Mobile Devices
The author considers the current segmentation of the mobile device market to be unsatisfactory - "ill defined and irrelevant" - as they tend to focus on the devices themselves (product based) rather than the ways in which people user them (behavior-based).
The problem is compounded by a lack of consistent definitions in the industry: what is a PDA, how is it differential from a smart phone, and how that is different to a plain "phone" is entirely unclear even in the industry - with each company trying to define its device into the most attractive category, or redefine the category to exclude competitors against whom they compare unfavorably.
With this in mind, the author means to scrap it and start over.
A DEVICE TAXONOMY
The author seeks to escape the categorization of devices according to their capabilities and, instead, look to the needs of the market and the ways in which they actually use their devices. In effect, a person who purchases a "smart" phone and uses it only for voice and text messaging isn't really part of the smart phone market segment and shouldn't be considered as such - they have chosen the wrong device for their needs, and is more likely to recognize their mistake and downgrade their device than to begin to make use of features they don't want or need.
Another mistake is the acceptance of the notion that the mobile computing device market will converge on a single multi-purpose device, any more than the automotive market is likely to converge on a single vehicle design that serves every driver's needs. However, it's likely that the market will sort into a relatively small number of classes based on their needs. The author defines four:
- Work Devices - The device will be used in the context of the user's profession, and generally only while "at work." This includes professional correspondence, viewing work-related data, accessing company information sources, etc.
- Personal Devices - The device is used for personal ends, which may vary from communicating with family and friends, to using leisure and lifestyle applications (calorie counter, pedometer. etc.), managing personal finances, sharing photos, etc.
- Entertainment Devices - This device is used for leisure purposes only. Watching movies, listening to music, playing games, leisure "surfing"
- Targeted Devices - A device used for a single task, or a small number of related tasks, such as a digital camera that connects to an online repository and enables the user to post photos to blogs or community sites.
The first three categories are served by general-purpose devices are intended to provide an array of capabilities to serve the user's needs, typically in a given domain (work, personal, entertainment). While the physical form can vary, they tend to gravitate toward a computer-like interface with keyboard/cursor/screen as these are the common methods for doing a variety of different tasks.
The last category remains valid, as the evolution of devices has typically been for a specific set of needs to be served by a targeted device (such as a digital video camera) until its capabilities are consumed by a general-purpose device. Even then, the user may retain the targeted device because the general -purpose device dos not do a given task as well (in the example of digital video, phone cameras are low-quality and the devices have little storage capacity - which may be fine for some users, but others may prefer to retain a digital camera).
(EN: The example seems a bit frivolous, but targeted devices also exist in the workplace. A hand-held device that scans bar codes for managing inventory and communicates wirelessly to a vendor's ordering system is by definition a targeted device - whose capabilities could be later evolved into a smart phone application.)
The author also notes that ease-of-use may also be a factor that drives users to one device or another. A person may opt to purchase a "smart" phone because it's more convenient for viewing and composing instant messages, even if they don't need the additional capabilities; or a person may opt to use a targeted device because it's difficult to use the same capabilities on the multipurpose one.
Work Devices
Most modern workers employ some sort of computing device in their professions. While the PC remains common among office workers, even workers outside of that environment are making increased use of computing devices: retail clerks, auto mechanics, waiters, salesmen, security guards, truck drivers, and numerous other non-office professions are leveraging computer technology.
Especially for these tasks where the full functionality of a desktop computer is unnecessary, mobile devices may become more commonplace as a method of tracking and reporting work-related information. These are largely enabled by targeted devices, though general purpose devices may be used by individuals who need computer assistance for a variety of tasks.
Presently, computer manufacturers dominate the general-purpose work device market, largely by virtue of the need of the mobile device to communicate and receive information from back-end systems via proprietary software and protocols. It's also reasoned that, due to the amount of information that is required, these devices will remain fairly large to accommodate a visual display and keyboard.
The author also dismisses they hype about speech recognition: it has been promised for decades and still does not function properly. In some instances, it has been successful at dealing with a limited number of predictable commands with reasonable accuracy. But given the lack of progress, and lack of investment in solving the fundamental issues, it's likely that the keyboard will remain necessary for quite some time.
It's also noted that the "truly" mobile worker, whose work is done in the field (a delivery driver or a meter reader) or in a location where the public comes to them (a clear in a retail store) need applications that are suited to a highly social environment in which there will be constant distractions and interruptions in their work flow, a problem that software design doesn't currently address (EN: application-switching and state preservation on mobile applications is rare and poorly implemented).
Personal Devices
In industrialized societies, communications technology is ubiquitous. Everyone has access to some communications technology, even if it's merely a land-line telephone, and exceptions to this are highly unusual. Access to the Internet is also becoming ubiquitous, and instances in which a person does not have a telephone and internet access are virtually as rare as a person who does not have electricity or indoor plumbing.
Mobile communications are merely an extension of this facility, and largely a convenience. Virtually everywhere a person is likely to be, they are within a short distance of a public telephone - having one in their pocket merely saves them a short walk, though as mobile is becoming more ubiquitous, the number of public telephones is shrinking..
(EN: an important difference, however, is that a public telephone enables a person to communicate with others who are predicted to be near a phone whose number is known to them - but it does not enable others to reach them. So it can be said that the mobile device isn't a convenience to the user to contact others, but a convenience to others who wish to reach the user.)
The author uses the acronym "PCD" (Portable Communications Device) as a synonym for personal mobile devices, thought it stands to note that not all mobile technologies are communications technologies.
(EN: Here, the author revisits the qualities of personal. Communicative, handheld, and wakable, which were already mentioned in the first chapter - "what is mobile anyway?" section.)
The author also presents an interesting pointing terms of "cost." Land-line phones and service remain much less expensive than mobile phones. However, in terms of convenience, they a offer a greater value. For two people to communicate via land-line, they must both be physically near a telephone at the same time and one must know the other's number - in many instances, this is not possible, and even when it is possible it is not often convenient or reliable.
Entertainment Devices
Entertainment devices largely evolved as targeted devices from analog counterparts: the walkman became the MP3 player, the film camera became the digital camera, etc. General-purpose devices evolved by consolidation of these devices into multi-functional ones: a mjusic player that takes pictures, an e-book reader that also enables you to surf the Web, a device that does all four.
The author draws a line between "primary" and "add-on" features, indicating that most devices begin with an essential function and others are bolted on later. The add-on features are generally less functional and more difficult to use because the device is designed for a primary purpose that cannot be compromised to accommodate the "also" functions.
There have been attempts at making devices truly multifunctional, but as a rule, any device that attempts to make all features equally good ends up making them all equally bad. As such, devices appeal to the specific needs of a user. If reading e-books is more important than surfing the Internet, the consumer will seek a e-book reader device that "also" has internet access; if the priorities are reversed, the consumer will seek a net-surfing device that "also" reads e-books.
Targeted Devices: the Information Appliance
Targeted devices are designed to help the user to a small number of specific tasks, and are generally designed to be highly effective and efficient for their precise purpose. In many instances, such devices evolved from mechanical or electronic devices for the same purpose: the mechanical watch became a digital watch, the film camera became a digital camera, the tape player became an MP3 player.
Historically, and in many cases currently, the value provided by a targeted device is self-contained: your wristwatch need not synch the time with your car's clock, nor do users expect a pocket calculator to communicate data with a laptop computer. On the other hand, certain "single use" devices have perished: people no longer purchase stand-alone word processors, and an attempt to develop one that would not exchange data with other devices would not be likely to enjoy commercial success.
The most basic functions provided by such devices are frequently included in other devices. A clock, for example, is often included in a computer, a coffee maker, a conventional oven, a television set, a car stereo, and numerous other devices. In some instances. the ability to monitor time is requisite to the value of the device (a coffeemaker that starts brewing at a specific time needs to have a clock); in other instances, it is merely a convenience to the user (while driving, checking the dashboard clock is easier and safer than glancing at a wristwatch).
When developing products for the current market, it's suggested that the value of a targeted device is in its ability to do a given task very well - better than the capabilities provided by general purpose devices. Also, given that most users have general purpose devices, there is little point in attempting to bolt "also" functionality onto a targeted device (e.g., the ability to manage your calendar on a digital camera is unnecessary because you already do that with a cell phone). Again, the value is doing a specific task well, not a lot of tasks poorly.
The author notes that "developers frequently want to add on features," and suggests that third-party developers (contractors) are often the "most notorious" for wanting to add on additional features. (EN: I think this may be a matter of perspective - in my own experience, it's often been marketing, or even executives funding the project, who make heavy-handed suggestions for additional functions - either to make the device more competitive with other companies' products, or just because they think it would be cool.)
From a design perspective, if you're unable to resist pressure to add "also" capabilities, your goal is to tuck them out of the way, so they don't interfere with the primary function of the device.(EN: But again, this leads to bigger problems. The person who had the bright idea doesn't think people can find it, and insists that is be made more prominent. There is no simple tactic for negotiating, especially if the person making "suggestions" is in a position of authority.)
In terms of design, targeted devices have fewer size issues than general-purpose ones. In many instances, the screen can be very small, or omitted entirely. The author gives the example of the iPod Shuffle, a tiny device with no screen and few controls (Apple's "click-wheel"), but which served the needs of its users perfectly.
It's also noted that applications for targeted devices tend to be device-specific. If the device communicates data, the informational interface may need to conform to certain standards in order to be compatible with systems that send/receive data, but there is no logical reason for using a standard technology or approach to anything that is inherent to the device - instead, use what works best.
Ubiquitous Computing
The notion of "ubiquitous" computing refers to the embedding of computer technology throughout the user's environment. The "computers" are not analogous to personal computers, but networks of sensors, indicators, and communicators. It's also been referred to as "calm" computing, which communicates information at a glance.
One example is the "Ambient Umbrella", which turns blue if rain is in the day's forecast - the user need do no more than glance at the umbrella on his way out the door to tell if he ought to take it with him. There is also potential value in the interplay of small devices, to communicate information to a "dashboard" display.
The author looks at four aspects of ubiquitous computing that will likely be relevant to the "mobile device ecosystem" in future:
- Pico Nets, also called "personal area networks," involve connectivity between the various devices a user may be carrying, to share and synchronize information. In some instances, an infrared "beam" can be used to share information among devices, though the Bluetooth technology was created to address the same need. However, device manufacturers have been reluctant, for reasons of security (or profit), to enable their devices to communicate with other manufacturers' devices. However, consumer demand is overcoming this, and a device that communicates openly is valued over one that does not.
- Home Servers. A common problem with integration among various devices is a lack of a central system to house data and coordinate information. This is becoming common for media (a few products enable you to store video in a central location to make it available on your various televisions, computers, and other networked devices), and it's a short step to including other forms of data.
- Shared Displays. One of the drawbacks to owning multiple devices is that each must have its own display screen. The ability for a device to display information on the screen of another device (such as the smart phone) would mean that the various devices would be smaller and less expensive.
- Public Interaction Points (PIP). This is vaguely defined, but seems to mean that in certain public spaces, information can be communicated between the user's device and a computer system For example, rather than consulting a directory in an airport to find a gate, various nodes in the environment could push information to direct the user to where he needs to go, just as GPS systems work in vehicles.
ANATOMY OF THE PCD
In most instances, discussions about "mobile" are specific to personal communications devices (PCDs) in the hands of private individuals, used for personal reasons. This is the most lucrative market for applications developers and an attractive opportunity for companies that wish to have a "direct line" to promote their products to potential customers.
The author regurgitates the personal/communicative/handheld./wakable list, with special emphasis on the "personal" (it is used by a specific person and not shared) and the "wakable" (it's always on and can demand attention), and then mentions that the PCD is a "general purpose" device as defined earlier in this chapter.
Aside of that, there are a number of aspects that bear consideration:
The Carry Principle
One key aspect of the PCD is that it is with the user at all times. This is largely unique to the PCD, as individuals do not carry other devices (such as a digital camera) at all times, and it's not uncommon for their "work" device to be turned off or left behind outside of working hours.
The author also brings back the Swiss Army knife analogy, in that a person will usably accept certain limitations to the capabilities of the device in exchange for having a basic level of functionality at their disposal at all times, in case they should need it. Also, the PCD is typically a phone and/or text messaging device, and any other feature is an "also" rather than the primary value of the device.
Input Mechanisms
The author details some of the common input mechanisms for PCDs:
An important input device, and one which is often overlooked and neglected, is the "focus control" facility - the method that the user employs to determine which of the device's function-sets to utilize at any given time: do they wish to use the device as a telephone, a text messaging device, a music player, or a camera? The most common methods are a mechanism to scroll through functions and select, whether done with a click-wheel or a touch interface. Sometimes, this is a specific affordance (the "rocker" wheel that allows movement up/down and left/right) or it may utilize other controls (such as volume up/down buttons) to navigate a menu of option. The use of speech commands and keyboard shortcuts are not unheard of, but are rare.
Command "buttons" are also common. For example, there is typically a "go" button that activates the most obvious or common command at a given time (to take a photo when in camera mode, dial a number in phone mode) and a "stop" button that cancels and action and/or returns to a previous screen or menu option.
There are often "soft keys", similar to the function keys on a computer, which have different functions (or none at all) depending on which application is in use. They may be physical or virtual buttons used to access specific functions (like back/next/stop/refresh when using a web browser). These are usually device-specific, and while they are intended to have universal functions, third-party software often breaks or ignores the intentions of the device designer.
The author mentions that some PCDs have voice commands, but it's noted that it's not a feature many users value, or that owners report using, and it's typically a source of more frustration than convenience, given the lack of accuracy in voice recognition algorithms.
Most devices have a tiny "chicklet" keypad (or virtual keypad) for text entry, which is notoriously difficult to use, but is considered a vast improvement over using the standard telephone numeric keypad to enter text (to enter an "R", press the number 7 three times). It's noted that most users take the device in both hands, typing with their thumbs, or hold the device in one hand and use the index finger of the other, when using a keypad.
Various companies have experimented with one-handed text input, either through modified keypads or "chording" techniques, but the notion has not caught on. Handwriting recognition has also been attempted by a number of companies with varying success. (EN: I recall once having a Sony Clie, which did a wonderful job for me, but others with the same model had a lot of complaints, so it probably depends on the person.) Offering full-sized keyboards (even flexible ones that can be rolled up) also failed to solve the problem - while they were easier to use, it required setting down the device, connecting the keyboard, and using both hands to type, which is not suited to the way user utilize a mobile device. Also, algorithms that attempt to predict words to save the user from having to type the full word have had entirely unsatisfactory res7lts.
The author also lumps together an array of different input devices under the category of "environmental data." The camera, microphone, RFID reader, GPS system, thermometer, gyroscope, accelerometer, and other devices are all considered "input" that collect information, whether through active means (you must push a button to take a picture) or passive (the GPS system is always on). Some of the applications that attract the most attention rely on these alternative input modes, but they also tend to be passing fads for most users.
Finally, the ability to connect the device to another computer is another form of input mechanism. In some instances, the device can be connected to a personal computer via cable or wireless connection, or they may connect to external systems via their networking capabilities to load data into the device.
Output Mechanisms
The most obvious output mechanism of a mobile device is the screen, and most mobile devices have evolved to include an LCD screen of some sort (though with some devices, it's quite small.) However, the screen is inherently problematic: it's too small, consumes a lot of energy, and its readability in bright lighting (such as outdoors) is poor.
Some work has done to address these problems, but they have not been entirely or effectively solved, merely mitigated. And some of the solutions have introduced new problems: for example, a screen that fades or dims after a few seconds of inactivity makes the mobile device impractical to read more than a small amount of text.
The author lists some of the technologies under development:
- The OLED (Organic Light Emitting Diode) has better visibility in sunlight, less power consumption, no polarization issues, and better visibility in bright conditions, but is more expensive and has a shorter lifespan.
- Electronic paper displays effectively use a physical "ball" (a very small one) white on one side, colored on the other, that is magnetically spun to show the colored side. It uses less energy and is ideal for bright environments, but is not effective in dim environments. Also, the current refresh rate of electronic paper is relatively low - four frames per second, which is sufficient for text but insufficient for multimedia.
- Electrowetting is a technology that uses a porous surface and fills tiny reservoirs with a colored liquid (mercury or oil), resulting in excellent color and low power consumption. It's a promising technology, but still in the very early stages of development.
Some devices have multiple displays (e.g., a clam-shell phone with a small display on the outside and a larger one on the inside), and could choose the appropriate technology for each display.
In addition to the screen, most devices also have audio output, either through a speaker on the device or a wired/wireless earpiece. The quality of the audio output varies, but as many mobile devices include music player functionality, it tends to be rather good when played through headphones. The quality of the speaker, whether the earpiece speaker or speakerphone capabilities, also vary by device, but are generally sufficient for their intended purpose - listening to a human voice.
Depending on the device, there may be a separate speaker for the "ringer," and there is typically a buzzer to produce pulses or vibrations for silent operation - however, not all manufacturers provide access to third-party software to access the ringer/buzzer. The same may be true of various LED indicators on the phone - their presence and software accessibility is varied.
Certain connection technologies such as BlueTooth, WiFi, and infrared can also be considered output methods, though they are used for output to other devices rather than the human user, and the receiving device must be capable of receiving.
Technologies
The author uses "technologies" to cover the operating system and various applications that are pre-installed on mobile devices. Any add-on application will be compelled to leverage the operating system, and should be developed to leverage native applications rather than duplicate their functionality (e.g., to use the phone's native "camera" application to take a picture rather than attempt to have its own functions for operating the same functions.)
- Most devices have native messaging service: the ability to send and receive SMS text messages is virtually ubiquitous. A standard has been defined for media messaging (MMS), but it is not as widespread, largely due to cost and interoperability issues. Also, there has been some thinking for voice messaging, but this is largely redundant to the voicemail capabilities of phones, and has not gathered much support.
- Media recorders are common. Most cell phones have a built-in camera and the ability to take still photographs. Video capture (with or without sound) is becoming more common, but because video formats consume a lot of storage and processing memory, capabilities are limited. And audio recording capabilities, while seemingly a good fit, are not as common.
- Media players are also common. Many mobile phones have the ability to function as a digital music player (MP3) and have native applications to display pictures and play video clips in standard formats.
- Most devices have some species of Web browser application, through these tend to be device-specific applications rather than mobile versions of desktop applications. Also, most are stripped-down: they support text and images, but media support is unreliable. (EN: support for Flash animation and games, which are fairly popular on the internet, is virtually nonexistent on mobile devices)
Aside of the native applications, most mobile devices have one or more platforms to support third-party applications: Java ME, BREW, Trigenix, Python, and Linux are some of the common "open source" platforms, and manufacturers often have a device-specific application platform - which have the benefit of having deeper access to device capabilities in exchange for having to develop a device-specific application.
(EN: The author goes into a bit more depth about several of these technologies, but my sense is that it is highly perishable information, and application platforms remain in a state of flux.)
Connection "Characteristics"
There are numerous problems with wireless data networking. Primarily, network access is unreliable and inconsistent - the user may not be able to get access in a given location, and the network may slow down or drop their connection at any time, whether the user is in motion or standing still. Even when the connection is made, there is considerable lag in data interchange. Also, the cost is prohibitive for a data plan, and roaming charges can be significant.
(EN: These issues are often understated, or dismissed as temporary issues that will eventually be addressed - but I'm not confident in such assertions. It took less than five years for the Internet to evolve from spotty modem service to reliable broadband, whereas cell phones are about 25 years old and still have not matured. My sense is so long as customers tolerate it, the connectivity will not be addressed - but "tolerance" means accepting that certain things cannot be done on a mobile device, which may be damaging to user experience and their willingness to trust the channel.)
While these problems persisted, application developers will need to consider them - specifically, to build any application that requires data to be persisted for a session in such a manner that interrupting the session won't cause data to be lost or compel the user to start over. It's suggested that this is the reason that native applications, which can store data on the device, are still preferred over web-based applications for mobile users.
Standby Screen
A brief mention is made of the "standby screen" that is displayed when the device is powering up (or down), etiehr from being turned off or, more frequently, from sleep mode. Currently, some devices allow the user to customize this screen, using photos or wallpapers or whatnot, but advertisers are greedily eyeing this "real estate" as a way to send a message to the customer before they can do anything else. (EN: I don't see users tolerating this, but anything is possible. Especially if there is a potential to offer free or reduced-cost service to the user in exchange for consent to be marketed, it could catch on.)
Some devices enable the user to access certain functions without leaving the "standby" screen - for example, if the user begins entering numbers, the device assumes them to be placing a phone call, and may either skip them to the appropriate screen or allow them to do so directly from the standby screen. In another example, one company provided the ability for the user to add application icons directly to the standby screen to enable the user to skip the menu system.
(EN: I'm not sure that this remains true. Most devices require the user to unlock them by entering a password before they can place a call or access any information on the device, for security reasons. Switching from standby to main menu is a visual cue that the user has entered the password correctly. I'm not aware of any device that returns the user to the standby screen after keying a password, and that would seem to be counter-intuitive.)