Is Apple in long-term trouble?

A survey of software developers shows a sharp drop in Objective-C popularity, Objective-C being the main programming language for Apple's iOS.

Source: tiobe.com

Fewer developers means fewer apps for consumers and businesses. One could argue that with hundreds of thousands of apps already available in the AppStore Apple should not worry about the trend. Furthermore, Apple's move into its own services, including media streaming, may also decrease the need for independent developers. In general, the mobile apps space has matured well beyond its heydays.

Nevertheless, it's hard to imagine a popular software development platform that is of limited interest to developers. We might be seeing the beginning of the end of Apple's rapid expansion.

tags: technology, apple, software, services, dominant design

Trade-off of the Day: Smartness vs Ease of Use

Steve Jobs shows how Apple broke the trade-off with the iPhone.



tags: trade-off, dilemma, interface, mobile, software, apple

Nortel/Rockstar US Patent 6,378,069 - a ticking bomb for the smartphone industry?

In July 2011, a group of companies led by Apple, Microsoft, RIM and others bought 6,000 Nortel patents. The patents were assigned to a holding entity Rockstar Consortium, Inc. One of the patents in the portfolio is US 6,378,069 "Apparatus and methods for providing software updates to devices in a communication network," issued April 23, 2002. I ran into it while reviewing references for the newly issued Facebook patent US 8,631,239.

The original Nortel patent looks incredibly broad. Its claim 1 covers any system that has a database of subscriber records and can provide software updates to the subscribers.

Arguably, the patent covers software update systems for smartphones, PCs, tablets, ebook readers (hello, Amazon!), and everything on the Internet that requires a software update. One could try to circumvent the patent by delegating software distribution to third parties, i.e. decoupling the entity that holds the subscriber database and the one that actually communicates to the destination device. It's not clear how this solution could play out in courts. Litigation over this potential workaround would be an interesting case to watch.

Rockstar has not asserted the patent yet, but the consortium still has several years to do that. They've already sued Google for allegedly infringing Nortel patents on relevant advertisement. Most likely, the Nortel portfolio contains more patent gems; litigating them can prove extremely expensive. The industry would be wise to set up a standard-like body to figure out reasonable licensing terms, instead of engaging in a series of all-out patent wars.

From our system model point of view, the patent covers key Control Points (see Scalable Innovation, Chapter 5, System Control Points: Where To Aim Your Silver Bullets), i.e. using Aboutness to direct a Packaged Payload - this is as basic as it can possibly be. An equivalent broad patent for Facebook would cover using information about users for sending messages, including ads. For Netflix, it would be using subscriber database for sending recommendations, etc. Powerful, if you can get it issued.

tags: patent, invention, software, innovation, portfolio, communications, aboutness, packaged, payload

Quote of the Day: Steve Wozniak on GUI.

In iWoz, Steve Wozniak writes about seeing for the first time the Graphical User Interface (GUI) at Xerox Park,

The minute I saw this interface, I knew it was the future. There wasn’t a doubt in my mind. It was like a one-way door to the future—and once you went through it, you could never turn back. It was such a huge improvement in using computers. The GUI meant you could get a computer to do the same things it could normally do, but with much less physical and mental effort. It meant that nontechnical people could do some pretty powerful things with computers without having to sit there and learn how to type in long commands. Also, it let several different programs run in separate windows at the same time. That was powerful!

tags: quote, interface, software, evolution

Landmark Supreme Court cases that shaped software patents.

GOTTSCHALK v. BENSON, 409 U.S. 63 (1972) - algorithms cannot be patented.

PARKER v. FLOOK, 437 U.S. 584 (1978) - novelty of an algorithm (process) resides in the implementation.

DIAMOND v. DIEHR, 450 U.S. 175 (1981) - physical process controlled by software is patentable.

BILSKI ET AL. v. KAPPOS No. 08-964 (2010) - abstract (business) method (including software) not patentable.

No Flash, no problem.

As I predicted a year and a half ago, Adobe Flash for mobile devices is destined to die.

(Nov 8, 2011. CNet). In a momentous about-face, Adobe Systems is scrapping its high-profile effort to bring its Flash Player software to smartphones and tablets.

This is another indication of a major shift in the development of the Internet. Technologies that created the web as we know it today are being swept aside by a new wave of innovation.

@ $730K per patent, Google buys Motorola.

Whatever the business wisdom behind buying one of the worst mobile handset manufacturers in the world, Google's acquisition of Motorola is great for patents. Assuming, generously, the value of Motorola hardware division is $0, Google paid about $730K per patent in the deal. On the per patent basis, this is 50% more, than in the Nortel auction won by Apple, Microsoft, and others. Now Google will have a free hand at ripping off Apple design and business strategies, hiring Apple's designers, and so on. Add to it Google's cloud capabilities, with Youtube, Docs, gaming, etc., and you get a very strong challenger, both in the consumer and enterprise segments. It might be a bit too early to sell Apple stock short, but the competition in the mobile world has be come a lot less lopsided. Unless, of course, Google suffers from the Not Invented Here syndrome, and tries to create its own brand of smartphones.

From an innovation theory perspective, this event is also significant because it confirms that Open Innovation does not have a working IP model, except costly patent acquisitions. Free software and open technology is great, as long as it does not involve a major commercial success.

tags: innovation, patents, growth, software, business, model, mobile, control, battle, theory

Mobile stalemate: Apple vs Google vs RIM.

Half a year ago I wrote about data on smartphone market share, which looked like that:

Google Android - 43.6%
Apple iOS - 26.2%
RIM - 24.2%
Microsoft - 3%

9 months later, market share distribution didn't change much, except Google lost few points to Microsoft:

Google Android - 40.1%
Apple iOS - 26.6%
RIM - 23.4%
Microsoft - 5.8%

Google's growth of 5.4% over last quarter seems like a seasonal variation caused by different release schedule from hardware OEMs. It would be interesting to see revenue margins for those phones/subscribers. Google is not making any money on the OS itself, only on app sales, ads, and subscriptions. Apple is making money on phones, apps, ads, subscriptions, media - songs & video, and accessory licensing - cables, connectors, car interfaces, etc. In addition to that, Apple sells iPods and iPads, which are not included in the market share data.
Putting it all together, I think in the mobile space Apple is a much more dominant force than Google. I also wonder how the whole enterprise market is going to work out. Microsoft will probably make a big push for it, trying to grab a share from RIM.

tags: mobile, business, information, apple, google, microsoft, evolution, market, control point, software

Business model innovation: scutage.

At around 1100, English king Henry II conducted a number of financial and administrative reforms, which allowed him to create a well-trained professional army and conquer territories far beyond the British Isles. Here's how it was implemented:

Rather than relying on the customary military services of forty days owed him annually by his tenants-in-chief with their retainers, he began insisting that these feudal services should be commuted into a cash payment or ‘scutage’ (from the Latin, ‘scutum’, a shield), a process as welcome to the barons as to Henry’s finances. With the proceeds Henry was thus able to build up a more reliable and more mobile, permanent professional army of mercenaries, or ‘soldiers’ as they became known thereafter, from the ‘solidus’ or king’s shilling that they earned.

History of Money, by Glyn Davies. p. 142.

What are the advantages of the new vs the old military service model?

The first advantage would be a greater degree of specialization, which is a common attribute of almost any successful business model innovation. With the introduction of scutage, the barons could specialize in making money, e.g. by improving agriculture, crafts, and trade within their domains, while the king could specialize in soldier selection, training, and military strategy.

Another one, would be a greater control over resources. The barons could now control the timing and scope of labor use, which in the old model could've been taken away from them at arbitrary times, e.g. during harvest gathering or seasonal forest clearing for agricultural needs. At the same time, the king gained direct control over army mobilization time, which in the old model depended on the willingness and ability of barons to provide requested manpower according to the kings needs.

Finally, scoutage led to growth of a new class of professionals - mercenaries whose only purpose in life was warfare.

The structure of this business innovation is similar to the transition undergone by the computer industry in the 1970-80s: from IBM enterprise model, where software and hardware were provided as one package with a series of periodic custom updates, to PC-based shrink-wrap software sales, a model Microsoft eventually rode to world dominance in 1980-90s. In both cases we saw a replacement of obligation-specific (in kind) transactions with obligation-generic (money) ones. As a consequence, a new class of specialists developed, mercenaries in the beginning 12th century, pure software firms in the end of the 20th century.

tags: business, model, innovation, history, software, industry, information, technology, computers

iPhone vs Android

Cnet put together a brief history of Android, starting with T-Mobile G1 in October, 2008. Since Android is software, not a device, the graphic below, showing Android development milestones, makes more sense. Further, in a real word comparing Android and iPhone timelines is a lot more interesting than just looking at Android alone.
The reason I'm not considering Windows Mobile, Nokia, and Palm is that they belong to the old Document-Mouse-Keyboard information access interface. After iPhone, Android became the second major technology that supported the new Stream-Zoom-In/Out application paradigm.

tags: technology, system, tool, evolution, mobile, software,  infrastructure, competition,  information

Второй раз на те же грабли

The latest numbers on mobile software are in:

Google's Android platform was running on 43.6 percent of all the smartphones purchased in the United States in the third quarter. It was followed by Apple's iOS, which captured 26.2 percent market share, and Research In Motion's OS, which tallied 24.2 percent share. Microsoft's mobile OS held 3 percent market share in the quarter.

For the second time in its history Apple created a computing platform, PC being the first one, and for the second time they could not hold the advantage. Unbelievable.

tags: mobile, evolution, business, apple, google, microsoft, computers,  software, control point

A $1.2B blunder

A funny paragraph from a CNet review of Windows mobile 7:

Above all, Windows 7 is--dare I say--elegant. Even my foreign-language spam looks beautiful on the device. It almost makes me wish I understood all those messages in Japanese, Korean, and Arabic.

Huh? Who cares about Windows 7 Mobile on a day when Steve Jobs introduces a whole slew of elegant mobile media devices? As far as mobile applications are concerned, Microsoft appears to be destined to fight with Google and Nokia over a distant #2 spot behind Apple. But that, in turn, means that HP's $1.2B acquisition of Palm last April looks like a total blunder. What kind developer in his right mind would spend any brain cycles on writing software for a marginal #5 OS?

tags: mobile, business, technology, market, 4q diagram, software,  source

Content as Software - 2

This is a follow-up to my yesterday's post about Content as Software (CAS).

Recently, Steve Jobs said that, despite abundance of Flash-compatible content, Apple is not going to support Adobe Flash on iPod or iPad. In his words, Flash is a CPU hog. Google expressed a similar view. Why is it so? Is one of the most popular web platforms going to die in the mobile space due to turf battles, or there's something else going on here? Is CPU hogging the only problem with Flash, or are there other reasons why it is going nowhere? To find answers to these questions, I looked up Flash timeline and compared it side-by-side with browser technology developments.

Remarkably, evolution of Flash looks like a continuous effort to overcome web browser deficiencies. First, it was a simple animation engine that allowed developers to embed dynamic content into static HTML pages. Then, media play-out and limited, but certain, scripting capabilities followed. At the time, web authors could not rely on consistent browser behavior beyond the very basics. In terms of browser as an application platform, Microsoft supported ActiveX in Internet Explorer (Windows-only), while Mozilla pushed JavaScript in Netscape and its open source descendants. MP3 audio was handled by a separate browser plugin, so it was difficult, if not impossible, to use most popular media to create a coherent browser-based web experience. On top of that, session management had to rely on cookies and annoying pop-ups, which could be easily disabled by the user. If you as a developer wanted to have a live page or stream content, you had to rely on Flash to provide a stable platform for your application.

Later, more video codecs, a full-blown object-oriented scripting engine (ActionScript), and web services integration layer were added to Flash. In 2007, after the acquisition by Adobe, new versions, Flash CS3 and beyond, could run whole applications inside its own application, had integrated support for Photoshop and other Adobe graphics manipulation products, 3D animation, and etc. The software has become a software platform that itself runs within a browser application that runs on top of a sophisticated windowing operating system, either Windows, Mac OS, or Linux.

A PC with plenty of electric power, processing power, memory, and storage space can easily afford this behemoth. A smartphone (or an e-book) cannot. And it doesn't need to, because its operating system, rather than a web browser, runs user applications. The OS, be it iPhone OS or Google Android, provides a consistent set of APIs that developers can rely upon when they write their code. Since browser on a mobile phone is no longer "The Web Application", but rather one of many web apps, there's no need for Flash to be a media presentation intermediary. The hog can be slaughtered.

Sales of smarphones are predicted to outnumber PCs by 2012. Unless a dramatic change in Flash architecture for mobile devices happens, the product is going to die a slow death. It's not about Steve Job's ego or his design preferences. Rather, it's about the process of creative destruction of obsolete technologies and business models.

tags: 10x, payload, information, mobile, content, software, cloud, system, evolution, niche construction, social, network

Content as Software (CAS)

An increasing number of providers are discovering that packaging their web content as a mobile application is a very attractive business proposition. The same content that is free on the net can be sold for real money when delivered through an iPhone app. NYT has a story about two such apps: Zillow (real estate) and Yelp (business recommendations). Both of them have a location-based component, which makes a lot of sense for people who have a purpose while on the go.

It's easy to predict that delivering content as software (CAS) will grow in popularity among users and providers. First, people are willing to pay for mobile apps; second, content providers have greater control over presentation (browser-independent implementation); third, CAS screens have a greater focus than PC, both time- and space-wise, which enables insertion of highly-relevant ads, including video. All signs point to a new stage in the development of internet commerce. For example, in the nearest future travel guides will stop being books and become integrated mobile apps.

tags: 10x, payload, information, mobile, content, software, cloud, system, evolution, niche construction, social, network

Beginnings of a tsunami.

That real-time collaboration is a thorny problem. It can be difficult to permit multiple people permission to edit the same document at the same time while ensuring one person's changes don't interfere with another's work. And showing simultaneous work complicates a service's user interface, too.

"With Google Docs it takes about 5 to 15 seconds for a change to make its way from your keyboard to other people's screens," the site [EtherPad] said. "Imagine if whiteboards or telephones had this kind of delay!"

We are witnessing the emergence of a new Payload. The web started with HTML, then moved to XML, and now we are entering the "Scripts+Data Streams" phase. All elements of the system, including browsers(script execution environments), servers, routers, and pipes will have to change to accommodate this step of the system evolution. Mobile devices are probably in the best position to take advantage of the trend.

In any case, whatever they say about cloud computing is just a small facet of what is coming to the world near you.

tags: computers, information, payload, evolution, problem, performance, software, niche construction

The software is the driver

Computers will play a huge role in the future of driving. Maybe GM should get out manufacturing and focus on software development. Its competitors surely move in that direction:

Audi achieves that torque number by using four electric motors, one at each wheel.

The four electric motors meant that Audi could design a virtual Quattro all-wheel-drive system, integrating the power distribution program with vehicle dynamics sensors. By default, the rear motors deliver 70 percent of the torque, the high number compensating for the 58 percent weight distribution to the rear wheels. But as conditions dictate, the power software can give any wheel greater or less power, which should make for incredible road-holding, beyond even Audi's current Quattro system.

The car's navigation and communication electronics are designed to communicate with external roads infrastructure, receiving information about traffic and green lights, for example, and adjusting routes accordingly to maximize driving efficiency.

As the amount and types of moving "informational parts" in the car increase, the value of the control function within the device will increase accordingly.

:: control, scale, information, payload, software

From "Cortical Computing", by Greg Snyder:

A back-of-the-envelope calculation is useful. A human cortex has a density of about 10¹⁰ synapses/cm². Today's microprocessors pack roughly 10⁹ transistors in 1 cm² of complementary metal oxide semiconductor (CMOS). Thus, to build biological-scale neuromorphic circuits, electronic synapses will have to be about one-tenth the size of an average transistor. This is one important reason intelligent machines are not (yet) walking around on the street.

A 10X change in computing is coming. From what I know, the hardware side of this change is going to happen within the next 10 years. But it will probably take a lot longer for the software to catch up.

Also see: DB Strukov & KK Likahrev, 2006. A Reconfigurable Architecture for Hybrid
CMOS/Nanodevice Circuits. FPGA’06, February 22–24, 2006, Monterey, California, USA.

Recently, I wrote about voice communications and books becoming virtual, i.e. an application on a mobile computer. Here's how remote control follows the same path (NYT):

And so now we have a bounty of applications and accessories that let us use the technology we already have to control the technology we already have. This is not only frugal, but upgradeable and flexible. Whether you want to control your music, your television or your PowerPoint presentation, there’s probably a solution using your phone.

 As we discussed in class a couple of months ago, iPhone is not a phone. Rather, it is a sophisticated software application platform. Apple is extremely smart in helping its customers to jump over the abyss between mobile voice and mobile computing in two easy steps: (1) iPhone -> (2) iPhone 3G. Neither Windows Mobile, nor Android demonstrate the ability to move seamlessly between the two worlds. 

Here's an early review:

 

Speaking of the App Store, it's easily the best thing about the iPhone 3G—although it also happens to be available on the old iPhone via the firmware 2.0 update. I've already posted my impressions of the App Store, plus a handful of the more than 500 available applications (Gina has also posted some reviews), but I'll say it again—if you're looking for something new to get excited about in iPhoneland, this is it. The 3-D, accelerometer-enabled games like Super Monkey Ball look amazing on the iPhone, the TypePad blogging app lets you post thoughts and photos on the fly, Pandora delivers free, streaming music channels, and Loopt lets you keep track of your pals in real time—and that's just the tip of the iceberg.