Through the Internet, humans have connected the world. People are closer to each other than ever while still remaining apart. The next phase of the the Internet will be about connecting things. The Internet of Things will be central to the infrastructure that we build. (The "Futurist's Cheatsheet" series surveys technologies on the horizon: their promise, how likely they are, and when they might become part of our daily lives. This article is Part 5.)

What Is It?

Think of a thing. Really, it could be anything. A chair, a toaster, parts of a car, the lights in your house, the electricity meter, the security cameras in your offices, a fire hydrant, traffic lights … really, anything or everything that can exist could be connected to the Internet. Another name for the Internet of Things is a network of things. The network can monitor your home, your car, infrastructure (utilities such as electricity or water), traffic patterns and a variety of other possibilities to create a more informed and responsive system through data analysis. 

How It Works

Do you really need an Internet-connected toaster? Probably not. But, the toaster is a good place to start when discussing the Internet of Things. 

What would you expect from a smart toaster? Perhaps a touch screen on which to schedule cooking. It could be connected to the coffee pot, enabling the perfect breakfast for you as soon as you wake. Your toaster could be programmed from your computer or a mobile app. Say you are laying in bed and know you are going to sleep in the next day, pull out your smartphone and reprogram the toaster to start an hour later.

A toaster could have its own IP address on the Internet. In theory, you could visit your toaster’s site. Giving things a full IP address is one way to tie a thing to the Internet. Another way, and the way in which many things will be tied to the Internet, is for a thing to just have the ability to connect to the Internet, without and IP address.

Now, imagine that there is no digital interface on your toaster. In this case it is just a toaster that happens to have cellular or Wi-Fi capabilities and sensors to monitor how well it performs. It sends sensor data back to the manufacturer through Internet nodes and portals without an individual IP address. The manufacturer uses this data to know how its product is working in the wild, how often it is used, and use this data to make a better toaster. 

Go back and replace the word toaster with anything, say, a power meter. The same concepts apply. An Internet of Things can use the Web as an interface, or just use the Internet to move data. That data can be used to interact with the network of things or just as a pipeline where data moves two ways, analyzed and used to make objects smarter and more responsive to people’s needs. 

Potential Impact

There are so many ways that an Internet of Things could impact people’s lives that it is hard to describe everything. Distilling it to a few key areas helps define what the scope of an Internet of Things could be: infrastructure (buildings and utilities), consumer (cars and homes), health care and businesses (consumer products and retail locations). 

Weather-related sensors could help agriculture by monitoring the moisture in the air or ground and give farmer’s warning about droughts. Smart buildings can provide enhanced security for the people that enter them or warning on disasters such as earthquakes. Connected cars can improve traffic flows or allow functions to be controlled remotely. Items within the home (such as the toaster) can be controlled and monitored and even connected to each other. 

Health care is an interesting avenue for the Internet of Things. Certain aspects of the body could be connected to the Internet. Heart sensors could give patients and doctors data to prevent disease. Sensors that monitor white blood cells could give cancer or AIDS patients warning of a relapse. 

The scope and impact of the Internet of Things is almost limitless. It is just up to the innovators of the world to be creative and find ways to make it work.

Challenges

Much of the base technology that will enable and Internet of Things is available. The challenge now is to refine that technology and make it ubiquitous. 

A truly connected society involves a concerted effort from many different industry sectors such as telecommunications (the lines that would do the actual connecting), to device and appliance makers that would implant sensors and connectivity into things. Software developers would then have to create the interfaces. There are also security and privacy issues, such as keeping this mountain of data safe and away from prying eyes. Wireless standards and infrastructure also need to improve to handle all of the data that would be generated. 

When Will It Be Ready?

Many of the innovations we have written about in The Futurist’s Cheat Sheet have seeds in today’s technology. That is the same for the Internet of Things. The technology is present, but the infrastructure and stability behind it needs to be improved.

Companies specializing in machine-to-machine functions such as Numerex and KORETelematics are already in the process of designing the connected world and building business models that will help define the Internet of Things. 

The progression will be slow. There is no event horizon where suddenly the technology that is only a theory becomes a reality.

The Internet of Things is something that must be built and refined, not something like quantum computing that is waiting for a significant technological breakthrough. In five years we will start seeing more connected cars and homes.

Infrastructure like smart grids and utilities will take longer to build and we will see it evolve over the next 10 years and more. The Internet of Things will become embedded in our lives and the growth will not stop during out lifetimes. 

Additional Information

European Commission: Cluster of European Research Projects: Vision and Challenges for Realising the Internet of Things (March 2010)

IEEE: Architecture and Protocols for the Internet of Things: A Case Study (2010)

GigaOm: United States of Connectedness: What works for Internet of Things

ReadWriteWeb: Top 5 Web Trends of 2009: Internet of Things

ReadWriteWeb: Top 10 Internet of Things Developments of 2010

ReadWriteWeb: Internet of Things Explained (Video)

Movies are filled with big, bright 3D holographic images. Whether it is Tom Cruise manipulating items out of thin air in Minority Report or Robert Downey Jr. going all mad scientist in Iron Man, 3D images have little real-world equivalent.

We can dream though.

Holographic and more advanced volumetric displays are just but a twinkle in scientists' eyes. True 3D projections for commercial or industrial uses is still years away. (The "Futurist's Cheatsheet" series surveys technologies on the horizon: their promise, how likely they are, and when they might become part of our daily lives. This article is Part 4.)

What Is It?

You remember in Star Wars Episode IV – A New Hope when Leia sent a message to Obi-Wan through R2-D2? The little picture of Leia imploring, “help me Obi-Wan, you’re my only hope.” That was a holographic display. Taking holographic displays further, there is also the notion of volumetric displays that create a life-like 3D object. 

According to 2002 research done by Actuality Systems, “volumetric displays create imagery that appears to float in a volume.” The image can be seen from a variety of angles without screens or goggles. 

How It Works

There are two ways that volumetric displays can be created: swept-volume display and static-volume display. Swept-volume displays use the persistence of vision to create volumetric images from “rapidly projected 2D ‘slices,'” according to Gartner’s description of in the 2012 Hype Cycle report. This can be done by projecting images onto rotating mirrors within an enclosure. Static-volume displays use “voxels” (the volumetric equivalent of pixels) to create 3D images. 

Potential Impact

A holographic display would allow you to talk with a representation of a person through a communicator terminal, like the small disk that Darth Vader used to communicate with Emperor Palpatine. A volumetric display could go further, showing a person that looks real but is instead only a visual representation. In the short term, there is little practical application for these technologies. 

The advantages of a high-definition volumetric display (HDVD) could be applied to a variety of industries. Drug creation, medical imaging, oil exploration, mathematical analysis, airport security and product design could all benefit. 

Challenges

Volumetric displays are expensive and difficult to produce and are barely beyond research phases. Holographic displays are a little more common, such as in retail environments. Perhaps the most impressive commercially produced holograph is Hatsune Miku, a complete holographic Japanese pop star that performs live on stage. Much of the challenge involved is finding a business model that supports the technology. In relation to our current technology (we are just getting used to FaceTime), there is as yet little need for holographic or volumetric displays.

Holographic and volumetric displays require a lot of computing power and moving parts, which could make systems unstable. 

“The use of holographic techniques is, by far, the most advanced, but due to the intensive computing calculations required to generate the holographic image, the cost of generating the display image is one of the hurdles toward real-life adoption,” wrote Gartner’s Stephen Prentice. 

When Will It Be Ready?

Very simple volumetric displays have been prototyped, and there are basic holographic technologies in use. But the full vision of the technology is much further over the horizon, perhaps 20 to 30 years.

While some commercial uses of holographic technology will be seen sooner than later, the true depth of volumetric displays will take a lot of time to develop and make commercially viable. The advantages of 3D displays, such as visualizing large data sets will eventually be a significant way in which research is conducted. It is cool to think of, but you will not be talking to your mother via hologram any time soon.

Additional Information

Actuality Systems (2002): 100 Million-Voxel Volumetric Display (PDF)

Stereoscopic Displays and Applications (2003): A Static Volume 3D-Laser Display (PDF)

Spie.org: Volumetric 3D display provides true perception of objects

Moore's Law describes the phenomenon that makes this year's computer more capable and less expensive than last year's. But it won't go on forever. While engineers have come up with various schemes to keep it rolling, quantum computing is the best hope for extending it indefinitely. The concept has been proven in the lab, but working quantum computers are not a foregone conclusion. Here is a quick-and-dirty primer on a very complicated technology. (The "Futurist's Cheatsheet" series surveys technologies on the horizon: their promise, how likely they are, and when they might become part of our daily lives. This article is Part 3.)

What It Is

Computers as we know them manipulate information encoded in ones and zeroes, known as binary code. Instead of the customary binary computation, quantum computers will encode data in the quantum states of subatomic particles known as quantum bits, or qubits. These devices will be able to execute parallel computations orders of magnitude faster than today’s computers.

The website AskAMathematician.com describes quantum computing, “A quantum computer can take many inputs, do many calculations, and produce many results at the same time.”

The concept has been validated by systems involving very small numbers of qubits. However, cobbling together full-scale quantum computing systems remains a distant prospect.

How It Works

Today's computers are built on silicon-based chipsets whose physical features are becoming smaller every year. At the same time, the speed at which these circuits run becomes faster, leading to ever more tiny and powerful devices.

Quantum computing takes the action of computing from the molecular scale of silicon to subatomic scale of particles such as electrons or photons. At this scale, paradoxical quantum phenomena come into play. A qubit can be either a one or a zero or both at the same time. Add qubits one by one and the computational possibilities start to rise astronomically.  

Potential Impact

There will be no greater advance in computing than quantum computing within our lifetime. In many ways, it is the Holy Grail of advanced mathematics. Once scientists can create quantum computers that are stable and reliable, advanced math that is nearly impossible today will be relatively trivial.

What are the possibilities of such computation? Really, there could be no limit. Scientists could figure out how to build smarter computers that could do anything asked of them (think Star Trek: “computer, make me a cup of tea") or design engines that could make deep space flight feasible. One of the most interesting possibilities of quantum computing would be the creation of true artificial intelligence. 

Challenges

Building qubits is extraordinarily difficult. Researchers have completed promising basic experiments, but the largest quantum computer to date has incorporated only seven qubits.

Generating qubits and getting them to work together is only half the job. The other half is coaxing them to turn input into meaningful output. Researchers are working on algorithms that form the basis of quantum digital processing.  

Timeline

Gartner's latest Hype Cycle report puts quantum computing in the “more than 10 years away" category. That might be generous, as the technical obstacles are daunting. Quantum computing on a large scale may never come to fruition. The earliest implementations will be in supercomputers unavailable to the general public. In the meantime, we will just have to make do with standard computation.

Further Information

Quantum computing as a vast field with new research published almost every month. The overview above is but a simple attempt at an explanation. Check out the resources below for additional insight into quantum computing.

Ars Technica: A tale of two qubits: how quantum computers work

Ask A Mathematician: How does quantum computing work?

How Stuff Works: How Quantum computers work

Andrew Steane, University of Oxford (1997): Quantum Computing

The problem is not new. One way or another, people have to validate their identities. I am trying to enter a building or a Web service that only Joe Smith should have access to, I need to offer evidence that I am, indeed, Joe Smith. For decades, authentication has required cards and passwords. In the near future, you might just use a part of your body. (The "Futurist's Cheatsheet" series surveys technologies on the horizon: their promise, how likely they are, and when they might become part of our daily lives. This article is Part 2.)

What Is It?

Use a thumb-print to unlock a door, an iris scan to unlock a smartphone. Maybe use your voice to interact with your mobile device, PC or television. Biometric data can be used for verification (say, allowing access to a personal bank account) or identification (say, identifying you to law enforcement agencies). 

How It Works

Pick a body part, any body part. There is a good chance that it has a unique identifier that can be used authenticate an individual human. Of course, not all body parts have practical applications in all situations. For instance, hormone analysis would be an awkward choice of authentication for entry to a building. 

Criminal forensics provided an early proving ground: Identification based on fingerprints became a viable form of authentication in the late 1800s. DNA performs much the same function today. 

Cloud technology is giving rise to new, ubiquitous forms of biometric authentication. Physical identifiers for large groups of people can be uploaded to a server and used for purposes such as accessing data on a company computer, gaining access to secure buildings or unlocking smartphones. Storing biometric keys in the cloud makes it much easier for devices to recognize and recover the data and for users to put it to work.

Potential Impact

The rise of a digitally connected society has led technologists to propose the notion of “one true login.”  Today, you may have one password for Facebook, another for Gmail and so on. At the same time, you may have an ID card such as a driver’s license. Depending on where you work, you may have an ID badge that you have to scan to get into your office. What if all of these functions could be replaced with one biometric identifier unique to you? 

Such an innovation could improve personal and data security an dalso improve user experiences across a variety of devices. Much of modern computing has been built around the standard user interface: keyboard and screen. That is starting to change as computers, smartphones, tablets, and televisions incorporate cameras that recognize your face, touchscreens that know your fingerprint and microphones that recognize your voice. Quick, convenient biometric authentication would tie these devices more seamlessly into daily life.

Challenges

The technology for biometric authentication is already widely available. The true challenge comes in building an acceptable infrastructure where the technologies can be easily implemented. Part of the challenge is cost in replacing or augmenting legacy authentication methods such as the magnetic keycard system in a hotel or an enterprise. Another challenge is legal. Many states and countries have privacy laws on how certain types of biometric identifiers can be used, inhibiting how enterprises and commercial ventures can deploy these authentication methods. These privacy laws are important as people are extremely sensitive in how their biomedical is stored and used. 

When Will It Be Ready?

Research firm Gartner focuses on the future business aspects of biometric authentication in its most recent Hype Cycle report, but the consumer realm poised to see practical applications. Smartphones can be unlocked through a variety of biometric keys such as voice, facial recognition or a fingerprint. Apple, Samsung and Microsoft will likely lead the way. Companies like Nuance are tuning mobile devices to the user's voice. And enterprises won't be far behind. Before long, companies will implement biometric authentication for onsite building access and smartphone security.

Additional Information

Book -- Anil K. Jain et al. -- Introduction to Biometrics

Michigan State University -- Partial Face Recognition: Alignment-Free Approach

Microsoft Research --  Progressive Authentication: Deciding When to Authenticate on Mobile 

As long as there have been humans, there have been dreams of super humans. Eyeglasses started sharpening vision in the 1200s, pacemakers have been implanted to extend lifespans since the late 1950s, and the first strength-amplifying robotic exoskeletons shipped earlier this year. But those innovations are only the beginning. With advances in technology, the ability to vastly enhance human capabilities is right around the corner. Here is an overview of current efforts and their potential. (The "Futurist's Cheatsheet" series surveys technologies on the horizon: their promise, how likely they are, and when they might become part of our daily lives. This article is Part 1.)

What Is It?

Human augmentation is the ability to supplement human brains and bodies with technological improvements. The notion has been part of science fiction lore for decades. Ever hear a sports announcer say, “that guy has a cannon for an arm!” Well, what if he had an actual cannon for an arm? RoboCop, Mr. Gadget, Star Wars (what is Darth Vader but an augmented human?), and the Bourne Identity all offer visions of how human augmentation could one day be achieved. 

But it's already well underway. With the buzz around Google Glasses and Oscar “Blade Runner” Pistorius' speedy artificial legs, the notion of creating a better human body through machinery and computers is the subject of much theory and research these days. 

How It Works

There are many paths to human-machine augmentation: wearable technology such as Google Glasses, sensor implants, using DNA and chemical processes to enhance brain function and muscle functions, nanorobotics, performance enhancing surgery. Some theorists, notably Ray Kurzweil, believe the brain will be encoded as software someday, allowing it to be reprogrammed, enhanced by peripheral technology, tethered to a robotic body, and immortal. (Until the next backward-incompatible system update.)

Potential Impact

The idea is to enhance the human notion of “normal.” At the same time, human augmentation can be used to repair parts of the body, such as cochlear implants for the hard of hearing. Laser eye surgery is a good example of both reparative and enhancive human augmentation, as it could be used to help the sight of the visually impaired or enhance the vision of people with normal eyesight. Many professional athletes, such as baseball players, get laser surgery. 

Research firm Gartner notes that there will soon be a market for human augmentation to create “superhuman” characteristics, such as a suit that improves endurance or adds extra senses to the body. There have also been recent advances in implantable technology that can monitor health-related data, such as heart rate or insulin level. 

The Challenge

Miniaturization and advances in wireless technology enable many sensor-based technologies to be implanted into human bodies now. Moreover, the combination of computer and genetic technology could enable people to retrofit themselves with superhuman characteristics going forward. 

In the short term, researchers are working with the tools already available. Advances in mobile technology and wireless data transmission along with sensor enhancements are creating a new field in the biomedical industry. As scientists continue to crack the human genome, DNA augmentation will become increasingly powerful and controversial. One day, doctors may be able to completely rebuild body parts with computer and mechanical engineering and have them look and function just like normal flesh and bone.

In the long term, society will be challenged to cope with superior human beings. The notion of a mechanically-enhanced human has already entered the thoughts of lawmakers. Several U.S. states have passed laws banning employers requiring employees to implant computer chips in their bodies. As human enhancement becomes more common in the decades and centuries to come, there is a real danger of discrimination between the augmented versus the standard human.

When Will It Be Ready?

Depends on the type of capability you are looking for. Strength-enhancing exoskeleton suits have been sold to the military and rehab hospitals. Google Glasses should be released as a consumer product in 2013 or so. Implantable, sensor-based technologies are just starting to hit the market. This is one field to keep an eye on as technology and biology merge to create the true ubermench that Nietzsche surely knew was coming.

Further Information

Technology Review: In Pursuit of Human Augmentation

Wired: Be More Than You Can Be

Bloomberg: Advances in Human Augmentation: We Can Rebuild Him

Association for Computing Machinery's Augmented Human

Wappwolf and ifttt are a bit like the chocolate and peanut butter in Reese's Peanut Butter Cups. Separate, they're pretty good. Together, though, is when the magic happens.

For folks new to the services, Wappwolf is an application that performs actions when you drop a file into your Dropbox. ifttt, on the other hand, can interact with websites and services and then do things like send an email or save a file in Dropbox.

Let's take a look at a several ways to combine the services for maximum effect:

For example, Wappwolf can watch Dropbox and convert files to Kindle format and then send them to your Kindle. You could use this to save company documents for later reading on your ebook.

ifttt might watch Instagram or Flickr for photos, and save them to your Dropbox when they appear. You could use that to make sure you never miss a polar bear picture again.

Separately, they're pretty handy. Together, you can do some really interesting things.

(If you haven't tried Wappwolf or ifttt before, don't worry. They're dead easy to use. See the How to use Wappwolf Automator page, and/or the About ifttt page.)

1. Send a Text Message When Files Are Saved to Dropbox

Say you share a Dropbox folder with co-workers, and you want to be notified when there's a new file in the shared folder. ifttt only has triggers for new files in the public folder for Dropbox, while Wappwolf can take action based on pretty much any folder you give it access to.

Start with Wappwolf and create an action for the appropriate folder. Choose the "email it" action and set up an email to go to trigger@ifttt.com (be sure to set up your email address with ifttt). Set the Subject to #Dropbox Notify, and the body to the filename (or whatever you like). You can add whatever you like to the body, but you probably want to uncheck the "Add file as attachment" box.

Next, go to ifttt and choose the Email channel. Choose "Send ifttt an email tagged and use #Dropbox as the tag. Create the trigger and then go choose your action channel, in this case SMS. Choose "Send me a text message," and complete the task. Now when a file is saved in the Dropbox folder, you should get a text message.

Don't care for text messages? You could also use ifttt to send Boxcar notifications if you prefer that app to texts.

2. From Google Reader to Kindle

iPads are great for reading, especially with the Reeder app. But for reading longer text, nothing beats the Kindle. What we'll do here is save an article from Google Reader to a PDF, then convert that and send it to your Kindle. This one should be very easy, because ifttt already has a recipe for it, so you're halfway there. Just visit the recipe and create the task.

Next, head over to Wappwolf and pick the folder you specified, then tell Wappwolf to "Send it to your Kindle." Add the action, save, and you're ready to go.

3. Copy Files from ifttt to FTP Servers, Box or SkyDrive

ifttt supports saving files to Dropbox, but it doesn't support the breadth of services that Wappwolf does. If you want to copy files over to Microsoft SkyDrive, Box, or upload them to an FTP server, you'll need to combine ifttt and Wappwolf.

Create any action in ifttt that saves a file to Dropbox in ifttt, then select the folder you're saving to in Dropbox to create the Wappwolf action for saving to the appropriate service.

Of course, you're not limited merely to uploading files from one service to the other. Wappwolf lets you combine actions, so you can save a file to Dropbox, use Wappwolf actions on the file, and then upload it to its destination.

4. Process Files via Email

Wappwolf works on files only when they're saved to Dropbox. Usually that's not a problem, since Dropbox clients are available on just about every popular platform. But what if you want to use Wappwolf with files coming in via email? No problem. You can use ifttt to save to a Dropbox folder, then create the appropriate Wappwolf action.

This might be useful if you want, for instance, to email yourself a file using a chron job on a server. Just attach the file to the email, mail it to ifttt and have it saved to Dropbox, then use Wappwolf to perform whatever actions you like.

Patience Is a Virtue

Note that some patience is required in setting up tasks with Wappwolf and ifttt. It can take a few minutes for each service to perform its tasks. For example, the chain required to save a file to PDF via Google Reader and ifttt, then convert and send to Kindle, then for Amazon to ready it for the Kindle took about five to seven minutes. That's not a long time, but it's slower than some folks might be used to with most modern Web services.

Have any cool tips that you want to share with the rest of the ReadWriteWeb audience? We'd love to hear about any cool hacks that use ifttt and/or Wappwolf. Share and enjoy!

The disk maintenance tools that Microsoft ships with Windows have always been, at best, tolerable. Now that there's an entire industry centered around archival storage systems and services, it's about time Microsoft gave its consumer versions of Windows a file archiving system appropriate for the 21st century.

Replacing Windows Backup, Windows 8 File History is the file archiving system that should have been in Windows 7 - and it points the way toward a post-PC future for Microsoft.

Ironically, the component that provides this functionality actually is part of Windows 7, and has been there since Windows XP. Windows has actually had a versioning system for files for a few years now. It's called Volume Shadow Copy, and it's a way for the system to maintain multiple backups of a file for different points in time. Windows 7 uses VSS (yes, that's the correct abbreviation... go figure) to back up certain system files that may need to be called back from the archives when you execute a System Restore - when you undo changes to the system, rolling them back to an earlier point in time.

Windows Backup can use this service... kind of. I don't know anyone who actually does this, but once you back up a folder to an archival device, in Windows Explorer you can right-click on a file in that folder, click on the Previous Versions tab, and recall an earlier version of that file. Of course, if you're going to restore an entire subdirectory full of documents, right-clicking and restoring each one this way is not a great way to spend one's weekend.

This is what a developer would call a service without a real interface. The real UI for this service has been added to Windows 8, and yes, folks will note that its inspiration probably comes from Mac OS. It's called File History, and it replaces Windows Backup. That name change alone will confuse some folks, and perhaps a shortcut from Backup to File History would help in the final edition.

The idea behind File History is this: If you continue to use Windows 8 the way Microsoft wanted you to use Windows 7, then you'll have bound your important personal folders to libraries. Your Office documents will be in Documents, your digital camera files will be in Pictures, your downloaded and transcribed videos will be in Videos. So it's silly to have to tell a different Windows program which files are important to you if you've done it once already.

File History already knows what's important to you. If there are files in your libraries that don't need backing up, you have the opportunity to make exceptions by clicking on Exclude from File History and adding exceptions into the box above.

This becomes important for the following reason: File History is designed to be something that works continually - not every week or overnight like an '80s-style backup, but every hour. The suggested use case of File History is for you to plug in an external hard drive via USB, but you can also map a cloud-based Microsoft SkyDrive location to a Windows network share. File History instantly becomes a cloud backup service. For Windows Phone users, that could mean anything you save to your media libraries will be automatically synced and available to your phone.

It also raises the question of bandwidth consumption. If you frequently copy over videos from your camcorder (as opposed to just your phone), then you won't want to use File History as a cloud backup service for that purpose. Unfortunately, that may impede your choices for backing up smaller media files and everyday documents - your choice of backup devices may apply to all your backed up files. If you've decided to use an archival drive like a Seagate FreeAgent to archive your camcorder videos, you won't be able to use File History to back up your documents and tunes.

Once again, the tool that's shipped with Windows is not perfect for all situations. Just as it's been for the last three decades, that leaves an opening for third parties backup providers - like Acronis, for example - to build a market.

Still, the evolution of Windows Backup into File History is important as the nerve center of people's digital lives shifts away from the PC. If Microsoft wants to maintain a handle on its customers' everyday life and work, it needs to stake a firmer claim on the services and tools that bridge all the devices that are at the nerve center now.

That means Microsoft needs a stronger cloud presence. SkyDrive is nice, but it's not as powerful a product as Dropbox or Box.net. The new File History is a compass pointer in the exact direction that Windows needs to go: toward a service that transcends both PCs and devices. As I've said before: Not Windows Phone, not Windows PC. Just Windows.

Perhaps you've seen the famous comic posted to Oatmeal.com titled How to Fix Any Computer. Not to give away all the secrets of the comic's trenchant forensic analysis, but Step 2 of the Windows side of the equation is unfortunately familiar to just about any Windows user: "Reformat hard drive; reinstall Windows."

A PC operating system is like steel wool. You can't use it in even the slightest way without mutating it. Installing a new program typically alters the System Registry, which to many Windows veterans even looks like steel wool. Inconsistencies in the Registry can affect the entire system, and much of the last 17 years of Microsoft's development of Windows has been devoted to adjusting, accounting and compensating for these discrepancies so that folks don't have to reinstall Windows every time something goes wrong. System Restore (a form of which premiered with Windows Me) was created to overwrite a newer, possibly damaged Registry with an older, hopefully undamaged copy, in hopes that the system could pretend the changes suspected of damaging the system never happened.

It's taken well over a decade for Microsoft to guide the evolution of Windows software to a state where applications are maintained separately from the operating system. We're not quite there yet, though we're close enough now that Microsoft feels comfortable introducing this useful "partial reinstall" feature to Windows 8.

Called Refresh, it's based on the Windows Image Manager (WIM) services introduced with Windows Server 2003 R2. Refresh replaces the kernel files in Windows, overwriting the existing installation with a new and freshly compiled image of the operating system. But using WIM, it adds back the separately maintained components from the new architecture of Windows 8, including existing WinRT "Metro-style" apps you've already installed. So you don't have to start all the way at square one, though it won't necessarily take you all the way back to where you started.

Test Setup

My test computer for this experiment is a quad-core Intel Core i5 2500K desktop system. I purposely did not use a fresh Windows 8 Consumer Preview installation, but rather one where I'd made software installations and changes, including:

• Microsoft Office, after I authored several documents and made some settings changes. I expected to have to reinstall Office, but I wondered whether it would remember me afterward;
• The Visual Studio 11 beta, plus some of the SDKs that go with it;
• Mozilla Firefox, along with some settings changes that I believed should get stored in a safe directory that survives the operation;
• A third-party screen capture utility called Screenshot Captor whose stored settings I also suspected should survive;
• A third-party utility I use to install software directly from ISO images of discs, called Virtual CloneDrive;
• Some Metro-style apps.

I also made the kind of adjustments that an ordinary user would make. I changed the Desktop wallpaper - which, for Windows 8, applies only to the Desktop mode. I moved a Vista-style Desktop clock gadget from the left to the right of the Desktop (would it get moved back left, or would it disappear altogether?). I also changed my Metro and Start screen background color from crimson (go Sooners!) to cyan (umm, go Seahawks). And I tweaked the network settings to add more sharing features to my homegroup.

I chose these actions to give me clues as to what parts of Windows get overwritten during Refresh. Since this operation will probably be undertaken only when Windows is acting weird, you should actually want some parts of Windows to be wiped clean. If too many things survive, the bad behavior might survive as well.

Test Results

The Refresh process consumed about 20 minutes - a little slower than the original install, but still well within reason. It wasn't long ago that a Windows XP installation could take more than three hours.

For the Windows 8 Consumer Preview, here's what I noticed immediately after Refresh:

1. My Desktop wallpaper was the same, as were customizations I had made to Windows Explorer (such as "View hidden files and folders"). This indicates that at least some Registry entries survive Refresh. Not all entries, however: My Control Panel, which I intentionally reset to View by Large Icons, reverted to View by Category. Administrators will want a hard-and-fast rule for which settings need to be checked and reset, and can be counted on to survive Refresh.

2. My Desktop gadget disappeared, suggesting that contents of the Windows folders were completely overwritten. That's actually good, because misbehaving components may be parts of system drivers or even malicious actors placed in system folders to feign legitimacy.

3. The Metro color reverted to crimson, even though my Metro background color remained cyan as the Refresh program was announcing success for the initial sign-in. Easy enough to change back, but the point is that the surviving data for the Desktop and the victim data for Metro appear to have come from two different places.

4. The Metro Start screen and Lock screen pictures stayed the same. Evidently, Metro doesn't store all of its data in the same place. This may pose some interesting problems with respect to System Restore (not Refresh), and the possibility that rolling a system's status back to a previous restore point may not restore Metro with the same integrity as it restores the Desktop.

5. The homegroup had to be rejoined, although Metro did remember the homegroup password. This is important for a slightly esoteric reason: If you have a dual-boot PC that also boots with Windows 7 (many people will, and I do), changes you make to the folders that Windows 8 includes in shared libraries appears to impact whether those same folders remain shared in Windows 7. I have no clue why or how this is so, but since Windows 8 folders start life as private and unshared (as they should be), rejoining the homegroup may mean you have to make adjustments in Windows 7 the next time you boot it up. This quirk may go away with the final edition of Windows 8.

6. Third-party application data ("AppData") was cleaned from its hidden folder. That's both a surprise and a big deal, because it means that not only will you have to reinstall your programs, but you'll have to start over with your settings. If you had a huge store of bookmarks in Firefox, and you weren't syncing it via some cloud service, it's probably lost. And if you had certificates, serial numbers or other data affirming your rights to use commercial software packages, they may be gone as well.

7. The contents of users' Documents folders remain intact. This is as Refresh's warning promises. When I installed the Visual Studio 11 beta, I had it create a number of sample files. Even though I typically keep my "My Documents" folder on a completely separate drive (which has saved me more times than I can count), VS tends to put its help files in the local "My Documents" folder on the system drive anyway. That said, they were intact, even though Visual Studio itself was not. Still, this is a good thing, because getting VS functionality back from here takes only minutes.

8. There's a nice "Removed Apps.html" file on my Desktop listing everything that Refresh had to remove, which I can print out and use as a checklist.

9. My IE10 Desktop home page reverted to Bing.com. Yeah, I caught that, Microsoft. Sneaky devils.

10. My Windows ID remained intact, as it now must in order for me to be able to log onto Windows 8 again. This also means my Xbox Live account remained intact, and Windows 8 games like PinballFX were able to sign in for me. For some users, having the Xbox Live profile will be item No. 1 on the checklist. However, things stored locally - such as my high scores on PinballFX - did not join the party.

One of the first programs I had to reinstall was Screenshot Captor, in order for me to take pictures for this report. The program was able to find my saved settings, which do not rely on the Registry. On the other hand, Firefox thought it was being installed in a completely clean system. So evidently, the rules for reinstalled programs needing a re-education in Windows will be... complicated.

Refresh or Reset?

The big question for many users will be whether Refresh will be any more of a timesaver than what Windows 8 now calls Reset. Reset is a fast way to start over with a completely new and unadulterated installation of Windows 8. Whatever settings you may have had are removed.

The "Reset your PC" warning (shown above) states, "All your personal files and apps will be removed." That's not exactly correct. If you sync your files using cloud services such as Apple's own iCloud, Box.net or Dropbox, or if you keep your important files on a separate drive, naturally, those files would survive even a Reset operation.

Reset wipes the My Documents folder when, and only when, it cohabits the same device as the system folder. Microsoft should consider selectively revising this warning. Perhaps: "All your personal files and apps stored on the same device as Windows will be removed." Even that may be harsh, because (we've been told) the new Windows Store may be called upon to reinstall lost Metro apps.

For me personally (this will not be the case with everyone), the difference between Refresh and Reset is minimal. Reinstalling Windows has become as common for me as, say, washing the car. If I have a tool that polishes the chrome for me so I don't have to, I might appreciate it a bit, but I won't call it a lifesaver.

But for most Windows users, anything that reclaims an hour or more of valuable work time will be as good as gold. If Refresh works as well in the final edition of Windows 8 as I believe it will, fewer folks may find themselves, as the Oatmeal chart so quaintly put it, quietly weeping.

Gartner. Forrester. IDC. And lots of smaller fish, too. You can't read a tech-industy news story, attend a conference or listen to a sales pitch without someone quoting an industry analyst. For tech companies, analysts are big news and big business, promising to help with transformation, monetization and a slew of other things ending in "-ation."

But what do technology industry analysts really do? And how do you find the one that's right for your company's needs. Let me try to explain, from the inside. You see, from 1999 through 2001, I was an analyst at Jupiter Research, now part of Forrester Research.

It's a tough job to explain. Analysts write and speak and pontificate - all the things you see in the news - but their true value lies elsewhere.

Companies pay analysts the big bucks to provide educated gut checks before making major strategy moves, for help mapping competitive landscapes and to get the dirt on vendor features and pricing you can't find anywhere else.

Analysts deliver numbers to justify your hunches, and they can even take the fall when execs need to save face with board members or investors. The best ones make life easier and more productive, but how do you tell which analysts are worth your time and money?

Well, analysts are fond of lists, and old habits die hard, so here's a list of key criteria:

Don't Believe the Hype

When I was young and green, I spoke to a crowd of IT managers about testing tools. They nodded, took notes, asked questions and seemed satisfied with my answers. An analyst from a rival firm followed me and ranted for 20 minutes. He was way out of his depth in the Q&A, and was obviously making up answers when he got stumped. He fell back on "You don't understand - the Internet changes EVERYTHING!" several times. The guy was a jerk, but I felt bad for him.

When the lights came up, the "jerk" had 15 people waiting to speak with him. I had four. Afterward, he shared some wisdom I'll never forget.

"Always tell them it's different. Especially when it's not." He may have been a goon, but the guy knew how to sell. Seeds of doubt grow into juicy contracts, and for marketing purposes, a controversial statement is worth 10 correct ones.

Marketing is the highest-profile part of an analyst's job. Don't get sucked in by the controversy. When evaluating an analyst firm, you can safely ignore just about any controversial headlines or projections, particularly if they show up in the subject of a press release. Whether you agree or disagree with an analyst's published findings, you'll need to dig deeper to see whether there's a useful business partner underneath.

The one exception to this rule is a finding so patently off-base that it's just silly. You know the ones: "80% of commuters will be using hovercrafts by 2015." In these cases, the analyst lacks either a brain or a spine. Any way you slice it, that's bad news.

Examine the Goods

When you buy a subscription to an analyst firm, you're buying access to the analysts themselves. The research is secondary.

Before you plunk down cent one, set up a call or face-to-face meeting with the analysts who interest you. You might not get to meet everyone you'd like, but the sales team will bring someone with an A-game. This will give you a good idea of the firm's character and how well its analysts can work with you.

Assuming an understanding of your business and their coverage areas, the basic ingredients of what an analyst has to offer are pretty simple. Look for communications skills, a network of vendor and client relationships, and a healthy dose of business sense. The trick is finding someone who balances all three in a way that works with your business.

1) Communication

Your first introduction to an analyst will probably be a written report. This can help you assess the firm's target market and general level of technical understanding, but a face-to-face meeting is essential to gauge his or her communications skills. You should feel comfortable explaining your issues, and the responses to your questions should seem considered.

The analyst may not be able to answer all of your questions, and that's fine. An "I don't know, but I know who to ask, and here's why" is infinitely better than getting steamrolled with pat answers or desperate fakery.

Beware of overused buzzwords and cliches, and don't be afraid to ask analysts to explain their statements in greater detail. Absolutes and one-line mandates are for sales pitches and keynote addresses, not one-on-one conversations. Analysts who won't listen to you probably won't listen to vendors or customers either, so they won't have a lot of secret wisdom to broker.

2) Connections

Prominent analysts at powerful firms command greater respect from vendors and can dig more details from them. This doesn't mean the analyst needs to be in bed with the vendors it covers. Some of the best, most respected firms are the most resented by vendors. For example, Real Story Group (formerly CMSWatch) is fairly small, but it covers only content management, and it's never taken a dime from a vendor. This focus and objectivity has earned the firm an incredibly devoted following, forcing CMS vendors to show their cards.

On the flip side, an analyst who consults with a large number of clients in a particular area will be able to broker a lot more collective experience on your behalf. One disgruntled customer of Vendor A could be a fluke, but knowing that 10 have complained to your analyst in the past week could save you from a very expensive mistake.

This is probably the biggest value an analyst brings to the table. We all learn from mistakes, and a well-connected analyst is learning from the mistakes of dozens of clients all the time.

3) Business Sense

Analysts will never know your business as well as you do, and you shouldn't expect them to. But they should be able to see patterns and linkages you haven't. The best question you can ask an analyst about anything during a meeting is "How does X apply to me?" X can be a report, a competitor's press release or something you heard on the news. It's always a valid question, and the answer isn't always the most important thing.

What matters is that the analyst listens to your question, considers your specific circumstance and comes to a logical conclusion based on the information provided. If you get an answer that's counterintuitive but still makes sense, even better. You want someone who can shake things up. If you get back a rehash of the same pitch you heard from the sales rep, move along.

Analysts aren't superheroes. At their best, they're smart people who sit in the middle of a lot of valuable information, which can make them a valuable asset to your business. And doing a little homework can help ensure you get what you pay for.

Installing SPDY Packages

Right now, Google is providing binary packages for Debian or RPM-based systems with Apache 2.2.4 or greater. If your system is Debian or Fedora/Red Hat-based, you should have little trouble getting SPDY running. If not, Google also provides the source code for the SPDY module, and you can compile the module if necessary.

Before installing SPDY, you will need to have an SSL certificate to use SPDY since it uses SSL. This can be a self-signed certificate, though Chrome will give some pretty evil-looking errors when accessing a site with a self-signed certificate. For testing SPDY, this is merely a small annoyance. If you're actually looking to deploy SPDY to a production site, though, it's probably much better to go ahead and buy a certificate signed by a recognized authority. (There is a way to turn off SSL, but it's not recommended for anything but debugging.)

If this is the first time you've set up SSL on Debian-based systems, see the Debian Administration tutorial on configuring SSL. The Linode library has a Fedora 14 tutorial that should work with later versions of Fedora. See the CentOS guide if you're using RHEL, CentOS or another RHEL-based distribution.

Next step is to get the right package for your system. My server is a Linode VPS and runs Ubuntu 10.04 LTS, and it's a 32-bit system. The 32-bit Debian package worked just fine for that, but Google also provides a 64-bit Debian package and 32-bit and 64-bit RPMs as well. All of the packages are on the mod_spdy page under the Google Developers site.

You might run into a few snags if you're using an RPM-based system, like openSUSE, that isn't quite what the Google packagers were expecting. Nick Robinson has provided some instruction for getting the binaries installed on openSUSE 12.1 that may prove useful if the RPM doesn't work for you.

For Debian systems, you should use dpkg to install the module, like so:

dpkg -i mod-spdy-beta_current_i386.deb

Note that that's the current package name; it will likely change soon when Google decides to move the module from beta to stable. (One hopes this will happen faster than Gmail moving out of beta status...)

Google also says that you need to run apt-get -f install, but this had no effect on my system. Presumably, some systems require installing additional dependencies.

Next, you'll need to restart Apache for the new module to be recognized. The packages automatically add the spdy.conf and spdy.load files for Apache, so that SPDY will start automatically when Apache starts.

Testing

Once you have the SPDY module installed and enabled, use Chrome to visit a page on your site with HTTPS. You can go to chrome://net-internals/#spdy in Chrome to see whether SPDY is enabled, and chrome://net-internals/#events&q=type:SPDY_SESSION%20is:active will show live and active sessions.

I do recommend doing extensive testing before pushing a SPDY deployment live. So far, I've been having no problem with my WordPress blog, but I have found a few glitches with my Piwik install. Apparently, other users have encountered this as well. I suggest watching the mod-spdy-discuss list if you're working with SPDY.

If you decide you don't want to use SPDY until it moves out of beta, you can turn it off without needing to remove the packages. Open the spdy.conf file (under /etc/apache2/mods-available on Debian systems) and turn SpdyEnabled to off. You'll need to restart Apache for that to take effect.

There's talk of SPDY becoming an official Apache module (though it's written in C++ and official modules need to be in C), so it may be even easier to get SPDY in the not-too-distant future. Until then, it's still pretty trivial to install SPDY, and you should be able to have it up and running in less than 30 minutes.

Those brickbats have often been well-deserved, but RIM should also have earned some respect, if not love, for the important role it played in smartphone development and popularization - not to mention a string of iconic-at-the-time devices that significantly advanced the state of the art.

I can still remember how excited I was back in 2006 when I got my first BlackBerry. Half the size of a paperback book, the BlackBerry 8700 had virtually no multimedia capabilities. But I could do my email anywhere I went. Just as important, that push email capability worked better than any mobile email I've used before or since. (I did miss my Palm Treo 650, but that's another story...)

I was even more excited in 2008 when upgraded to a BlackBerry Curve. Now I had a real screen that could show pictures, and a way to play music on my phone, not to mention a few games and apps. Oh, and a camera. A surpassingly BAD camera, but still a camera that I used to snap - and email - photos everywhere from San Francisco to 12,000 feet up in the Chinese Himalayas. And because it ran on BlackBerry Server, it wasn't fazed by Gmail or Facebook blockages.

Yak Meadow, via BlackBerry.

But I was already realizing that my Curve was no longer on the leading edge. My friends all had iPhones, and while they couldn't do their email as fast or as securely as I could, they could do a lot more things that seemed more and more important.

That mattered to me, but it didn't seem to matter to RIM. So in 2010, when I finally got the option to choose my own smartphone, I didn't even consider getting an upgraded BlackBerry - only whether I needed an iPhone or an Android. I've never looked back.

And yet...

Now that RIM is on the ropes, it's all too easy to recall the things that I will miss about the BlackBerry:

A physical keyboard. Sure, I've gotten used to the virtual keyboards on modern touchscreen phones. And it turns out that I don't do all that much typing on them anyway. But when you needed to write on that phone, nothing beat the sure feel and positive feedback of a well-designed physical keyboard. (Of course, RIM played down that advantage as it introduced models with only standard phone keyboards or touchscreens.)

Secret confession: I miss the little wheely thing on the side of early BlackBerries that let you quickly and accurately scroll among the various menu options. The little ball in the center of the Curve was never as easy to use, and touchscreens are a whole 'nother ballgame.

Blackberry Messenger. Sure it worked only with other Blackberry users, but it was an awesome implementation of instant messaging. Way better than any third party solution on the Blackberry or other phones I've used.

A screen you could actually use in broad daylight. That screen may have been blocky and low-res, but at least you could see what was on it.

An email-first mentality. I'm old school. I still rely on email. And while I get email on my other devices, I still appreciate how RIM put email front and center on the Blackberry.

Battery life measured in geologic time. I never ran out of power during a long day on my BlackBerry. Never. I could even skip a day - or two - of charging and still stay in touch. Try that on your iPhone and you're carrying around a pretty little glass brick.

The ability to concentrate on my work without the distraction of all those apps, websites, games, and whatnot. Except, of course, for...

Brickbreaker. I spent hours sliding that little paddle around the screen, and got some decent scores. Problem is, being good at Brickbreaker is like being good at pinball - a sure sign you have too much time on your hands.

But that's just me. Enterprise communication managers will have many more reasons to mourn BlackBerry's passing. Corporate uses were key to RIM's DNA - and that's something no one ever says about Apple or Google. RIM always took enterprise security seriously, Apple and Google had to be dragged kicking and screaming to even think about the issue. If RIM is no longer a viable competitor, how long will it take Android and iOS to resume ignoring security and manageability?

So even if RIM stops making Crackberries, I'm truly hoping that BlackBerry Enterprise Server doesn't go away. In fact, I'd like to see RIM put all its emphasis on making BES the back end for all kinds of corporate mobile communications, supporting all kinds of mobile devices.

That would be a much smaller - though perhaps more profitable - company, and management still defiantly claims no interest in going down that road. But while few folks will miss BlackBerry devices, many businesses would miss BES. Focusing on the corporate market is the company's last, best hope to remain relevant. Because that little wheely thing ain't coming back.

Lead image courtesy of Shutterstock.

An amazing run, and it wasn't all just luck, though the company enjoyed plenty of that. To boost its chances to win the startup lottery, Instagram did eight very important things right.

1. Instagram Operated as a Nimble Start-up

Despite its meteoric growth, Instagram kept its overhead low with a total of only 13 employees, and its headquarters weren't anything to brag about, either. Instagram is housed in Twitter's old digs on 164 South Park Street in the trendy but still slightly seedy SOMA neighborhood of San Francisco.

More important, it stayed nimble enough to switch gears from its original idea of a location-based social network called Burbn to an iPhone app focused solely on photo-sharing. And instead of hiring workers or looking for revenue, the company was able to focus on growing the platform, its community and its aesthetic offerings.

2. Instagram Cashed in on the Mobile Explosion

Instagram's timing was perfect, capitalizing on a historic increase in smartphone use. Starting out as an iOS app, it successfully targeted the burgeoning legions of hip iPhone-app users. It waited until Android reached nearly 50% mobile marketshare to launch the long-awaited app on that platform. Meanwhile, Instagram mostly ignored the no-longer-as-hot Web space, although there is a Web version called Webstagram.

3. Instagram's Interface Stayed Junk-Free

Spend 20 minutes on Instagram and chances are you'll be hooked. Beautiful pictures filtered through hazy lenses flow down an otherwise cold, glass screen. Instagram charges up a smartphone screen, filling it with emotion - not piles of distracting controls and icons. On Instagram, the image stands alone. Instagram successfully created a new world where images - not functionality - are the main focus.

4. Instagram Didn't Get Creepy

Instagram managed to keep its creep vibe low. It doesn't push users to post, to share or even to like - it is an open space, available to use as you wish. Stay on the sidelines and just observe - no one will ever know. Pop into a conversation and then quickly leave. You won't be the only one: Instagram users are notoriously fickle.

That's why Instagram's easy follow/unfollow option doesn't require a commitment as heavy-handed as "friending" on Facebook. Follow whomever you want, or just wander off and find new images to look at. You won't accidentally see some old fling pop up in the Instagram stream, unless you purposely follow that person.

5. Instagram Married Visceral, Visual Communication & Community

Instagram built a devout community based on a single idea: capturing and sharing beautiful images. Powerful images are inherently emotional - think about The Atlantic's In Focus section.

Instagram gave people an easy way to connect around images, without the added pressure of complex social relations. Communities popped up organically around filters, around using too many filters and around just friends.

"There are a lot of things in [a] photo that someone can respond to, [that] promote conversation - then you get a wonderful interaction out of it," says Piictu Community Manager Zachary McCune. "I hope that continues, because that's what's beautiful about being able to relate to photographs."

A talented photographer can capture a single emotion in a square image, and make the viewer stop and feel something, if only for a moment. Instagram makes that moment easier to share.

6. Instagram Created a Valuable New Data Set

Every time an Instagram user snaps a picture, the app can capture a rich set of data, including location, time of day and other data points that can be associated with a smartphone's sensors, explains ReadWriteWeb's Dan Rowinski in How Instagram Will Help Facebook Monetize Mobile.

Instagram may not have been monetizing that data, but you can bet Facebook will. "Facebook is adding another crucial set of data points/edges to analyze people's activity online," says Pixable CEO Inaki Berenguer. "With Instagram in the fold, Facebook can now quantify what people are taking photos about based on the tags they put on their photos."

7. Instagram Didn't Worry About Making Money

In its short 15-month lifespan, Instagram focused on building its community, user base and functionality rather than worrying about how the platform was going to make money. By not chasing the dollar, it was able to focus on making the app better.

Of course, Instagram did pay attention to acquiring VC funding. Right before the Facebook acquisition, it closed a $50M Series B round from Sequoia, Thrive, Greylock and Benchmark, valuing the company at a sweet $500 million.

Figuring out revenue streams is now something that Facebook COO Sheryl Sandberg and her team will have to worry about.

8. Instagram Embodied a Cultural Shift Toward Photo Inboxes

As the future of photo-only inboxes approaches, and more people use photo-sharing apps to connect with their friends, the need for more visual communication will only intensify. The idea that social interactions will increasingly become visual and mobile is attracting widespread interest.

"Increasingly, we see people using photos to share an experience with a friend, whether it's to tell them what they had for lunch, show them a cool spot in their city that they found or to share special moments like birthdays or weddings," says Pixable's Berengeur. "The ultimate goal of sharing mobile photos is to broadcast your life to your friends instead of keeping a memory. The rise of smartphones, photo apps and social networks have made taking and sharing a photo easier than ever."

What Other Startups Can & Can't Learn From Instagram

What worked for Instagram isn't guaranteed to work for every startup. It's always easier to look back and uncover the right and wrong strategic moves. Only in retrospect is it clear that Instragram did everything right. Along the way, there were plenty of critics carping at just about every decision the company made.

There is, however, one thing that other startups can certainly learn from Instagram: Timing is everything. If Instagram had come along six months later, things might have turned out very differently. Some other company might have already executed on its ideas and it would be the one preparing to cash some very big checks.

Instagram images via @bralts.

This level of personal interaction is precisely what makes consumers not entirely trust push notifications. It is a mixed bag: They love notifications when they are useful, hate them when they become a vehicle of spam. For developers, this is a fine line. There is a fundamental disconnect between technologists and consumers when it comes to push notifications. Many developers think push is a wonderful, useful tool. Most consumers would prefer to be left alone. Developers: what do you think consumers think of push notifications? That is the subject of this week's ReadWriteMobile poll.

Not All Push Created Equal

There are many subtleties to the push notification conversation. How, when and why to deploy push notifications depends on the type of app (a news app, social app or game, for instance) but also on the type of platform they are being pushed from. Android, iOS and Windows Phone all handle push differently and that can be a source of frustration for developers.

The consensus among many developers that I have talked to is that the way push works on Android is the most preferable in terms of user experience. For instance, if an app sends 20 push notifications to an Android smartphone, only the most recent notification will show in the users' message tray. That is when Android push notifications work; between the three operating systems, it is also the most unreliable. Notifications sometimes get lost or take a long time to show up.

In terms of a purely technical distribution, iOS has the most stable push notification system. Notifications almost always arrive in real time and are gathered in a drop-down notification tray that was released with iOS 5.0. iOS notifications are also the most intrusive. Unlike Android, all notifications are shown in the tray, and they tend to build on top of each other. Many an iOS user knows what it is like to have that little red number hover over an app that constantly sends push notifications. When the notifications never stop, that can be a frustrating experience.

Windows Phone has the oddest push system: Only the most recent applications will be allowed to send push to a user. This is governed by how many "Live Tiles" a user has one their phone. With Windows Phone Mango, only 30 Live Tiles are permitted (up from 15), creating a bottleneck of endpoints for developers sending push notifications. Users with highly customized Live Tiles may not receive push from apps they do not create a tile for. Windows Phone also has several different kinds of push displays, including "Deep Toast" and multi-tile notifications. Android and iOS also have gradations of the types of notifications that can be sent.

Chart: Push Notifications Overview from Microsoft

Different Use Cases, Different Reactions

Some push notifications are exemplary uses of the technology. Others are borderline and can be annoying, depending on the user. Others are outright spam.

In the industry, the best use of push notifications are often cited to come from apps like The Weather Channel and Words With Friends. If there is severe weather, like a large tornado coming my way, you best believe I want a timely push notification. The Weather Channel is parsimonious about how it sends notifications, usually only pushing news when something dramatic is about to happen. Words With Friends is the best example of a game using push to tell a player when it is their turn. It is one of the features that makes the game so addictive.

On the other hand, we have games like Urban Crime from Gameloft that sends constant, intrusive push notifications that are highly annoying. This is where the line gets tricky. To a certain extent, Urban Crime is just trying to pull you back into the game through game mechanics. On the other hand, Gameloft sends too many, and it is hard to make them stop. Other entities, such as the new Color app, send push notifications whenever a user "visits." The only good thing about Color's use of push is that nobody actually uses the app, so you will not be bombarded with a constant flood of notifications.

"From a gaming, social content, news methodology, if they are tailored to be pushed when I have said they want to be pushed, it is fine. If you are hitting me up with ads, if you are hitting me up with information that is not useful to the core experience, consumers really hate that," said Bill Gianoukos, VP engineering and product management at Boston-based HeyWire.

Developers believe that there are distinct uses for push notifications and put trust in users to know how to turn them off, if wanted. This may be putting too much trust in the casual user that is more likely to delete an app that sends too many notifications.

"I think that Apple's intended purpose with them has kind of been shaded. Right now, predominantly, most games I see them being used as spam," said Adam Telfer, VP of game development for Toronto-based XMG Studio. "In our own games, we try to only use them for game mechanic reasons. So, you think about a game like Words With Friends, they are only using those push notifications for pulling the user back into the game because they have a new move. Not because there is some sale on an item or [other instances]."

Image: Powder Monkey game logo from XMG

Gianoukos offers great advice on the ideal scenario to use push notifications:

"Ideally, push notifications are meant to enable the users to get the content or the event that occurred that they want to know about sent to their phones so they can be prompted to go into the application. So, it is a prompting mechanism, ideally based on user preferences of when they want to be interrupted and notified of event, data, content or something occurring that is important to them, and they need to go into the app to interact with that piece of information," Gianoukos said.

Developers: What say you? Do consumers want push notifications? Where is the line between trying to re-engage users and spamming them? Vote in the poll and let us know in the comments.

Images: Kids pushing and sale courtesy Shutterstock

Mobile app testing is not easy. Whenever I talk to a developer, testing inevitably is one of the phases of creation that is both exciting and excruciating - exciting because a product is finally near completion, excruciating because... well, if you have ever tried to dig unknown bugs out of software, you understand. To do testing right, it is a multi-layered approach that takes time, resources and patience. There are several ways to go about app testing and not all situations make sense for every developer.

Multi-Layered Testing

We enlisted IBM's Leigh Williamson, a Distinguished Engineer and a member of the CTO Team, to guide us through the different ways that developers can go about app testing. Testing is not as simple as a developer trying to troubleshoot the native code on an app. Mobile apps are often vertical software systems with a variety of moving parts on the front end, in the middle and in a back-end cloud. Your code on the device may be running perfectly fine, but you would never know it because it is being corrupted from the back end that feeds it information. Or maybe some of the middle-tier services, like third-party SDKs, are running improperly. When something goes wrong, sometimes it is easy to figure out what is broken. Many times, you have no idea.

There are a variety of ways to automate testing. The most painful way to do it but probably one of the most essential is the manual test. Put the code on the device and use it like you would in the real world. When something goes wrong, log it and dig back through the code.

"There is manual testing for mobile applications. Even though that is the most inefficient and costly, we do still think there is a place for it in a well-balanced strategy. Because we really haven't gotten a way yet that we've seen that can automate the valuation of the usability and consumability of the general end-user experience for the mobile app. That is something that a real human being has to evaluate ultimately," Williamson said.

In an ideal world, manual testing is one of the last steps in the application development lifecycle. Everything has been built and it is almost ready to ship. There are ways to make it less painful, such as services from startups like Boston-based Crashlytics that provides a SDK that will tell you why your app crashed to the precise line of code along with all the environmental data at the time of crash.

Is That Cloud Really a Wrench in Your Machine?

As mobile app developers turn to the cloud, there is then another tier that needs to be tested. In IBM's mind, this is where its strength in the testing realm lays. The company can simulate or isolate the middle and back-end tiers to give developers an idea of how the app will perform before making actual integrations between device, servers and clouds. This is function of IBM's acquisition of Green Hat, a testing platform for a variety of cloud services. A "back-end as a service" mobile cloud service like Kinvey will give developers an instance of Kinvey's cloud structure (that runs through RackSpace, Azure and Amazon) that runs on the desktop to perform a similar service.

"What that allows an organization to do is to actually set up a continuous test agile methodology for developing the mobile application, checking in some code change which triggers in an automatic build of the mobile binaries for the app for the different mobile devices, and push the updated version of the application to some real mobile devices, and execute the code on the mobile devices with a simulated middle tier and back-end tier because it is always going to respond to the mobile device the same," Williamson said.

A Cloud of a Different Sort

A developer recently lamented to me about testing, and his main comments were along the lines of "there are too many mobile devices out there running too many damned versions of operating systems across too many mobile carriers. It is freaking impossible to test all of it." That was the gist of his complaint. The actual words were ... saltier.

He has a point. There are upwards of 300 different types of Android smartphones on the market across the world. There is an increasing amount of Windows Phones from different carriers on the market, not to mention developers that might want to run applications or mobile websites on BREW, Symbian or Bada devices. Even the iPhone has become more difficult to test applications on with three different levels in the market (3GS, 4, 4S) that most consumers use. There are also different levels of iOS (not everybody knows how or cares to update) running on different carriers. iOS is not fragmented like Android is, but there are still variables to be tested within the ecosystem.

To test different devices, a developer could acquire all those smartphones and tablets, rack them up on a server, and pound away on each device specification. That is impractical, time consuming and expensive. There are several services that developers can turn to for a "device cloud" that developers can use to test applications on many devices at once. IBM uses Perfecta Mobile and Device Anywhere for its device clouds but there are also solutions from companies like Apkudo for device-cloud testing.

"That is also part of our strategy, recognizing that there are many different techniques available for mobile testing and some of those techniques are applicable for mobile apps but for other reasons customers would find the technique not acceptable," Williamson said.

Choices ... Or, None at All?

IBM has been around for a long time - more than 100 years, in fact. It was one of the founding companies in the computing space and to develop consumer hardware and later make the switch to enterprise software. There is chutzpah to being IBM. They take a deep look at issues facing developers and create verticals to solve those problems. And charge for it, sometimes handsomely.

There are benefits to working with IBM. For instance, there are not many other companies out there that offer the end-to-end testing services that IBM can offer. If the company does not have the solution itself, it will either acquire the solution (Green Hat) or partner with a company that does (Device Anywhere). As for Williamson, he understands that end-to-end testing can be done without the services of IBM, but it requires a lot more work than perhaps necessary.

"There are other alternative ways to reach that same well-integrated solution but it requires the customer to do a lot of the integration themselves and in essence be a systems integrator of their own whereas we do that in IBM as a part of our mission," Williamson said.

Bug and cloud keyboard images courtesy of Shutterstock

Trello is developed by Fog Creek Software, a company best known for developer-oriented software like FogBugz, and for co-founder Joel Spolsky.

The Idea Behind Trello: "Five Things"

When Trello was launched last year, Spolsky wrote that it came out of an idea at Fog Creek to help manage developers. "After ten years in management I still never knew what anyone was supposed to be working on. Once in a while I would walk around asking everyone what they were doing, and half the time, my reaction was 'why the hell are you working on THAT?' So one of the teams started working on finding better ways to keep track of who was working on what."

That led to the idea of Five Things. Every employee would have a list of exactly five things they were allowed to work on, two active jobs and three jobs that they'd tackle once the first two were finished. The number of things employees could be assigned was five, and no more. Four was too few, and six was right out.

Turns out, the Five Things idea didn't work so well, but it led to Trello. "We started dogfooding the product when it was only 700 lines of code, and even in that super-simple form, we found it incredibly useful. By the end of the summer, we realized we had a hit on our hands: an incredibly simple, easy-to-understand way for teams to collaborate online."

How Trello Works

Before we dig into the guts of Trello, let me explain why it's head and shoulders above any other online collaboration software I've tried. In a nutshell, Trello really is as easy as using a whiteboard and Post-It notes. It's great for visualizing tasks, and it's dead simple to use. A couple of clicks and keystrokes, and you've got an idea down or put an action item on your list. Moving cards around on the Trello boards is as simple as drag and drop. If the keyboard is more your thing, no self-respecting developer tool would be caught dead without extensive keyboard shortcuts, and Trello is no exception.

Now, let's back up to the basics of Trello. It provides users with boards, lists and cards. Each board starts with three lists, which can hold one or more cards. Boards can be used to organize any project - a software release, an editorial calendar, a user's to-do list or whatever you like.

The boards can be private, or belong to members of an organization, or you can make them public. If you'd like to get a quick idea of what a complex board might look like, the development board is a good one to start with.

Each list holds to-do items, suggestions, features or any other category you'd like to track. Cards track individual suggestions, features, etc. When you create a new board, Trello defaults to three lists: To Do, Doing, and Done. You can, of course, revise the names of the lists and add new ones. I'm not sure about the upper limit of lists per board - the Trello Development board has eight lists, which seems close to the upper limit that you'd want for a project whether or not the software will support more.

The individual cards hold (almost) as much, or as little, information as you'd like. At the most basic, you might give a card a title like "finish Trello article." Then move the card between the default lists to show what you're planning to do, what you're doing and what you have done. But cards will do much more for you, if you want.

Collaboration

Click on a card, and you can assign it to members of an organization, give it a due date, assign labels and leave comments. Cards also allow you to add checklists, so you could include a list of steps that need to be finished before a project is complete - or just use a card for your grocery list, if you like. Trello will also give a progress bar that shows how close a project is to completion based on the items that have been checked off.

As you'd expect, you can also leave comments on cards detailing your activities or leaving information for other members. Trello cards also allow for attachments up to 10MB, so you can leave documents, images, etc.

You might find more complete collaboration tools, but Trello seems to have struck the perfect balance between the amount of information that you can convey with the system and ease of use.

Trello also provides a full activity feed on the side of the interface, which shows what you (and other users in the same board) have done recently. You can also opt to have Trello send email notifications of activity "periodically" (about every hour) or "instantly" (every minute). You can also choose not to have notifications at all.

Finally, Trello also has really good filtering/search options so you can skim boards by keyword, member, label, etc. You can combine these, too - it's easy to search a board by label(s), member(s) and specific search strings.

Trello on Mobile and Add-Ons

The Trello experience works great on a desktop browser, but not so much on a less-than-5-inch screen. But as luck would have it, there's an app for that. At least, if you use iOS. The iOS app lets you do most of what you can do from the Web-based application. You can create cards, lists, etc. You can set due dates, add activity updates and so forth.

The Trello site works pretty well on the iPad in Mobile Safari, though it doesn't support drag and drop. You have to click the menu at the top of each card and select the "Move" option instead - not quite as slick, but still usable.

As far as I know, there's no Android app yet but it is in progress, according to the Trello team.

Work in Progress

It's worth noting that Trello is a work in progress in general. The team seems pretty busy adding features and responding to requests. Some very minor things don't work quite as you might hope. For example, you can archive lists, but you can't delete them outright.

You can print out cards, but as far as I can tell, not entire boards. You can also export cards as JSON, which might be nice for developers.

If there's a feature you think Trello should have, but doesn't, don't fret too much. Trello is in heavy development and Fog Creek has a really nice board for suggesting ideas and voting on existing ideas. I'm a little bit disappointed that Fog Creek isn't planning a paid or supported version of Trello in the near future. I'm pretty happy with Trello and plan to use it a great deal, so I'd like to be a paying customer rather than depend on a free product.

That said, Fog Creek has committed to keeping the current product free forever, though the company says it might add pay-only features in the future. It also has committed not to "make a for-pay feature that forces you to compromise on privacy, security or portability." So far, Fog Creek has had an excellent reputation with developers, so I'm willing to trust the company as it works toward a for-pay model.

If you're curious about how Fog Creek is putting all this awesome together, check out the tech stack behind Trello. It's an interesting collection of tech. And if you haven't tried Trello yet, what are you waiting for?

Most people think of Autodesk as the maker of AutoCAD, the design software of choice for architects, engineers and other design professionals - typically running on high-powered workstations. So why is Autodesk CEO Carl Bass so hung up on the "democratization" of technology - spreading technology to the cloud computing platforms and mobile devices?

At the company's media summit in San Francisco this morning, Bass told a crowd of journalists, analysts and customers gathered in the company's slick design gallery (see pictures below) that the combination of mobile devices, cloud computing and social collaboration is more profound than the shift to PCs.

Bass sees the world changing from a PC-centric model where workers promise to "email you that file when I get back to the office," to an environment where mobile devices and the cloud make where ever you are the computing center of the world.

It's already happening, he claimed, citing a list of impressive usage figures:

• 2 million unique visitors a month to Autodesk 360, the company's cloud offering


• 30 files a minute uploaded to AutoCAD WS, the company's cloud-based AutoCAD editor


• 10 million downloads of SketchBook in 2 years, now averaging 150,000 per week on PC and mobile platforms


• 13 million unique visitors - more than Pinterest - to the company's Instructables community


• 21 million unique visitors a month to Pixlr, its online photo editor

On the low end, naysayers like to denigrate the importance of mobile products, Bass said, calling them "juvenile" "toys." But he pointed out that "consumers by night are often professionals by day."

He also claimed that professionals can do serious work on today's portable devices. "I think we're underestimating these small devices... in the work that we do. They can run serious apps" for engineers and other demanding users, and they are getting more powerful all the time.

Meanwhile, on the high end, the cloud lets anyone take advantage of analyses that used to require dedicating expensive workstations for days at a time. Now, "You can do it in the cloud in an hour," he said.

The cloud, Bass added, "is an infinitely scalable resource," limited only by how much you're willing to pay. For urgent jobs, you can pay more and get it done faster. Other tasks can be done more cheaply over time. And that raises a fundamental question: "What would you do differently if you could compute answers faster?"

Autodesk may be a bit ahead of its time. The vast majority of serious design work is still being done sitting at powerful workstations, just as it has been for a while. But Bass couldn't be more correct about the trends. It's hard to argue that more and more computing tasks are going to stop moving away from the desktop. Big, data intensive jobs will move to the cloud while smaller, more UI-focused tasks are going mobile.

There will always be some things best done sitting at your computer. But the number of those things is clearly shrinking, not growing.

Tomorrow, I'll write about Autodesk's cloud-based PLM (Product Lifecycle Management) offering. But in the meantime, enjoy a couple more shots of some of the cool designs on display in the Autodesk Gallery:

Me? I'm a fan of HTML5 solutions, not so much the Adobe Flash solutions – especially since Adobe is essentially abandoning Linux. But you can't assume that all users are using a browser that can support HTML5 playback. And you can't assume they're on a platform that has Flash support, either.

The jPlayer project handles this by providing fallback support for either one. You can choose to go with Flash by default, HTML5 as a fallback or HTML5 by default, with Flash as the fallback. One way or another, almost all bases should be covered. jPlayer has support for Firefox, Chrome, Opera, Safari, IE6 through IE9, Mobile Safari, the Android 2.3 browser, and BlackBerry's OS7 phone browser and PlayBook browser.

It also has support for plenty of media types: MP3, MP4 (AAC or H.264), Ogg Vorbis/Theora, WebM, WAV, and FLV.

One of the reasons I like jPlayer is its extensive documentation and API reference. The project also supplies plenty of demos to see how jPlayer might be used for video or audio players on a site. (Note, I do believe that "Big Buck Bunny" has now established itself as the "Hello World" of video examples.)

There's an instance on JSFiddle if you want to mess around with jPlayer in real time. The Quick Start Guide should, well, get you started. The downloads page also has a zip file of the demos which should help.

jPlayer is developed by a small Web development firm, Happyworm. It's available under the MIT and GPLv3, so one of those licenses should work for just about any project you're interested in. You can grab it, and send pull requests, on GitHub. The team is prepping for a new release so now would be a good time to get requests in.

Let me say at the outset, I'm not suggesting that everybody should move away from Gmail or Google services. It's merely the right choice for me, and it might be interesting to other folks who depend on Google right now but want to find alternatives.

I've been a Gmail user for years, and moved my domains (that I use for most work and personal correspondence) to Google Apps for Your Domain nearly as soon as it was available. At first, I was pretty happy with the experience, but I've become increasingly dissatisfied with Gmail (and Google) over the past few years.

While I'm concerned about privacy issues with Google, it's not the primary reason I'm jumping ship from Gmail. My qualms have been more along the lines of Gmail being less and less appropriate for the way I want to use email. Google's attempts to make Gmail more friendly may work for many users, but overall I've found Gmail less and less friendly over the past few years.

For example, take Gmail's bulk filtering and trying to classify messages as "important" or not. When this works, it's fine. In fact, I'm surprised by how much I have come to rely on it. But I've also noticed that Gmail is pretty iffy about its classifications. Somehow, it manages to label many of my messages both important and bulk. So those messages (often missives from PR folks, some of which include call details) get shuffled off to the Bulk folder. When they're unclassified as "bulk" messages, Gmail helpfully closes the message and shoves you back into the Bulk folder – with no real indication what it's done with the message.

Moving from Gmail

The hardest part of migrating from Gmail? Deciding where to move to. I've experimented over the years with different providers, and for a while I hosted my own email. If you're hard-core on privacy, self-hosting is the way to go. But managing IMAP and SMTP for myself is just not something I feel like mucking with these days. The cost in time and trouble for setting everything up, including SSL certificates, just isn't worth it.

Finding a provider that offers email hosting isn't hard. Finding a great provider, though, that's a little trickier. Initially, I thought about going with Rackspace's email, but I couldn't find anyone I knew personally who'd used the service. After asking around a bit, I kept hearing good things about FastMail. I spent quite a bit of time looking over the FastMail site and saw that it offered replacements for most of the Google services I was using (related to email), and then some. Reliable SMTP and IMAP, Webmail, spam filtering, and XMPP (Jabber) support.

Above and beyond Gmail, FastMail also offered file storage over WebDAV, and LDAP access to the address book – which meant it'd be easier to sync my contacts between the Web interface and Thunderbird on a couple of different computers.

The actual move from Gmail hasn't been too difficult. I had to change my MX records and add a few CNAME records to direct mail from Google's servers to FastMail. FastMail offers an IMAP migration tool, so if I wanted to move my email archives from Gmail into FastMail, it should be easy.

Google talks a lot about its "Data Liberation" stuff, but they don't provide a good way to get all your mail out of Gmail. Yes, you can grab mail via IMAP and POP to get it away from Google, but if you've built up several gigabytes of email, you can count on a long slog archiving your email. It'd be nice if you could grab all of it as an mbox or something like that.

Hello, Spam! I Haven't Missed You

One of the things that Google gets very right is spam filtering. They don't catch everything but Google does a pretty good job of spam filtering.

So far, FastMail hasn't been quite as effective at stopping spam before it hits my inbox. Which is to say, I've seen seven or eight pieces of spam in my inbox in the last few days, where I usually don't see any.

On the other hand, FastMail allows you to set the level of spam protection you'd like to use. After looking into it, I saw that the default is to "normal" spam protection, but their "aggressive" settings are more to my liking. We'll see if that improves things from here on out.

In addition to spam, I've also had to start dealing with a lot of mail that used to get shuffled off to the Bulk folder. Gmail does get bulk filtering right much of the time, I have to say. The move has also helped me realize how much quasi-spam I've been getting, and lists I've either signed up for and forgotten about or never signed up for. (Hi there, 1-800-Flowers.com!)

So if you decide to liberate yourself from Google to another email provider, you might be in for a minor shock when you realize how much mail you don't have to think about.

Webmail, IMAP and SMTP

The FastMail Webmail interface is a bit clunky. They're working on a more modern interface, but the current Webmail setup is a clunker next to the Gmail interface. It has its charms, but I suspect most users would prefer Gmail's interface over the Webmail for FastMail.

But what it lacks in looks, bells and whistles, it more than makes up for in terms of speed. I've found Gmail to be pretty boggy of late, whether I'm accessing it via Chrome, Firefox, Safari or Mobile Safari on iOS. It doesn't seem to matter if I'm on my iPhone, iPad, ThinkPad, or iMac, it's not at all uncommon for the interface to hang for up to a minute when searching or just waiting for to switch between labels.

IMAP access and sending via SMTP is also far, far faster. Right now, I'm using Thunderbird for most of my mailing, and the difference between sending mail via Gmail's SMTP server and FastMail is like night and day. You'd think Gmail would be the winner here, given that Google has more servers than many small countries. Alas, no. Gmail wheezes through sending mail with the speed of an arthritic snail.

Mail That's Not Dumbed Down

The big benefit, depending on your point of view, of moving away from Gmail is getting a more "traditional" email experience. To put it another way, mail that's not dumbed down.

For example, being able to sort mail by sender or subject. It's nice to be able to actually sort my email when using Webmail, rather than depending on search. FastMail's interface for creating mail filtering rules is more robust than Google's too.

The Webmail interface also has some features that you'd find in older mailers that Gmail doesn't support, like redirecting (or bouncing) an email to another account. This is slightly different than forwarding a message, and allows me to route an email from my personal domains to my @readwriteweb.com account if I want to reply from there instead.

FastMail also has tools for purging mail and removing duplicates, whereas Gmail just sort of leaves users on their own to manage space.

It's worth mentioning that I've also switched from a "free" service to one that's running about $45 a year. Free, of course, is relative. I wasn't paying Google anything, but it also meant I was not really their customer. In the end, I'm much happier trading a few bucks a year in exchange for not using an ad-supported service. If I have problems with FastMail, then I have paid for the right to expect support and help in a timely fashion.

FastMail has a bunch of different plans, I'm on a "family" plan with a single user right now. That's not much cheaper than it would be to pay for a single user on Google's paid Apps for Your Domain plans. But if I want to add another account I can, potentially, do so much more cheaply than with Gmail.

The downside to choosing FastMail over Gmail is that the service doesn't provide anything for calendaring. My next task is going to be either finding a service that provides CalDAV for a reasonable fee, or setting up my own CalDAV server. I'm really hoping to avoid setting up a server for just one user, but so far I haven't found a lot of calendar providers – especially not services that are geared for individuals. I'd consider 37Signals, for instance, but their free plan doesn't include a calendar and the paid plans start at a hefty $24 a month. Might be worth it for small team, but not for a single user. Any suggestions are welcome.

If you ever have your friends and/or relatives help you relocate, be sure to keep the security cameras rolling. There will be packing and unpacking decisions that are made for you, some of which you will never be aware of unless there's evidence in front of you. Sometime during the late summer of 1994, I lost a small briefcase.

It had no valuables in it; in fact, it had but one purpose. I carried magazines in it, typically computer mags. I had been married almost a year. My wife, not yet my business partner, was an editor and contributing author for what was then Prentice-Hall on the north side of town (it's where I met her; she was my editor on a book). There were picnic tables set up outside the office, and as I waited to pick her up for lunch, I'd bring this briefcase of magazines with me. It's a sign of how importantly I treated computer mags: I intentionally scheduled my time to allow for one hour for lunch with Jen, and one hour reading my mags.

They were my connection to a brighter world - the continual, monthly affirmation that everything I believed to be important as a kid (not that long before), actually was. I often used a computer magazine as a prop for my presentations for editors or business executives. I described it as "a more important convergence than the TV with the PC:" the fusion of journalists, programmers, and teachers. We're all becoming the same group, I'd tell audiences, and the economy your children will inherit will be powered by their ability to discover, to teach, and to code.

Two months ago, at the bottom of a cardboard box of spare bed linens we had intentions to donate at some more convenient year, someone had stuffed my magazine briefcase. It had been resting there since Jen and I moved into our first house, waiting there ever since we had our child. Inside it was, for me, a love letter from the twentieth century.

Software Development

One of the feature articles in the August 1994 edition of Software Development was about the IEEE's efforts to develop a standard for creating new documentation for old software. There was a new programming paradigm that businesses were just now waking up to: client/server. It was a multi-tier model of computing where logic was distributed and communicated through these emerging digital networks. But businesses and especially governments were realizing they could not justify the near-term costs of replacing existing software - much of it having been conceived in the FORTRAN era - versus continuing to expense the existing costs over the coming decade, the turn of the century.

The job of CIO had not yet caught on in most corporations; budget specialists and procurement agents evaluated software replacement costs in terms of lines of code. As ridiculous as this sounds in hindsight, if newer software was shorter than the programs it replaced, budget officers would compute "lines of code destroyed," and treat it as a cost.

So the IEEE thought perhaps an interim solution would be to produce a standard for reverse-engineering existing code, with the objective of producing documentation. This way, it might be easier for companies, from now until the end of time, to continue using the old software since they'll never replace it. "Maintenance as usual is the origin of CEO cynicism toward remedies to cut the costs of maintaining long-lived systems," wrote physicist Moisey Lerner, at a time when tech news stories were written by physicists. "The new IEEE Standard 1219-1993 for software maintenance is destined to cure this cynicism."

Windows Magazine

Windows Magazine was edited by a man I would later the privilege of calling a friend and colleague, Fred Langa. His July 1994 scoop was the first pictures of "Windows 4.0" - the system that Microsoft had been calling by its code name "Chicago," and would soon be re-dubbed Windows 95.

The existing edition, Windows 3.1, was perceived as a most welcome correction for the slow, overbearing monstrosity that was Windows 3.0. It was more reliable, easier to maintain, and more standardized in its tools and features. It crashed less often, and gave the user fewer reasons to exit back to the old command line, MS-DOS, to do "serious work."

But the next version was due to introduce something not copied from the Macintosh: a curious new creation called the Taskbar, which had an omnipresent button marked "Start." It implied that Windows was taking over from DOS as the operating system.

It was a turning point in the industry, and Langa knew it. "Chicago sports a completely redesigned user interface with new ways to navigate and use the system," he wrote. "If Chicago delivers on this promise, mainstream computer users will have more power at their fingertips than ever before. The industry will respond with a flood of Chicago-specific software and hardware... and the entire computing industry will take on a new Chicago-centric aspect. Almost every piece of hardware you use and almost every piece of software you run will be affected. If Chicago fails, the downside could be just as significant... We need a modern 32-bit platform, and we need it soon. If Microsoft can't provide it, someone else (maybe IBM? maybe Apple?) will, and we'll be faced with an entirely different set of hardware and software choices. Win or lose, Chicago will shake the industry and affect us all."

One of Windows Mag's reviews for July 1994 was for a personal information manager (PIM) package called Polaris Advantage. It was a $149 scheduling, note-taking, and contact management program, in the era before Outlook and in a world that still viewed the central destination of all data storage as the PC. As quickly as they could, developers were conceiving new ways to import data from these weird new portable data collectors, and store it all in the hard drive where it belonged.

"Importing data is fast and flexible," wrote James E. Powell. "You can save an import setup with its field mappings and reuse it. Advantage will accept dBASE, Paradox, Excel, Btrieve, and text files, as well as files from the Sharp Wizard or Casio BOSS."

A major advertiser for Windows Magazine in 1994 was Apple. Always the company to bring out these cool devices for PCs, the QuickTake 100 Digital Camera was being touted as "the fastest way to give everything from proposals to catalogs more impact... And at $749, accountants love it, too.

Byte

Byte Magazine is one of the reasons I'm here today, doing what I do. Every month, Byte set its sights on the bigger picture, a significant trend that might be far ahead or way far ahead. And in July 1994, Jon Udell (to this very day, among the most insightful people ever to sign his name to an article) was setting his sights on the inevitable convergence between the computer and the telephone.

The FCC had set a deadline for April 1995 for telephone carriers to implement Caller ID, which meant that digital content would be carried alongside the phone call by law. Engineers had already foreseen the benefits of gathering this digital information and leveraging it for computer-based telephony. But this was the '90s, and DEC, Rolm, and the other behemoths who tried and failed the first time, were being swept aside by sleek, savvy, streamlined newcomers to this industry like Novell.

"Novell says you'll pay $75 to $200 per seat for NetWare Telephony Services," Udell wrote, "depending on the number of users (and not including the cost of the link). If you're not running current PBX hardware and software, though, you'll need to upgrade, and that can be painful." Udell then cited one source who was given an upgrade quote of $30,000. But since the upgrade came in the form of software, he suggested, rather than hardware (usually a capital expenditure along with furniture), accountants could find a way to amortize the costs long-term.

Elsewhere, Bruce Dawson reviewed a ruggedized notebook PC from IBM, that even featured the little track pointer you still see on Lenovos today. A magnesium alloy case protected a PowerPC 601-based system running AIX (not Windows), with plenty of memory (16 MB) and more than enough hard drive space (340 MB), all for a low price ($11,995) that rendered it potentially "indispensable," he wrote.

Lost Bits

There's a warning frequently given to children about protecting themselves when they do anything online. "Everything you do online is searchable and is permanent."

Don't tell this to your kids, but statistically, there is no greater falsehood. Of the tens of thousands of online contributions I have made in 28 years' time, less than 10% of it is obtainable through the Web at this moment. If it weren't for my obsession with archiving, most of my work would not exist. In a half-century, my daughter may uncover a hard drive buried in a box of linens whose donation worthiness she happens to be reconsidering. Will she even remember what it is?

The computing publications of 1994 were our connections to a broader, brighter world of insight, audacity, wisdom, and adventure. Each one had a collective character, an amalgam of all its editors and contributing writers. Its voice had personality, intelligence, spirit. What's more, each voice was unique, in the same way people are unique. Each magazine's structure was its own psychology, its arrangement of cover stories and features and reviews as great a statement of its character as any one of its articles. You didn't read Byte by accident, or because you stumbled upon an article or someone ripped it out, folded it into a paper airplane, and tossed it at you. You read Byte because you were a Byte reader. We didn't call the act of reading a computer magazine "consuming." And we didn't call the messages it conveyed "content."

There is a message from that time that is missing in the message from ours. Sure, there were writers who were pushing dBASE and trumpeting the marvel of megabytes. But they were the first to believe and uphold the ideal that they could speak to you, the reader, as an equal and as an intelligent thinker. They didn't write above you. They wrote to you.

They wrote to me. Some 12,000-plus responses later, I'm still just catching up.

This was the topic of Jesper Andersen's talk at Strata, "Building a Data Narrative: Discovering Haight Street." The idea, to "see how far we can go" in understanding San Francisco's famous Haight Street through as much data as possible.

Tell Me a Story

Before diving into the data, Andersen made the case for providing a narrative for Haight Street. Pure numbers work for some types of data, but they're not good at expressing social networks or explaining what a street is "like." For example, Andersen talked about Klout and the way it tries to express a person's social influence with a number. Klout, says Andersen, "is reductive and destroys value" in trying to sum up a person's social network with a single number.

Andersen didn't spare himself, either. He talked about some of his former work with Visible Path creating a "horrible interface I'm now ashamed of." The problem, says Andersen, was trying to reduce users to scores. "Don't give users scores, give them stories."

So how do you create a story from the data available for Haight Street? Andersen noted that you can find a lot of data about Haight Street, but when you search WolframAlpha about Haight Street you don't really get to know the street.

What you get is cold facts. A map that shows Haight Street as perfectly straight, but doesn't give the elevation. It certainly doesn't give the atmosphere of the street. To dive in to find out what the street is like, Andersen delved a bit deeper.

For example, he looked into Foursquare checkins, and mapped them along the length of Haight Street and compared them to a map of the elevation of Haight Street. He took Google Street View and tried to extract the colors of Haight Street to see what the primary colors of Haight Street are.

Andersen tried to answer questions like "where is it safe?" using DataSF.org and looking at the crime statistics. Again, he mapped these according to the length of the street and compared upper and lower Haight Street. (Lower Haight Street is less safe, statistically.)

What Are the People Like?

The hardest of questions? "What are people like on Haight Street?" To try to puzzle this out, he looked at Twitter data by location and language. Surprisingly, you'll find a lot of Swedish folks on the upper half of Haight Street. Not surprisingly for San Francisco, many people on Haight speak Spanish or Japanese.

Andersen also tried sentiment analysis of Tweets and found that, by distribution, people were more negative on the lower half of Haight. He was also able to get some idea of what people found interesting on Haight by mapping pictures from Instagram to the street as well.

All of this information is out there, and can be used to start telling a story about a location past hard facts. It's not a perfect picture, but it's a start towards storytelling with data instead of just a jumble of numbers and facts that don't paint much of a picture at all. Andersen's attempt suggests that there are ways to use existing data for storytelling. What's more, there could be a big opportunity for app makers to look for new ways to collect and display data to paint better pictures of real life.

Haight Street photo courtesy of Wikipedia under the (CC BY-SA 3.0).