

Those two things together have the potential to be a very big deal.

It's a big deal because the iPod had all of this but its size meant the number of apps you could write lived in a small petri dish. The size of the iPad's petri dish is much larger, when compared to the iPhone.

For example as they showed: Keynote and iBooks.

And for us to design: Electronic Restaurant menus, interactive car manuals, Medical Industry, mapping/aviation, un-maned drone control (planes, vehicles, robots), and so on.

The above list are applications that neither the iPhone or a laptop can do. We are entrepreneur we should see this, we should not be the ones complaining iPad sucks because x or y.

I don't believe it has GPS, so not so sure about the locational awareness bit.

If I'm not wrong, I believe the 3g version has A-GPS.

You are correct: http://www.apple.com/ipad/specs/ (left-most column under "Location")

It looks like both the wifi and 3g versions have a digital compass, as well.

Wow. I never even considered that they would differ on something like that.

That's a substantial hit on the "it just works" front. Some apps will not be able to do anything useful without this. Some people won't understand why.

The reason why is that Qualcomm and most 3G chip manufacturers bundle GPS, as it is required by federal law for 911 service. Just adding a standalone GPS module to the Wifi version would probably cost about as much as adding a 3G chip.

Btw, does anyone else find it striking how much patent licensing adds to the cost of a device? I'm pretty sure Apple is passing the $130 3G upgrade along at almost cost. These 3G chips have been out for years now and probably only cost about $10-20 to manufacture, yet every 3G cellphone manufacturer gets raped by patent licensing costs to the tune of $100+.

Given that Ericsson F3507g mini-pcie card is $80 give or take, I doubt that the $130 3G upgrade is at cost.

Does the Ericsson card have a GPS radio and quad-band G3 as well, or just 3G?

This is already the case with the iPhone vs. iPod Touch.

  Digital compass
  Assisted GPS (Wi-Fi + 3G model)

Looks like both models have A-GPS

They refer to the 3G model as the "Wi-Fi + 3G Model", so I believe that line item indicates that only the 3G version has GPS.

Ah, that makes sense now. Thanks!

Perhaps not in this 1st gen version, but if it proves to be a success then just give it (and the inevitable competing tablets) some time before it does.

Also, the AppStore means never having to worry about malware or viruses. That means no hesitation, if you see something you like, just get it, it's just a single click.

Fear of malware, viruses, or just accidentally changing some setting can be terribly paralyzing.

> Also, the AppStore means never having to worry about malware or viruses.

I've heard this idea over and over these past few days, and it really puzzles me. Do we really believe that the same App Store reviewers who are so incompetent that they can an e-reader because you could use it to read the Kama Sutra (or whatever your favourite ridiculous rejection of the day is) are, on the other hand, so competent that it is impossible to get malware by them? Or do we just believe that the iPhone/iPad/whatever is naturally resistant to malware, however cleverly written --in which case it doesn't matter what we install on it?

Has a single malware or virus made it passed them? Please provide sources for this if you have anything but a hunch.

Is it really the case that the burden of proof is on someone who thinks that a system might be vulnerable? Should we believe by default in the security of new cryptosystems because they haven't been cracked yet?

I am neither asserting nor implying that there is malware anywhere in the iPhone ecosystem at any point. I meant merely that

1. the reviewers wouldn't, and couldn't be expected to, catch it, unless it were very clumsy (do you check the sources of all the software you run? Even if you do, do you think that the reviewer does? Now consider how many more apps a reviewer must vet daily than you run);

and

2. that there are much cleverer things to do with malware than loudly announce your presence, such as quietly funnelling out interesting personal data, so that it is very dangerous to confuse a lack of reports of malware with a lack of malware.

To the last point it may rightly be objected that arguing something on the basis of “If it were there you wouldn't see it, and you don't see it, so it's there” is indulging in a particularly base sort of logical fallacy (identification of a statement with its converse). I emphasise again that I am in no way claiming that there is malware; merely that the belief that there isn't, can't be, and won't be as long as we stick with the closed model seems to represent an indulgence in a peculiar sort of optimism.

I think the burden of proof lies with the person claiming things that nobody has seen does in fact exist.

I am not making any claims about the actual, present presence of malware. My only claim is that it is unduly optimistic to believe that a vetting process offers a complete immunity from malware.

Maybe I mis-interpreted your position, and you meant only that malware is less likely in a ‘gated community’ such as that offered by the iPhone and pals. If that is what you mean, then I agree.

If you would like a more grandiose claim, then I am willing to strengthen it slightly by making the definite claim that, some time between now and the heat death of the universe, there will be malware on the iPhone. However, this statement, being (practically) unfalsifiable, isn't worth much, which is why I didn't make it originally.

It's not the reviewers, it's the fact that there is a paper trail between the malware author and the malware.

That's a good point.

The paranoid in me wonders whether a malware writer might then work on infecting an app-store developer's machine, to get the code in via such indirection; but I agree that seems like an awful lot of work when so many unsecured machines are out there for easy and anonymous direct picking.

It'll probably happen eventually, and I'm sure there are several app-store developers whose machines are already members of botnets.

Very good point - especially for the first one. They won't need someone's help. Mainly, our help.

One could say Steve Jobs has "taken" the market from us, now controlling even the developers. Yet I believe, as the post pointed out, that theres an escape hatch through using the internet as a delivery mechanism. HTML5 and responsive web apps are truly the future there. Learn UNIX programming if you haven't.

This moves the developers, from say, GeekSquad and "Best Buy Optimization" rip-off services to productive application development. One could argue that this will make the smart smarter (as the barrier to entry is higher) and the "dumb" (less experieced a better word?) interested in something else.

I think the iPad is "meh" because to some degree we haven't seen what it will do to the landscape yet. We've been focusing on the technical aspects of the device - and putting up against a rubric we've built over the last 30 years. I believe Jobs aims to change that rubric.

Other people have said it - computers are complicated for laymen users and they want it to act like an appliance. Hes giving it to them.

From developing for the iPhone the most notable thing I have learned is that the relationship between UI ease of use and UI complexity is very different than for a desktop. A few touches and a swipe on the iPhone becomes a few clicks and a drag on the desk top. By increasing the screen size available for touch based UIs I think the iPad and its successors will eventually replace laptop and desktop computers for most people.

I commented on a previous post that I wouldn't read another iPad article, because all the ones I had read were tripe that bludgeoned the device for ideological reasons, or praised it because it's shiny.

However, I respect Joe Hewitt's iPhone work and other writings, so I checked out his article, and I finally got some real insight. This is certainly the best iPad article I've read so far. The idea that we need to re-imagine all of our current software with the capabilities of a large, responsive touchscreen is a good take on the release.

The only thing that worries me is the risk... what's the chance people don't buy these things? Apple has had a flop or two in its past, and even though I develop iPhone apps, I'm a bit leery of developing iPad apps.

Risk/Reward

By getting in early you have a chance to get to know the device early, you get to have apps out there with less competition, you have the chance to pioneer something and the buzz that comes with it, et. etc.

That weighs against the risks: noone will buy it, something big will change between v1 & v1.1, etc. etc.

Now, some people will buy it and the app store will not have fewer than 1000 apps when it launches, so neither the risk nor the reward are huge. You need to work out which way it leans.

During the keynote, they presented the iPad as "magical and revolutionary". Of course, this is marketing, but I was thinking about it to try to see what could be so magical and revolutionary about it.

For the magical part, sure it's a wild hyperbole. But I suppose you could say that, in that it feels like it's the real beginning of the kind of interfaces that have been imagined and dreamt of for many years on TV, in movies… and considering the size of it, it is pretty amazing.

As far as revolutionary, my first impression was "how is that revolutionary since everything it does an iPod Touch has been doing for a couple of years on a smaller screen?". But I came to a similar conclusion as Joe Hewitt, in that it's revolutionary (though it's also hyperbolic) because it shifts the paradigms of computing and the way we think about computers. Yes, the iPhone/iPod Touch are similar technically but the apps were still apps for an advanced mobile phone and the size of the screen was in fact very limiting.

In a way, I feel that the iPad is what computers should have been from the start if it had been possible at the time. For example, take the mouse. It was a great invention to interact "directly" with what's on the screen, but if it had been possible to have touchscreens then, I doubt mice would have been used. Applications have always been a bunch of files in the filesystem and we got used to that, but really that's something we don't really need to know and something that Apple has been removing for quite some time on the Mac even, with app bundles.

So, of course, we'll see what happens. But I think it's an interesting step.

PS: yes, I think I drank too much kool-aid

I still think that for most office uses, a mouse is better than a touch screen. If I'm spending a lot of time entering data, a all-in-one touch screen device is miles behind a monitor at eye level several feet away, a good tactile keyboard flat on my desk, and a mouse a few inches away. Maybe I'm constrained by what I already know, but I can't picture a touch device being better.

From a developer's point of view, how does the power of Android compare to that of the iPhone OS? I haven't developed on either, but articles I've read so far seem to indicate the two have pros and cons but are pretty evenly matched overall. Of course the iPhone still attracts more developers because it has more users.

Let's assume that Android-based smartphones become increasingly successful as many predict, and that users find Android smartphones about as appealing as iPhones. Tablets running Android have been announced, though no apps have been optimized for the larger screen size yet. If Google released tablet-optimized versions of the stock apps, and made the appropriate tweaks to the SDK and the Market, wouldn't Android-based tablets be a close match to the iPad?

How come there hasn't been any excitement about the potential of Android-based tablets?

I can't claim to be an Android expert but coding Java in Eclipse is a big pain in the butt. Some people love Eclipse. I am not one of those people.

With respect to iPhone/iPad development, Cocoa is the most beautiful and consistent framework I have worked with. It's the best implementation of MVC I've worked with. Objective-C is a little crufty but message passing languages are so much more elegant than method calling languages like C++/C#/Java.

With respect to Android based tablets, Google has done a huge disservice and fragmented the market by pushing Chrome OS for tablets.

I think the disservice was more that Chrome OS is locked-down Linux vs scaled-up Android.

It should have been Android all the way up -- handhelds to desktops -- freeing the market to throw all sorts of crazy hardware variants into any number of form factors.

I've commented here before about the shittiness of Android as a platform: http://news.ycombinator.com/item?id=1064415

I think Android is a huge mistake on an unpocketable portable device, and I don't think Google disagrees. It might work out on a TV / Media Center device, but Boxee is so far ahead there that it's quite unlikely.

Google's plan for Smartbooks and iPad competitors is ChromeOS. I think they'll succeed handily if they ship it with only a content-addressed-filesystem exposed (think 'media picker'), and with a fleshed-out Google Native Client for local apps. I'd love for them to get some competent hardware manufacturers lined up with solid build quality, but Apple has that market mostly cornered (the Nokia netbook looks great though).

I think Android is a huge mistake on an unpocketable portable device

Why? Or, more, importantly if iPhone OS work why doesn't Android work?

Probably because all the Android tablets that have come out so far have been pretty terrible. The Archos device got slammed pretty hard. I haven't used it myself but checking out the specs it's really just not in the same ball park. Slower, smaller, lower quality display, stuck on a year old version of the Android OS -- in fact it got the 1.6 update 8 days ago and it's fragmented off into it's own Archos App Store/Market/Catalog/whatever. Not encouraging. In fairness the Archos device is primarily a PMP and has some unique features that might make it a better choice for some people. This seems to be the direction most of these Android tablets are taking. I don't know of any tablet running Android that is directly comparable to the iPad.

As a developer for Android devices, i must say this article makes me really sad. How can a developer be satisfied with limitations and just put the "but it's how it is to be more secure" stamp on it.

Let's see those 2 sentences that make really no sense: "it can't just read and write willy-nilly to the file system, access hardware, or interfere with other apps. In my mind, this is one of the best features of the OS." and: "As a developer, it's a bit sad losing the ability to come up with crazy plugins and daemons and system-level utilities, but I believe it's a tradeoff worth making."

I've now owned Android devices since the G1 (the first), now switching to the Nexus One. The nerd i am i probably installed hundreds of those apps, that are not approved by someone. Not one of them did somehow brick my phone! The worst thing that happens is that apps don't do what i want, some even crash (but that's surprisingly rare). But still, they can run in the background, communicate with each other, etc.

So, why am i telling this: If smart engineers can think about what a smartphone OS should be like, it will support save mechanisms for multitasking, for daemons and plugins. Behind Androids Interface are a lot of very neat concepts working.(intents/broadcasts, man, i love that concept!)

From what i can tell:

- iPhone OS is some stripped down Mac OS version, which probably has its deficits in some areas. I believe that Apple just doesn't give us multitasking because it has heavy impact on batterylife or security (because the OS wasn't made for mobiles, in the beginning).

- When chosing between a _free_ development platform on Java, with Eclipse on every major platform and a subscription based development only with Mac OSX and a language that is basically used nowhere else (Obj-C)... well, the choice is clear for me. Google did a pretty good job with the Eclipse integration and it's toolchain.

Cydia, the app store for jailbroken iPhones has a backgrounder app. Give it a try. What they don't tell you is that it sucks, maybe because the iPhone doesn't really have the ram to run apps in the background or maybe because the iPhone OS wasn't designed that way, but for some reason I'm leaning towards the former as an explanation.

I found the Backgrounder app to be a great way to multitask on iPhone OS. First of all, it gives you a visual indication on the icon of every currently running app, including the Apple ones that are always running like Phone, Mail, Safari, and iPod. Second, it's very intuitive and easy to use, you simply hold the home button down for 1 second instead of quickly pressing it, if you want to background the app instead of just exiting to the home screen. Third, it passes the number one use case very handily: I can listen to Pandora while surfing in Safari seamlessly, and it almost makes my iPhone seem like a functional computer.

iPhone 3GS with backgrounder is seriously a powerful device, and lets me have an insight into how powerful the iPad will be.

Don't get me wrong, I'm not a fanboy. I won't be buying an iPad until I'm sure it can be jailbroken, and probably not rev. 1, because I'm sure rev. 2 will have the missing features like webcam, etc. I'm just saying that a jailbroken iPhone/iPad will be an impressive device.

As a developer for Android devices, i must say this article makes me really sad. How can a developer be satisfied with limitations and just put the "but it's how it is to be more secure" stamp on it.

I find it ironic that an Android developer is saying this. Android only has what, 256MB of available disk space for 3rd party programs? That is extremely limiting for all kinds of useful apps that might need a local database, or for large apps like games.

Also, the Android marketplace has similar limitations to the App store, but none of the safety. Google can and will pull your app if it doesn't like it, however, their post-release approval process lets malicious apps get through. There's already been one phishing app released. It seems like Google has all of the bad features of an open PC (viruses, worms, malware) and all of the bad features of a sandbox (complete control over code execution, unless you root your device).

Don't downmod me because you disagree. Reply and tell me why I am wrong.

The 256 MB (or whatever) is not Android limitation, it is limitation of specific device. Large applications can put data on SD card, where they are limited by the card size.

I have the ADP1, which has around 70 MB partition for applications, yet I use navigation app, which has 2.8 GB worth of data (entire EU + few surrounding countries).

Well, you should get your facts straight first. "Android only has what, 256MB of available disk space for 3rd party programs?"

That's nonsense. It depends on the device. Also, the G1 has far less then 256 MB free for apps, yet a typical app is around 100kB large. Also data can be on the SD Card and future version of android will be able to install apps on the sd card as well (maybe even the current version, don't know).

"Google can and will pull your app if it doesn't like it"

Where is the proof for that? Afaik they still have to show the behavior Apple is showing constantly by pulling normal apps. Guess what, you can easily install and use another market application and Google won't forbid you to do so: http://slideme.org/ is a good example. You can install it easily on an _unrooted_ device like other applications.

I agree that the android market could use some approval process for applications that need security (like banking apps). Something like the SSL-Info in your browser. But that's about it. Also could you please point me to any viruses and worms you are mentioning?

By the way, what do you mean by "complete control over code execution, unless you root your device"? who is controling what exactly? If you mean native code (i suppose you mean the sandbox is the dalvik vm), i should point you to http://developer.android.com/sdk/ndk/1.5_r1/index.html .

> How come there hasn't been any excitement about the potential of Android-based tablets?

In the propaganda video Apple has on their site there is a brief talk by the tech guy. Incidentally he is the only guy in the video who doesn't come across to me as a total douche-bag when talking about the iPad.

He mentions that Apple owns the hardware and the software. Usually that is just Apple marketing BS but in this case I think it actually really applies. There is a tight integration between all of the hardware, software and even the uni-body aluminum casing. The decisions they've made for everything (from chip design to UI design) is with the entire device in mind.

Android, not so much. It must necessarily be a more generalized framework. There are some reasons that is a benefit but for this type of device I wager it is actually a detriment.

The main positive the media is taking away from the iPad is its speed. I was impressed by the claimed battery life. I don't believe any Android tablet will be capable of delivering a similar level of performance on these two key factors.

"The store may not be open, but the iPhone/iPad platform itself could hardly be more open to tinkerers of all ages."

Considering it as an Apple product in the vein of the iPhone, with all the limitations that go along with that? Yes. But considering it as a computer? No.

"...could hardly be more open...?" Except that you have to pay $100. Oh and you have to develop on a Mac, so you should probably buy one of those too.

$100 is only required if you want to publish your apps on Apple's App Store. If you just want to tinker you can build the apps for yourself and let friends use them also. I don't recall what the limitations are on that. As for the requirement to own a computer to develop for it -- yeah... that's common.

Not true. Apple requires you to sign your application before the iPhone OS will allow it to be installed. Also, the check is on the device itself, not in iTunes.

There's only two ways to get the apps installed: 1) Pay $100. 2) Crack the device.

Reference and instruction via networkpx: http://networkpx.blogspot.com/2009/09/compiling-iphoneos-31-...

One of the few iPad articles that I actually enjoyed reading. Facebook on the iPhone was a real reinvention of UI. Something we don't see often enough.

I think the iPad is good for what it is - a personal media player. It is NOT the laptop/touchscreen hybrid next generation technology that we had hoped for. I suspect as a PMP it will see wide success, but there is still a massive void yet to be filled with a proper touchscreen laptop that can accommodate all of our computing and graphic needs. We can only hope another company steps up to the plate and delivers something truly revolutionary. Not just another iPhone/PMP.

I have a hard time accepting his new found love for Apple when in the article before last, he was ranting about the evils of Apple's app store approval process. He claimed that he would no longer be making iPhone apps, and would make web applications instead. Now, he says that the glorified iPod touch is going force everyone to "re-imagine" their applications. Sorry, but I'll remain skeptical thank you very much.

> The store may not be open, but the iPhone/iPad platform itself could hardly be more open to tinkerers of all ages.

Really? Can I install an iPhone/iPad IDE and emulator on my Ubuntu workstation?

iPad, as a device is a new opportunity for developers. But not an incredible one. Open up the platform, and we can consider it incredible.

He calls the demand for open-computing and removal of restrictions a "nonsense" .

I'm not sure if you are familiar with Joe Hewitt's backstory, but he's been very vocally against the app store policies. In fact, he left the Facebook iPhone application team over this:

"My decision to stop iPhone development has had everything to do with Apple’s policies.” – Joe Hewitt"

ref: http://www.techcrunch.com/2009/11/11/joe-hewitt-developer-of...

Quoting him & disagreeing with him calling iPad platform an "incredible opportunity" - gets me downmodded.

I laughed a little.

Enough with these iPad stories, please?

Reply on news.ycombinator.com