conormulligan.com

Useful OAuth Sign-in Controllers from Google. Still, I wish there was a better way to authenticate than throwing up a web view, it's messy as hell on an iPhone.

The introduction to Apple's published App Store Review Guidlines (requires a paid-up developer account) is kind of hilarious. The highlights:

We have lots of kids downloading lots of apps, and parental controls don't work unless the parents set them up (many don't). So know that we're keeping an eye out for the kids.

Bullshit.

We don't need any more Fart apps.

Someone tell that to Phillip Shoemaker!

We have lots of serious developers who don't want their quality Apps to be surrounded by amateur hour.

Really? If apps suck they won't gain any traction and no-one should ever see them; I can't see how this is a problem at all.

Apple has revised the rules by which submitted apps much abide:

In particular, we are relaxing all restrictions on the development tools used to create iOS apps, as long as the resulting apps do not download any code. This should give developers the flexibility they want, while preserving the security we need.

The changes affect the notorious section 3.3.1, along with sections 3.3.2 and 3.3.9. This isn't actually a huge shift — it mainly restores the status quo pre-Flash CS5 — but it does make things a little easier for developers using frameworks like Unity.

Between other projects I (finally!) found the time to update the site. The only visible change is the much more minimalist theme that should look a lot better in crap browsers and mobile devices. The backend is where most of the work went though; I gutted out the old framework which was more or less a patch on Google's webapp framework and replaced it with my own WSGI-based framework. The interfaces between the two are radically different, so this involved tediously updating each request handler.

I also started making use of App Engine's cooler features like the blobstore (which has a miserable API¹) and task queue (which has a great API). Almost everything is memcached now so page loading should be much snappier, unless your request happens to spin up a new instance.

There's a little more to come, mainly on the JavaScript front: look out for JSON-fueled asynchronous pagination and forms soon.

1. It's forgivably bad though, given the environment constraints. Performance is good too.

On another UIWebView note, Matt Gallagher has a typically excellent post on substituting local data for remote resources using NSURLCache.

This is really cool; dynamic image scaling for mobile devices:

If you're designing web sites for mobile, you need to make sure your graphics work on thousands of differently-sized screens. Let tinySrc take care of the hard work.

http://i.tinysrc.mobi/http://mysite.com/myimage.png

(Yup, that's it.)

I've already played around with this by automagically modifying the href attribute of each anchor element in a UIWebView document; works like a charm.

John Cole is the latest in a long list of people to be screwed by Paypal's crazy verification policies.

Great new Financial Reports module on iTunes Connect.

Apple will hold their Autumnal, musically-inclined event on the 1^st of September; I expect we'll see the revamped Apple TV, iPods with Retina displays and cameras, and hopefully some cloud-based services like iTunes Live. Also making an appearance: incredible sales numbers for the iPad and iPhone 4, featuring unnecessary hyperbole.

Apple have introduced a nifty program where developers can sell their apps to schools and colleges in bulk for a 50% discount:

The App Store Volume Purchase Program makes it easy for education institutions to purchase multiple copies of your app.

You can choose to offer special pricing that is 50% of your list price to education institutions when they purchase 20 or more copies of your app. You have the option to apply this special pricing to all of your applications or only individual applications.

It's a great idea, although I wish they'd give developers a little more flexibility in setting the discount rate. Also, I wonder how they're going to determine what counts as an education institution; typically, Apple use university email addresses for verification but this would be fairly cumbersome for primary and secondary level institutions.

For the last couple of months I’ve been putting together a Stack Exchange client for iPhone using the recently released API. The app, Stack Up, is on sale for $0.99 until the 8th of August; after that, you can get it for $1.99.

For the more information check out the project page, the dedicated site, or the stackapps page.

Apple have seeded the Xcode 4 preview shown off at WWDC to developers, and man is it a big update. No word on when the final version will ship, but I expect it'll shown up in November with iOS 4.1.

Matt Drance takes an interesting perspective on the iPhone 4 debacle:

Whether you are the CEO of the second largest corporation in the U.S. or a general in the U.S. military, it should be no surprise when the press reports what it sees and hears. If a media outlet reports garbage, then all things being equal, its reputation should pay a price. If it has information that it believes to be both material and credible to a relevant topic, it is not obligated to sugarcoat that information, but to report it.

Antennagate is news exactly because Apple has hit so many home runs—and, by the way, received countless glowing headlines to match. Apple should be nervous when this sort of thing isn’t news: it would mean nobody cares anymore.

As I said earlier, I think the press conference was by and large a success; however, Apple will soon run out of goodwill if they continue to respond to criticism the way they did before Friday.

So, Apple will give out free bumpers to iPhone 4 owners, but only until the 30^th of September, which strikes me as a cheeky but ultimately cost-effective way of dealing with the problem¹. As expected, the New York Times article describing a second software fix was bogus. I was pleasantly surprised by the tone Jobs took, which was notably less combative and much more conciliatory than I expected. In fact, it was downright humble by Cupertino standards which should go along way to sowing some goodwill after the initial "quit holding it wrong" response. Overall, I think it was a very effective presentation by Jobs, and I expect the mania to cool dramatically over the weekend.

1. I think there's a good chance Apple will continue to ship bumpers with new iPhones after September if they haven't come up with a better solution; Jobs seemed to suggest as much in the post-presentation Q&A.

Every single time the New York Times quotes an anonymous source, they use the exact same phrase somewhere in the text: "agreed to speak on the condition of anonymity because he was not authorized to discuss the matter." It's difficult to put into words how annoying this is.

So says the New York Times:

One person with direct knowledge of the phone’s design said Thursday that the iPhone 4 exposed a longstanding weakness in the basic communications software inside Apple’s phones and that the reception problems were not caused by an isolated hardware flaw.

...

The person said the problems were longstanding but had been exposed by the design of the iPhone 4.

I find this hard to believe. An exposed antenna which attenuates when bridged by one's hand is something that really can't be fixed by software¹, no matter how often Apple says so.

1. Unless of course I'm completely missing something, but this just strikes me as implausible no matter which way you look at it.

I couldn't find any decent high-resolution arrow icons on the web suitable for use in a UINavigationBar or UIToolbar, so I put my limited Photoshop skills to the test and created my own. There are eight icons in total: four directions in both standard and high-definition sizes. They're basically a recreation of the up and down arrows found in Mail.app and the forward and back icons found in Mobile Safari. The images are shared under a Creative Commons license, so you're more or less free to use them however you see fit.

You can grab all eight icons here. Licensed under a Creative Commons Attribution 3.0 License.

John Gruber thinks something big is going down tomorrow. It's difficult to tell, but I will say this: a press conference to simply push back against industry criticism is not something I would have expected from Apple, who normally only take the stage to announce a product.

A beta version of iOS 4.1 was seeded to developers today (you'll need an active iPhone Developer account to follow the link), which includes, among other things, the software side of Apple's response to the iPhone 4's reception woes. Apple are also holding a press conference this Friday, presumably to try and put a lid on the whole thing once and for all. I still think it's unlikely they'll go any further than this weak software fix, although there's probably still a slim chance they'll start shipping free bumpers.

I know I sound like a broken record, but I'll say it again: Apple's biggest weakness is in cloud-based services. They do offer MobileMe, but for €80 a year and with a limited set of features¹, most of which are available for free elsewhere. Obviously Google were always going to excel in the cloud so Android outclassing iOS in this area is to be expected, but now Microsoft will offer a free package of tools, predictably dubbed Windows Phone Live, with a panoply of functionality for devices running Windows Phone 7:

A feature we’re discussing for the first time today is the new Windows Phone Live companion site that gives people a central place to see pictures they’ve published, view their Windows Live calendar and contacts, exchange OneNote files and access other information shared between the phone and the Web. The site will offer 25GB of SkyDrive and host the Find My Phone service, which allows people to find and manage a missing phone with map, ring, lock and erase capabilities right from your PC – and all for free.

It looks pretty competitive, certainly better than what either Apple or Google are offering at the moment. It goes without saying that you shouldn't judge a product based on a spec-sheet taken off of a Microsoft website, and it could end up being a UI disaster or plastered with ads like Hotmail, but at the moment it's good stuff.

As John Gruber said a few months ago, Apple doesn't do loss-leaders so I don't really expect them to open up the full MobileMe suite to everyone for nothing. But they certainly could offer some functionality like remote-wipe for all users. Surely they could factor the cost of providing this service into each iPhone, knowing it will only be usable by iPhone owners. In fact, MobileMe may already be heading in this direction; the new account status that began popping up before WWDC would certainly suggest Apple are at least toying with the idea of a tiered pricing structure.

1. Highly polished features, it must be said.

Apple will eventually have to address the reception issue again; once they ship the somewhat silly software update, and all that happens is iPhone users see less reception than before, I can't really believe the dissatisfaction will simply evaporate. They've done themselves no favours by pointing the finger at an algorithm; it only makes it more difficult for them to actually address the hardware issue.

There are no good solutions, at all, from Apple's point of view. A recall is simply out of the question, Apple would never go for it; I think they'd gladly weather the PR storm and hope users start holding the iPhone the right way. I also think they're unlikely to rev the hardware; the backlash would be immense and they'd probably have to accommodate customers wanting to trade in their current iPhone for a revised model. One simple and relatively cost-effective solution would be to ship the currently outrageously over-priced bumpers with new iPhones, for free. This too seems unlikely for a couple of reasons: 1) Apple said they wouldn't, and if they figured it would make the complaints go away they probably wouldn't have come out so forcefully against it, 2) It would be admitting that there really is a hardware issue, something Apple clearly does not want to do, without actually addressing it which creates problems for them down the road, and 3) Gizmodo have been pimping the idea for weeks now, and if Gizmodo wants it chances are Apple won't do it.

Anyway, one thing is clear: fairly or unfairly, the perception of the iPhone being a crappy phone but an otherwise great device is pervasive and Apple needs to come up with a real, non-diversionary solution and end the "there is no problem" theatrics if they want to combat this trend.

It can be difficult to acknowledge that a product you otherwise love has a couple of significant flaws. That said, the white-washing of the iPhone 4's reception issues by the company and, much more regrettably, some writers is still disappointing. So, it's extremely refreshing to hear Marco Arment say this:

Apple’s arrogance and indifference in issuing this response is offensive, insulting, and disappointing. It’s as if they’re expecting this issue to go away if they just wait long enough and ship enough iPhones. But it won’t. It’s only going to get worse as more people try to exchange their iPhones at the Genius Bar for these two issues, thinking it’s just a problem with their iPhones, and encounter the same problems with every replacement.

Read the whole thing.

It's an interesting idea, although the only way it can possibly succeed in driving development on Android is if it functions effectively as a gateway drug to more powerful tools. The next logical step is to facilitate apps written in dynamic languages; a fairly undaunting task given Android's architecture. Learning basic Ruby or Python would be an obvious progression for Inventor veterans, even if only to inject more complex logic into Inventor-based apps.

Adding high-resolution artwork to an iOS 4 project is extraordinarily easy; all you need to do is add the new files to your project in a particular format (imagename@2x.png alongside an existing imagename.png), and UIImage will automatically load the appropriate file depending on the screen scale.

However, if you're loading images remotely, you'll need to do some of the work yourself. Here's an example of how to fetch a gravatar portrait depending on the device resolution:

float size = 40.f;
if ([[[UIDevice currentDevice] systemVersion] floatValue] >= 4.0) {
    size = size * [[UIScreen mainScreen] scale];
}

NSString *url = [NSString stringWithFormat:@"http://www.gravatar.com/avatar/%@?s=%i", emailHash, size];

NSURLRequest *request = [NSURLRequest requestWithURL:[NSURL URLWithString:url]];
NSURLConnection *connection = [[NSURLConnection alloc] initWithRequest:request delegate:self];

// implement the NSURLConnectionDelegate methods,
// and use the resulting data in [UIImage initWithData:data];

As you can see, it's relatively straightforward: just grab the screen scale from [[UIScreen mainScreen] scale] and use it to determine the resolution of the image you want to download.

So, Apple have published a press release addressing the infamous iPhone 4 reception issue:

We have discovered the cause of this dramatic drop in bars, and it is both simple and surprising.

Upon investigation, we were stunned to find that the formula we use to calculate how many bars of signal strength to display is totally wrong. Our formula, in many instances, mistakenly displays 2 more bars than it should for a given signal strength. For example, we sometimes display 4 bars when we should be displaying as few as 2 bars. Users observing a drop of several bars when they grip their iPhone in a certain way are most likely in an area with very weak signal strength, but they don’t know it because we are erroneously displaying 4 or 5 bars. Their big drop in bars is because their high bars were never real in the first place.

To fix this, we are adopting AT&T’s recently recommended formula for calculating how many bars to display for a given signal strength. The real signal strength remains the same, but the iPhone’s bars will report it far more accurately, providing users a much better indication of the reception they will get in a given area. We are also making bars 1, 2 and 3 a bit taller so they will be easier to see.

Weird. I imagine that signal-strength calculation is a difficult algorithm to crack, and subject to a large degree of uncertainty -- just as power calculation and runtime calculation are difficult algorithms to crack -- but it's a little odd that Apple would get this wrong and fail to fix it before now. The press release claims the error has been around since the original iPhone launched, which makes me thing the software issue is a relatively minor one. Obviously signal degradation is a real problem on the iPhone 4, though not nearly as widespread and devastating as some make it out to be. As Apple mention in their press release, this is an issue that affects all mobile devices; and although I'd wager that signal strength on the iPhone 4 is measurably more affected by the placement of one's hand relative to the antenna given it's unique design (which seems to more than make up for any degredation, but probably makes the signal strength prone to prominent fluctuations) it certainly doesn't look like a faulty product. Anyway, it looks like the imminent software update is one part bug fix, one part fig leaf and probably all unnecessary but at least it should cool the hysteria a little.

Boy Genius Report has it that Apple are finally ready to make a serious push into the cloud. I had expected them to show something off at WWDC given how poor they look in this area, but at least we're hearing about their plans now now.

Anyway, it's nothing unexpected, mostly just a fleshing out of what Lala offered and some eagerly anticipated features like wireless syncing. Expect it to be impossibly slick, and marketed as the second coming. It'll be interesting to see if Apple also decide to shake up Mobile Me, which is pretty weak tea once you figure out Google's Gmail, Calendar and Contacts Exchange support and sign up for a service like Dropbox.

A prediction: it'll launch as iTunes Live at a music event this Autumn, alongside some new iPods.

It should never have been released. At least now Microsoft can focus on Windows Phone with a singular focus, and hopefully release something competitive. Just about the only interesting thing about the Kin was that Microsoft sold it and branded it as its own entity, like Zune and Xbox, and not a platform living on other vendor's hardware¹. I think if the Windows Phone platform is going survive it'll need to be on the back of Microsoft-branded hardware, because they certainly won't be able to compete with Android in selling the OS to OEMs -- unless they're willing to license it for free to buy up market share² or the initial wave of Windows Phone devices are appropriately successful. The Zune HD is an attractive piece of hardware, certainly nicer than the vast majority of Android phones, and its industrial styling is unique enough to distinguish it from the iPhone. What's stopping them from releasing their own high-end, premium handset?

1. Both Kin models were made by Sharp, but not explicitly branded so.
2. I wouldn't be shocked if Microsoft did offer the OS for nothing, at least initially. Loss Leaders are nothing new for Microsoft.

Alongside iOS 4, Apple shipped iBooks 1.1 today which brings the app to the iPhone, and adds some other really nice features like PDF support. Now all we need is a Cocoa port, preferably one that lives outside iTunes. Download it here.

Apple released an iTunes Connect client for iOS today, which is great news; anything that allows me to avoid the iTunes Connect web app is a good thing in my book. The app is functionally solid but uncharacteristically buggy; it seems to throw “No Data” error messages more often than not and occasionally can't find the server at all. It's free, so I guess I shouldn't complain too much, but I expected a little more polish from Apple.

Apple’s stock dipped a little after Steve Jobs’ keynote, presumably because most of the things everyone expected to make an appearance — a new Apple TV, a revised Mac lineup, “Magic Slate”, Mobile Me, iTunes.com — didn’t (Apple did release a new version of Safari as expected, although it wasn’t mentioned in the keynote; instead, Apple chose to quietly announce it in a press release¹). In a way, maybe we were wrong to expect anything else; Apple hinted at the utterly iPhone-centric nature of the conference a month ago when they revealed an exclusively iPhone OS (now iOS) slate of design awards. So, with no new products² to announce, Jobs treated us to two hours of iPhone 4.

There was almost nothing we didn’t already know about the new iPhone in terms of functionality or specification; in fact, I believe the Gyroscope is just about the only hardware addition that wasn’t revealed when Gizmodo ‘acquired’ a prototype back in April. FaceTime was probably the biggest surprise of the keynote. Everyone knew the iPhone was getting some form of video chat — the front-facing camera revealed as much. What was surprising is that the implementation looks so smooth and elegant that it actually appears to be worth using, unlike every other video chat implementation, ever. Also, the fact that Apple wants to standardise FaceTime is pleasantly surprising and welcome, Google would do well to get on board³.

A few other observations:

• The Gyroscope is a nice addition, although it’s up to developers to make apps that use it in interesting ways.

• The display looks absolutely incredible, although we knew it would given the monstrous pixel density. It'll be interesting to see if and how developers will be able to take advantage of the new resolution, beyond just including higher-resolution graphics in apps. John Gruber made some good observations on this a couple of months ago.

• The iBooks update seems solid. I especially like the PDF support (this will all but kill off the low-end PDF-reader market).

• It’s disappointing that we didn’t see iOS 4 running on an iPad.

• The iPad now badly needs a front-facing camera; FaceTime for iPad will be a big selling point for the next model.

1. Long term Mac users have got to be a little worried that their platform has been gently put aside by Apple in order to focus on iOS devices. I’m not too concerned — 10.6 is a great operating system, and I’m patient enough to wait another year or so for 10.8 — although it would have been nice of Apple to have included a couple of OS X design awards. I think the real test with be next year’s WWDC.

2. I’m not really counting iMovie for iPhone as a new product, although it does look amazing.

3. Interoperability between iOS and Android stands to benefit the latter more than the former given their respective market-share.

WWDC is right around the corner and despite the uncharacteristic iPhone leaks I expect a few surprises. Here's what I think we'll see:

• There's very little we don't already know about the 4^th generation iPhone, but at least one unspoilt feature would be nice. Unfortunately, this is highly unlikely given how much information has leaked out.

• Developers will probably get a look at a 4.x build of iPhone OS 4 running on the iPad. Will the multi-tasking workflow change at all? I suspect not, given how consistent Apple has kept SpringBoard.

• Almost everyone expects a new version of Safari to make an appearance after John Gruber’s not-so-subtle hints that Safari will (finally) support extensions. I wouldn't be surprised if they closely resembled Chrome extensions: A JavaScript API with HTML and CSS for presentation^1. This is pure speculation, but Apple might also ship a new version of Dashcode with support for Safari extensions along with the final iPhone OS 4 SDK.

• There are rumours that Apple will reveal a new version of Xcode. Xcode has always been a little spartan when stacked up against other IDEs; that said, I do like it, and with a couple of additions² it would compare more favourably.

• Apple will inevitably release a revised iTunes.com to replace Lala; hopefully they'll reveal something at WWDC (although Apple may want to wait for their Autumnal music event, where they traditionally show off new iPods).

• I strongly believe Apple needs to and will eventually make significant pricing changes to Mobile Me; at the moment they just look weak when stacked up against Android. Ars Technica seems to have some evidence that changes are imminent.

• There’s probably going to be a couple of revisions to the Mac lineup; the MacBook Air, Mini and Pro are all due for an upgrade.

• I doubt the rumoured new Apple TV will show up, but you never know.

• Don't be surprised if Game Room is shown off a little more thoroughly, and we'll probably learn more about iAds. It'll be interesting to see how hard Apple tries to sell Game Room to developers; Jobs kind of phoned in the Game Center part of the iPhone OS 4 event.

• Jobs will wax poetic about how staggering, magical and revolutionary the iPad sales are, and how he’s humbled by blah, blah, blah.

Steve Jobs will give his keynote address on Monday at 10AM pacific time.

1. It would be really cool if Chrome extensions and Safari extensions were one and the same, although I don't think it'll ever happen for obvious reasons.

2. There’s nothing I want more than Safari-style tabs in Xcode.

Apple just launched a gallery of interactive, iPad friendly HTML5 tech demos, presumably to (a) reassure uneasy developers that they really are on board and don't plan to purge iPhone OS of Mobile Safari in order to ensure the App Store's supremacy and (b) rub salt in Adobe's gaping wounds.

The examples are nice, although slick HTML5 demos are a dime a dozen these days. Bizarrely, they're all WebKit only – even demos that other engines like Gecko are perfectly capable of rendering – so it's really only showing off what WebKit can do, which sort of defeats the purpose. Worse still, the site will block all non-Safari browsers so even those using some flavour of WebKit, like Chrome (which has a greater market share than Safari) will be turned away. Stupid.

A few years ago, one of the most important goals of responsible web design was to get your site looking identical, or as close to identical as technically possible, in as many browsers as possible. Typically, this involved writing valid markup and valid CSS, and then tacking on a panoply of hacks to make it work in Internet Explorer. This is still standard practice for a lot of designers, especially those who choose to or must support IE 6. However, a significant number seem to be much less doctrinaire about making sites look absolutely the same for absolutely every user, even as it's gotten substantially easier to do so. More important, it seems, is making websites look as good as possible in more capable browsers, then making sure they degrade gracefully and subtly when rendered using older engines.

Here's an example: the latest versions of Chrome, Firefox and Safari all support the CSS3 border-radius property¹. If you visit twitter.com with any of these browsers, some of the UI elements, including the main content area, will have rounded corners. It's nothing superfluous, and nobody would really miss it if it weren't there: which it isn't for Internet Explorer users. Instead of building a messy hierarchy of nested divs and images to get rounded corners like designers used to do, the twitter guys are forgoing the effect altogether in older browsers, and enabling it in modern browsers by simply adding a couple of CSS declarations. Almost everybody wins; the twitter designers get cleaner markup and save a small amount of bandwidth by not having to serve excess images, and a lot of users get pleasantly round corners and slightly faster page loading. The only loser is the guy with an old or crappy browser, and he has an easy upgrade path to the nicer site: download a better browser. Plus, all he's losing out on is a visual flourish that has no practical impact on how he uses the site, so if he chooses to keep using old and crappy software he can do so without being adversely affected.

I've noticed this becoming more and more common, particularly as Chrome and Firefox eat into IE's market share and more developers target mobile browsers². If you take a quick poll of the websites you visit most often, I'd wager that at least a few look better in Chrome, Safari and Firefox³.

I'm not suggesting designers are going to allow their work to look drastically different between rendering engines, but I won’t be surprised if everything starts to look a little bit staler in Internet Explorer as CSS3 usage creeps more and more into the mainstream.

Overall, I think this is a hugely positive trend. If more developers start using and field-testing CSS3, it should prompt users to demand more from their browser, and encourage vendors to support web standards promptly and thoroughly⁴.

Anyway, the moral of the story is this: don't be afraid to start using cutting edge web technologies before they gain widespread adoption. That said, some functionality is more suitable for use now than others; border-radius, text-shadow, @font-face and gradients⁵ are all good properties to play around with. If you need a quick guide to what's supported in each browser, check out html5readiness.com.

1. Some versions require the -moz and -webkit vendor-specific prefixes.

2. Mobile browsers are almost all based on WebKit, which is on the bleeding edge of web standards support. I don't think the impact Mobile Safari has had on HTML5 and CSS3 uptake should be understated.

3. This is doubly true for sites whose audience is largely made up of browsers that aren't IE. Also, I suspect that some big media sites like The New York Times and Newsweek are more reluctant to use emerging technologies because they have a greater stake is making sure that their sites remain consistent across all platforms. In other words, when you see The New York Times using text-shadow, you'll know CSS3 has reached critical mass.

4. To be fair, the major straggler, Microsoft, is making a big push to support HTML5 and CSS3 in IE9. Look out for it in 2011.

5. The syntax for gradients is fairly different in WebKit and Gecko, and both engines require a vendor-specific prefix. Gradients are supported in IE by way of filters; meaning cross-browser CSS gradients are entirely possible right now, if not quite as cleanly as CSS3 will eventually allow.

My latest project, CharMap, has just been approved by Apple and is now available worldwide.

CharMap is a Unicode character map and font viewer for iPhone OS. For more information, check out the project page or jump in and buy a copy from the App Store for $1.99.