Merry Christmas

Happy Christmas, everyone.

Just a note about subscriptions

Companies love subscriptions because it gives them recurring revenue every month and a compulsion to use their product, because customers are usually either too lazy to cancel and sign up for something else or believe that the money already invested represents the overall price of the product. However, some companies and/or groups, such as Microsoft, the RIAA, some of the digital music downloads purveyors, let the allure of recurring revenue get in the way of a user experience. Customers are confused when they go to look for a song and suddenly it is not there, because exploding media is defective by design.

Why there is nothing but Java for android

So, I've been thinking a bit more about Java and C on the Android phones, and the conclusion that I have come to seems to make sense to me, at least.

Google is using a walled garden strategy with regards to application security: each application gets its own virtual garden and virtual resources which it then uses. Applications can then make connections between walled gardens if they want to share data. I have written about this before.

This implementation is done in Java, because it makes sense to run a separate Java VM for each application. The phone is not viably going to be running more than the frontmost application at one time, and the background processes are probably going to be nice'd through the roof. This enables the frontmost app to be responsive to user stimulus, at least with a keyboard. This is a system that works fine in Java, but what if we add another programming language to the mix, such as C++ or Objective C. Without building its own virtual machine to run non-interperetted programs in and optimizing it all by its lonesome self, Google is using a language that already has an optimized interpreter, enabling it to do as little work as possible. This prevents Android apps from being written in any language but Java. Quite frankly, I'm surprised that Sun is not in on the open handset alliance. Probably because they are designing their own Java-only phone. Ooh, more competition in the market.

There is yet another problem with using Java in a device with a not-so-fast processor: if you need something constantly updated with quick redraws, like something using a 3D engine, it will be pretty slow. An example of this is the earth demo on the example android phone. The globe was lagging just enough to be annoying and seem too slow to be usable. Java generally is a very slow language, which is the cost of a very dynamic language. I like Objective C because it incorporates the right amount of dynamism and static typing for me to use effectively, plus the fact that it is really easy to use.

AHA!

Another article agrees with me on the matter of android phones and echos exactly my point.

The main difference is of course that while the Apple iPhone platform only has to work on one device, Android has to support an extremely diverse range of hardware devices with vastly different capabilities. This will undoubtedly lead to a whole heap of issues.

I wrote about this earlier, and I predicted the exact same thing. The android SDK is a bit clunky as well for those of us that don't like to use Java. I would much rather get a nice C compiler and write apps that run without requiring an interpreter.

A feature that I just realized I want

I just tried to use quick look– in Xcode. I pressed the space bar a couple of times before realizing that it would not work because it is not the finder. Wouldn't it be nice to have quick look in Xcode, and not just the little preview pane. It's a feature that I want.

Front Row not as in front as you think?

I was looking through the private frameworks folder, and I noticed that Leopard's Front Row seems to have had very little changes from the original AppleTV binary and framework, which I conjecture from the framework still being named Backrow.framework. Interesting.

Workable Apple TV DVR?

I was reading this article on Roughly Drafted, and I thought of a way to deliver a workable Apple TV DVR: through the USB port. The DVR would be opt-in, so the tuner part would be sold seperately, but it would be a good way to get HD content on the Apple TV. Current Apple TV owners could take advantage of it as well, via a software update. And best of all, Apple is deferring the income from the Apple TV for two years, so that justifies offering a database of listings for at least two years. It would support both satilite and cable by means of a satilite tuner and a cable tuner.
The downside? The USB port is full, so an external Hard Disk could not be plugged in to store all of your HD TV on. However, Apple might also enable remote storage of recordings on an Airport Extreme with a hard disk attached.

Reality Phones

With the iPhone, the sum of the whole is greater than the sum of the parts.

Most of the content of the iPhone is in its operating system, not the hardware. Apple has spent two years on this phone, which was admittedly part developing the touch screen, but probably also part writing the software, including Core Animation. Handset makers need to get something: licensing Windows Mobile or using a slightly modified version of your own OS is not enough. It has to be fully customized for the handset, right down to the volume switch. It also needs to provide a good user experience. For example, most smartphones allow storing documents in the phone's filesystem. The iPhone hides all of the filesystem from the user, instead opting for a simpler approach of storing things in Core Data databases (I presume). On a normal smartphone, one might have to type in a filename as C:\ and lots of slashes, or navigate a directory tree. Phone OSs are generally not tailored to the phone, nor do they aim for usablity. Microsoft is following the same strategy as it did with Windows: just licensing its OS to the hardware makers, all the same on every phone to be customized in small ways by the vendor.

The fact is that too many chefs spoil the soup. Carriers want their services integrated into the phone so that the consumer pays for them, and often spoil the experience. Handset makers do not have a good backing in software and therefore tend to fail in that area.

What makes Apple different is that it has an accumulated library of technologies to build its programs on. Apple has experience delivering a stellar user interface, and can extend its existing technologies and paradigms to the phone (Core Data, Cocoa, Syncing). Apple has an experience in software, and it really is the software that makes the iPhone come alive. Normal phone vendors have very little software experience, and therefore do not have technologies to do what they want with the OS.

But that is only one path to success in the cell phone market. The entire cell phone market seems to be stuck twenty years behind the computer market, as nobody seems to be able to grasp the concept that a good user experience is often going to be more rewarding, instead of locking customers in with contracts. All it would take is for carriers to hire really good customer service people, not try to kill or maim the customer and then take all of their money, and drop or shorten the contracts. A carrier that positions itself as friendly instead of a monolithic giant can succeed. Couple that with a superb software experience on the same network, and you've got a winner.

Technorati Tags: Apple, Cocoa, iPhone, Mac OS X, Microsoft, Mobile Telephony, Cell Phones, Windows

Time Warp

I have had a visitor to my blog using OS/2. Strange.

Numbers Table Controls in Keynote and Pages

You can get Numbers style controls for editing rows and column widths in Keynote and Pages. Just type "=" into one of the cells as though you are going to begin a formula. Numbers headers should appear, and you can grab one of the lines between the headers to resize the rows (keep dragging after the headers disappear).

A strange Leopard glitch

This is very strange, and it only seems to happen with fullscreen apps. When it enters full screen mode, the dock glitches up. At first I thought that it only happened with EV Nova, but then I found that it also happened with a pinball game that I was playing. Strange. Resizing the dock to its smallest size and then back repairs the left side and then opening each stack individually repairs the right. I normally just type "killall dock" into a console, but I thought that I should document this glitch.

Regarding Cell Phones...

So, this story was on digg.com. The story says that the proof that cell phones disrupt an aircraft's navigation systems may not be true, but the FAA and the FCC are still investigating whether the technology is viable. I have no problem with that point.

The point that I have a problem with is what they left out: using a cell phone in flight triggers every cell tower near the plane, which being high up in the air stands to be a lot of cell towers. The article states that they tested some flights for cell calls and that on average there would be one to four calls per flight. One to four calls do not disrupt the cell network, but imagine if half of the people on the plane were placing calls, or using a data network (like EDGE). Add in multiple planes, and you have a recipe for increased power consumption, decreased reception, an increased number of dropped calls, and some people not being able to make a call. Until a better technology if found, cell phones on airplanes will probably stay on hold.

Technorati Tags: Airplanes, FCC, Mobile Telephony, FAA, Usability

Goggle at google

So, I read about the new iPhone interface for Google, and guess what, it adds a fourth interface in addition to the other the other three, and is only available on the iPhone - not the google phone, I tried it in the emulator. Interesting, but it makes the many faces of Google more confusing.

Respect the iPod, and tame the iPhone

I have been thinking more about the iPod being the dominant MP3 player, and how other companies would perceive the iPod majority with respect to the iPhone. For example, most cell phone makers are attempting to make phones with touch screens, and cell phone providers are obviously worried that the iPhone will become the dominant phone; even AT&T is worried, that's why AT&T is also carrying competing phones. With a Verizon accepting AT&T's standard for 4G phones, it is expected that the 4G iPhone will run on Verizon as well as AT&T (whether the 4G iPhone will be unlocked is an entirely different issue). Most of these maneuvers I interpret to be to attempt to move the Cell providers into a position to profit from the iPhone. AT&T is competing against the iPhone because it does not want the iPhone to be the majority standard when its contract runs out.

I have noticed an increase in advertising for cell carriers recently, which can be attributed to the iPhone causing ripples in the cell industry. Also, no cell provider seems to actually care about its customers; they all seem determined to sell more and to gain subscribers. I think that all that it would take to cause a reform in the cell phone market is for one company (possibly one bidding on the 700MHZ spectrum) to focus more on caring about its customers instead of focusing on causing them pain. Such a cell phone company would rapidly gain subscribers, mainly through word of mouth marketing. However, the massive costs incurred in setting up a cell company would require a company with large amounts of money already.

Astute readers may have conjectured that Google would be an optimum company to do this. I believe so, as it would provide an optimum method of cash flow for Google (subscriptions). Google may also be able to subsidize the costs of running a cell phone network by running ads on its phones. This would be an optimum way to do cell phone services, as it incurs less cost to the consumer. I would imagine Google offering cell plans with data starting at about $40 per month. All that Google would have to do is set up a high speed data network and then create and run a Google VOIP. You would also be able to use cell minutes with your computer in this system.

There is much competition in the cellular market at the moment, and I see a possibility for a company with a lot of money to make a lot more money through setting up a cell network. All that a company would need to do is see that hole in the market and pounce on the opportunity.

EDIT: It has been pointed out in the comments that the iPod is not a monopoly, it's a majority. The text of the entry has been changed, and I am sorry, but I sometimes think one thing and write another.

Technorati Tags: Android Phone, Apple, AT&T, Google, gPhone, iPhone, Mobile Telephony, Smartphone, Verizon

Apple's Devotement to Professionalism

One of the things that I noticed looking at the picture of competing MP3 players to the iPod and competing phones to the iPhone is that Apple's seem so much more polished. However, they also seem a lot cleaner, using simple things such as highlighting menu items and pure white backgrounds. The use of images is sparing and the pictures are often less important than the text. Something that I noticed on the competing players is that they use absurd fonts or layouts, and the images are used without the text to intuitively tell what they actually are. Icons are not enough, you have to have the text to back it up. The idea of spatial intelligence plays a large part in UI design. Things like the Zune are pitifully trying to be 'cool', while the corporate giants do not realize that the spartan UI is actually the coolest. Such distracting UI designs remove the focus from the intent (music or telephony), and instead provides the player virtually saying, "Look at me! I'm cool!" This is also a problem in phone UIs, which were never cool before the iPhone.

The real joy of Leopard

Overall, Leopard seems more of a release building on Tiger for the average user. Although it incorporated X.Org over XFree86 and built in a lot of new developer technologies such as Core Animation, it provided only a couple of enhancements for the average user, with the main changes showcasing features such as Core Animation and other improvements. Time machine: Core Animation and Core Graphics, the new Finder: Core Animation, Spaces: Core Animation. Although these provided a lot of useful features (I use spaces a lot), there is no one "killer app". Some may say that Time Machine is the killer app, but I think that there is still a lot of room for a delicious generation app to take advantage of all of the new features in Leopard and become a smash hit.

Today I downloaded and built Scott Stevenson's Photo Gallery, a Core Animation demo app which I think is really a harbinger of the new Leopard apps. For some reason I am not able to post a picture right now, but believe me, if you have the developer tools installed and can compile it you should. The coolest effect is when a picture is clicked on a ripple emanates from your mouse and the area within the ripple turns to black and white. It is really cool, but you have to experience it for yourself. I do believe that this is a preview of what is to come as third party apps for Leopard, and perhaps iLife and iWork '09.

Technorati Tags: Apple, Cocoa, Cover Flow, Leopard, Mac, Mac OS X, Core Animation, Quartz

So, with the Kindle stuff over...

I think that now is an excellent time to start thinking about the Kindle 2.0, which I think about as the Timber (or Tinder, whichever floats your boat). I have made a mock up of what I think would overcome the Kindle's weaknesses as a 1.0 product (click for larger version).

In addition to the features mentioned above, it is more rectangular and I think looks better (even though it is obviously a mock up). I like it, and between it and an iPhone for the same price, I would still choose the iPhone, but I would choose it before buying $400 worth of books.

EDIT: Oh, and you can't stick an XKCD actual size sticker on there because it isn't actual size, I would imagine it to be a little bit bigger.

What the iPhone/iTouch SDK will mean

I have been thinking since my last posting on the limits of the iSDK, and I have put together a chart of what may be feasible and what probably is not (this is all conjecture based on guesswork).

• Connect to the internet over EDGE (iPhone only):
  This probably cannot be done, thanks to the rumored limits on the SDK. Most likely apps will not be able to access the low level EDGE connection and share it with a laptop.
• Wireless exchange of files with a computer:
  Probably, over bluetooth or WiFi, as apps will need some way to communicate with other devices (including computers)
• Disk Mode:
  I think that this may be possible in a limited way, a.k.a. Access all directories except for /System, /tmp, /usr, /etc, etc.. Apple would have to move their applications into /System/Library/CoreServices/, just like on a desktop mac.
• Wireless Sync:
  Probably not, as 802.11g is only 54 mbps, which may seem fast, but the experience would be similar to 7 MB per second, compared to USB 2.0 speeds of 60 MB per second. Oh, and you would need wires to charge it with, so there are two differences: ten times slower and only needs to be plugged into a wall, not your computer.
• Microsoft provided MAPI protocol:
  Anything is possible
• FREE Microsoft provided MAPI protocol:
  OK, maybe not anything
• "Productivity" Apps (Office, etc.):
  Possibly, though we have viewers for Word and Excel documents, and PowerPoints can be exported to PDF for viewing on the iPhone. My guess is that that is good enough.

Why Microsoft Office 2008 is going to be Office's Vista

I have written about this before, but I believe that Office 2008 is going to fail. After thinking about it some more and re-reading the Apple Insider series, I have come up with a new set of reasons why it is going to fail.

1. Not Mac like enough. I wrote it before, and I will write it again: Office 2008 is not enough like Office 2004 or other mac applications to get the benefit of a consistent user interface. Even inside of the four office apps there is not a consistent user interface; Word, PowerPoint and Excel use nonstandard toolbars while Entourage uses a normal Mac toolbar.
2. Different for the sake of being different. This was Vista's problem, it did not have enough in common with XP, so users chose something they were familiar with. Sure, you could take time to learn it, but to me the interface seems a little jarring each time I compare it to a normal Mac app
3. Looks too much like Vista. I know this is a repeat of the first one, but seriously, this takes too many of it interface cues from Vista and not enough from Leopard or Tiger. The toolbars, the overlays, and the widgets in general are just too "glassy". And there is a lot of unnecessary work being done too: the re-implementing of a lot of Carbon widgets to look more like Vista. I think Office v. X looked more Mac-like, and that people are going to stick with Office 2004, as it fits a bit better into the Mac look.

I think that Microsoft mainly implemented a complex look for Office (Windows and Mac) to make it harder for Open Source projects to look good by comparison. Gloss and transparency coat the proprietary implementation, while the Open Source one looks like the last version. Seriously, Office customers might move to OpenOffice if they are not able to understand the new UI and want something familiar.

Technorati Tags: Leopard, Mac, Mac OS X, Microsoft, Microsoft Office 2004, Microsoft Office 2008, Open Source, OpenOffice.org, Vista, Windows