Happy Birthday, iPhone!

Today is June 29, and that marks the anniversary of the launch of the first iPhone. This also means that the one-year warrantees for the first iPhones bought expire today. Mine expires in September, so I need to get my iPhone-skydiving experience in soon. Arguably the warrantees expiring don't mean much, because if it hasn't turned out defective or buggy in the first year, it probably isn't. In addition to the warrantee thing, the iPhone contracts of people who bought their iPhone on the 29th last year are half over. Apple's deferred revenue from the iPhone is also half over. Happy Birthday, iPhone!

Is a broken box model too much?

For so long, web developers have had to deal with the quirks of internet explorer. Today, most focus on their web site working under versions six and seven. Apple has decided that they were not going to put up with it anymore, and decided to drop support for IE 6 in the upcoming MobileMe. According to the linked post, 31% of the people that use IE are using IE 6. According to Wikipedia, 75% of browsers are IE. Now, anyone can multiply those two percentages and come up with the relatively small number Apple is willing to ignore. WHAT! One fifth of the internet! Apple did not have to think about this decision. The decision was made for them the moment they picked up SproutCore, because SproutCore programs do not run under IE 6. Not even a little. The two example programs on the SproutCore homepage do not. SproutCore wants a more complete CSS implementation. Although it works around IE's Javascript DOM (see here), its box model just breaks down in IE 6.

Apple's decision to not support IE 6 in MobileMe is probably for the best. Everyone wants to see a move away from IE 6, including Microsoft, who has been pushing Internet Explorer 7. Although ignoring a fifth of the internet is difficult, that fifth of the internet may just decide to change browsers so that they can do more. After all, most web browsers are free. Legacy users go through a one-time process of switching and the internet is better for everyone. Internet Explorer 6 is now six years old. Isn't it time the web moved on?

Technorati Tags: Apple, Internet Explorer, Microsoft, MobileMe, Web, Windows

Levreging the Amazing Power Trapped in a Video Card

Apple has its new "Grand Central/OpenCL" feature planned out for Snow Leopard. OpenCL is a feature that uses the immense parallel computing power in a video card for other purposes. Has it occurred to anyone that Apple may be looking at other applications for OpenCL besides the desktop? The fact that this comes from the same company that had the novel idea to put a video card in a phone makes me believe that:

1. The iPhone 3G may come with a faster video card, both to help Core Animation renders as well as take some of the load off of the processor.
2. iPhone Firmware 2.0 may include OpenCL already. In the same way that iPhone Firmware 1.0 was a Leopard-based system, iPhone Firmware 2.0 may be Snow Leopard based, or at least take advantage of the graphics card when Core Animation is not being used.
3. OpenCL will set the iPhone apart from Android, as it will make the already fast iPhone experience even faster. Although users do not normally run very intensive tasks on their phone, the phone may be called upon more and more to do things such as blogging and the like. Parallelism is also useful in things such as fast Javascript interpreters (critical for a mobile device.)

Speed and responsiveness are of the essence on a mobile phone, because to do a task in thirty seconds or less, the hardware definitely needs to cooperate. OpenCL contributes to that speed, for although developers are not expected to fill up a graphics card with their threads, they are at least expected to employ parallelism in their applications. Multiple threads have many uses (background polling, perhaps? If Apple did not already have a solution it would be a good argument), and it is possible that Apple may deliver the first mobile phone that really makes use of multiple threads.

Technorati Tags: 3G, Apple, Cell Phones, Core Animation, iPhone, iPhone SDK, Mac OS X, OpenCL, Quartz, Smartphone, Snow Leopard

Software Vs, Hardware: Android and iPhone

Two major smartphone platforms have dominated 2008: Android and the iPhone SDK. However, these two phone systems have vastly different models: the iPhone chooses to focus on software, while Android focuses on hardware. This is evident in the release of the 3G iPhone, where the only new hardware feature accessible to developers (so far as I can see) is the GPS. Android has a myriad of hardware features, but also a myriad of possible features to support. For example, in Google's Maps application for Android, thene is a feature within street view that requires the device to have an internal compass as well as GPS. It is logical to assume that phones without compasses will not have this feature available to them. The thing is, consumers do not like to see features "grayed out"; disabled features only add complexity and a feeling of "my phone is bad because it Cannot do <insert feature here that the phone does not support but is in the interface anyway.>" Consumers like to be able to use all of the features of their phones, because even if they do not use any of them besides the part that makes calls, they like to have that feeling of "my phone can do this" satisfaction. This also creates a headache for developers, especially those developing installer software, because apps have to worry about dependencies. Because there is no central way to specify the hardware dependencies of an app in android, people may end up installing an application and then finding out that it doesn't work because they don't have a specific piece of hardware. Compare this to Apple's model, where everything will run on either an iPhone, iPhone 3G, or an iPod Touch, which all have pretty much the same feature set. And for those 2.5G iPhones that do not have GPS, it can be assumed that smart developers will have an intervening if statement that uses cell triangulation instead.

In the aforementioned scenario, Android is beginning to look a little fragmented hardware-wise, as it embraces the Microsoft model of providing only the software and letting the hardware makers build the phones, while Apple is taking its traditional stance of keeping its cards close to the chest and exerting a tight degree of control on its hardware and software, which enables it to provide deeper integration. And with a centralized application distribution system, similar to the style of the Linux package manager with money added, Apple manages to make a tidy profit out of it. Google manages to get its money's worth too, just in a different form: people using the internet more often. And with more people to use the internet, more people use search, and thus see its adds. This is not the only effect however, as more people using the internet makes internet advertising that much more viable, due to a larger and more diverse audience. Therefore, Google sells more ads, and as more people set up web sites, Google buys more ad space. Although both strategies are valid, I prefer Apple's strategy of militant compatibility across devices, because it guarantees a better overall user experience with the device. I would imagine user experience would be important for Google, as the better user experience people have with the web, the more they use it. Although each platform is different, Apple and Google may just have to compete for the high-end smartphone market (both phones require fairly good video cards.) Apple and Google have gotten along in the past, the question is: can they do it in a market they are both attempting to flesh out in vastly different directions?

Technorati Tags: 3G, Android, Apple, Cell Phones, Google, gPhone, iPhone, iPhone SDK, Linux, Open Source, Smartphone, Usability

Open Source is Magic, Not an Excuse For Laziness

Nokia wants open source developers to "embrace the old-school cell phone business model of DRM, Intellectual Property, SIM Locks, and subsidies. Nokia expects the open source developers to do their work for them and make them an OS, albeit an OS that goes against the Open Source philosophy. In the article, it was said that Nokia had once tried to spin its own version, but that the community had moved ahead of them in the time it took to do that. Nokia needs to embrace Apple's model of open source: take the original project, then write some third party drivers and add-ons that can be laid over the original project. These add-ons can have DRM. These add-ons can have Intellectual Property. Now, I am not saying that Apple's model is perfect (their version of most of their open source is a little behind), but it works, and Nokia has the chance to improve that model by allowing the open part of the project be updated under the open source model. Nokia may have to invest a little work in this (Kernel modules have to be recompiled with a major kernel update), but they have everything they need, and with a little intelligence, they can make the system work on their side.

Certificates and Their Place in the Web

I have implemented zero sign on in my myOpenID settings, and let me say that it is ... interesting. It is similar to having a keychain, not having to type in passwords, however it is also different in that you are already logged in on all sites that support OpenID, and you appear as the same person on all of them. Client Certificates, the method behind zero sign on, may become more common once they are simplified. At the moment though, they remain an esoteric oddity.

I Sense a Disturbance in the Source

Hmm. I believe SCO owns the Me-inc. trademark. And they provide software for mobiles, such as syncing software... SCO vs. IBM... History may repeat itself.

Sorry to tease you SCO, but...

Motion in the Semantic Sea

The semantic web is evolving quickly; Apple is expected to unveil MobileMe, a revised .Mac with integrated desktop-quality applications, Google is encouraging semantic web development (and does quite a bit itself). Here are some prime examples of great semantic web applications.

• 280 Slides. This is a great example of what I expect the semantic web to become: Dynamic languages interpreted into Javascript. This one uses Objective C, but Google Web Toolkit allows you to do it in Java. This is a web app that imitates Keynote. I have a sneaking suspicion that the people at this company either work for Apple as part of the MobileMe initiative, or are very soon going to be working for Apple. I also suspect that they are using a WebObjects 4.5 backing.
• The Waterfall Software Store (WebKit and Konqueror only). Amazing. Astounding. Probably the fastest drag I've seen, and that's because it's operating system-level. This store uses the CSS property -khtml-user-drag and javascript to provide an awesome integrated shopping cart. It degrades well too. Users of Firefox and Internet Explorer will get 'buy me' boxes on the items.
• .Mac Web Mail Apple really did well here.
• Acobat.com The concept of VNC inside a browser window is pretty cool, even if the performance penalties may be a bit greater. In addition, the PDF editing and such are... interesting.

After that there are pretty much the usual culprits: Google Reader, Photoshop Express (even if it does use flash), etc.. The semantic web is starting to grow up, and that is a move that could make or break companies.

Is Snow Leopard really that big of a surprise?

It is rumored that Mac OS 10.6, a.k.a. Snow Leopard, will be previewed at WWDC and announced at the Macworld Expo 2009, just fifteen months after Leopard. What people don't seem to realize is that Panther came fourteen months after Jaguar, and Tiger came seventeen months after Panther. In fact, it is only Leopard that took two and a half years. For a visual representation of this, see this chart. Maybe 10.6 represents Apple's return to normalcy in its release schedule.

Are touch screens the way of the future?

I have previously written about the difficulty of touchscreens in larger computer screens. The main difficulty: real world scalability. Touchscreens do not scale well from 3.5 inch phone screens to 17 inch computer screens. I have previously seen a good demonstration: touch the top, then bottom of your computer screen for three minutes. I wrote about this earlier. Imagine moving your finger like that for three hours. The thing about touch screens is that they add cost to computers, in addition to the keyboard. And most of what I do when I am using the computer is writing. Most of the time, I am not using the track pad, because reading and writing email involves the keyboard, writing blog posts and papers involves using the keyboard, and writing code definitely involves using the keyboard. The places where a touchscreen would be useful are in multimedia applications such as iLife. Organizing photos is definitely a good application for a touch screen, as is using iMovie. However, note that fingers are a lot less precise than mice, trackpads, or styli. Multi-touch is also a good technology for gaming, as it is possible to select multiple thing, and touch multiple targets, etc.. However, each user's preference for multi-touch may depend on what they do with their computers. If they are heavy writers, they are probably not going to use their touchscreens very much, and they may become extra pricing. If they are heavy gamers, they may use touchscreens a lot.

Although having a touchscreen does not preclude the use of a keyboard, it does drive the price of a computer up. With Microsoft's initiative for Windows seven to employ more touchscreen and multi-touch capabilities, they may drive the price of a normal PC up. With Apple scraping the cream-of-the-crop expensive computer market, Microsoft may have more trouble with Windows Seven, as computers become less affordable and Macintoshes become more viable in that price range, Microsoft may see more switchers. And when consumers are looking to get PCs that do not require touchscreens, and computer manufacturers are looking to eschew the cost of a touchscreen and an expensive OS, Linux may become a more viable option to Windows on low-end PCs. With Apple moving in from the top, and Linux moving in from the bottom, Microsoft may begin to lose market share, or it may fumble it in the same way as it fumbled Vista: higher hardware requirements and a paradox of choice in its pricing scemes. And Microsoft may become less relevant as the tech industry moves on to less expensive things and bigger and better things, with Windows caught in the middle.

At what point?

The Golden Compass has incited a proposal to censor books based on age. At what point do books such as Fahrenheit 451, 1984, and Animal Farm become "Inappropriate for all ages." Will it even matter, because people will be too complacent to comprehend the symbolism? Isn't this net neutrality in printed form?