Towards a National Cybersecurity Policy

The US presidential election has made cybersecurity a major focal point for government policy. After numerous serious breaches and denial of service attacks at retailers, banks, media companies, defense contractors, government agencies, telecommunication providers, and Internet service companies, it's about time that we had a serious national response to the threats posed by state-supported and private actors. However, I'm concerned that we could end up in a bad position if we rush to react to perceived threats without careful consideration of the systems and policies we implement and their potential side-effects.

Cybersecurity Basics

Our current approach to cybercrime prevention requires each information system operator to secure their own environment by implementing best practices for threat prevention, detection, and response. These practices include:

• using network encryption (SSL/TLS, VPN, etc.) to ensure all data flowing between the system and its users cannot be intercepted
• using multi-factor authentication to ensure that system access is restricted to authorized users
• implementing strict controls around physical system access and encrypting all data that is stored in the system
• implementing strict procedures for patching known security defects in all device firmware, system software, and application software
• ensuring all systems are properly configured to prevent unauthorized access or data leakage (operating systems, web servers, e-mail systems, database servers, etc.)
• ensuring complete awareness and accountability of all devices and applications in the operator's network, to prevent a rogue device or application from being inserted into the network and leaking proprietary information
• implementing appropriate network security measures (firewalls, IDS, and IPS) to prevent unauthorized data transmissions to or from components in the network
• implementing security information event management (SIEM) to observe all application, system, and network events and detect suspicious events that could indicate illicit activity
• implementing denial-of-service prevention solutions to protect against brute force attacks like the one that took out Dyn on October 21, 2016 and with it many popular Internet services
• regularly inspecting all code developed internally for security weaknesses and immediately patching any defects discovered
• regularly performing third-party penetration tests to verify that cybersecurity defenses are operating as intended
• implementing e-mail and messaging filters to protect employees from exploitation (e.g., spear phishing, social engineering)
• performing routine malware scans on all servers and personal devices
• implementing recurring cybersecurity training for all employees, including specialized classes for software developers and system operators
• implementing an internal cybersecurity management and incident response team with skilled professionals and 24x7 coverage

As you can see, there's a lot of stuff that has to be done, and most of them require ongoing effort and expense. The awful truth is that most companies are not doing these things, and many others are doing them in an incomplete or haphazard manner. And the bad guys know it.

A Big Problem for the Government

Cybersecurity is just as big an issue for our government as it is for the private sector, yet in many ways the government is even less capable. The breach at the Office of Personnel Management, which resulted in the loss of over 20 million government employee records with extremely detailed personal background information, could have been prevented if the agency required two-factor authentication, but they didn't. It went on for over a year, but it wasn't detected because they didn't have an effective SIEM platform in place. It's also been widely reported that Chinese hackers stole large amounts of data from the F-22, F-35, and C-17 programs at the Department of Defense.

I believe – or should I say hope – that some government systems have better security in place, but the complexity of systems and networks, and the constant evolution they experience, means that the probability is high that access to even the most sensitive information can be gained through unexpected and potentially unguarded pathways.

Even our political system is under threat. The 2016 election cycle started and ended with the drama over Hillary Clinton's private e-mail server and its potential implications for national security. As if to underscore this issue, thousands of internal e-mail messages were stolen from computers at the DNC as well as the Google e-mail account of Clinton's campaign manager John Podesta. The DNC breach was carried out by malware installed on their computers, and the Podesta hack was a straightforward spear phishing attack that convinced him to give his Google mail credentials to the hackers. If the DNC had mail filtering and malware detection systems and if Podesta had enabled two-factor authentication, both attacks would have been thwarted.

A National Cybersecurity Policy?

The incoming administration has proposed a national cybersecurity task force to determine appropriate policies and actions for defending our nation's information systems. What might this entail?

First, there might be a set of regulations which require companies and government agencies operating vital information systems to implement a specific minimum set of cybersecurity capabilities from the list above. Companies that fail to meet these requirements will be cited and must comply within a given period of time or face stiff financial or other penalties. This will require a national cybersecurity enforcement agency that will perform inspections and penetration tests, and deliver warnings and citations to organizations that fail to comply with the law. There is legal precedent for this; consider that Ford can't sell a car in the US unless it has airbags and seat belts, so why should a bank be able to launch an online banking system without the capabilities required to protect its customers' identities and accounts?

Second, there might be a national cybersecurity monitoring function observing all network traffic passing over major internetwork links, especially those traversing our national border. In this way, unusual or suspicious traffic patterns can be detected; e.g., OPM data being sent to China, a major bank's computers being accessed from a TOR node, or a large data dump from a US telecom provider's network to a server in Romania. This would then let the potential targets be informed, hopefully before too much damage is done.

Third, there might be a traffic management function to block or interrupt network traffic if it is deemed suspicious or harmful. This would amount to a national firewall, where all US network operators are required to deploy secure border gateways with traffic inspection and control software and allow control of those devices by some national cybersecurity agency. This mechanism might be able to prevent large volume data theft and block large-scale DDoS attacks.

Fourth, there might be a set of regulations which require manufacturers of networked devices to meet a minimum set of cybersecurity requirements. Consider that a lot of the hacking going on today is enabled by things like home routers and webcams with known vulnerabilities that have been compromised by malicious actors and are being used to penetrate internal company networks or launch large-scale denial of service attacks. Companies might have to demonstrate compliance before their devices are allowed to be sold in the USA, so this would give them a massive incentive to get it right.

Finally, I believe the task force will propose a single agency to implement and enforce these policies. This will close the gaps that exist today across Homeland Security, Commerce, Justice, Defense and other federal agencies that each cover a part of the national cybersecurity landscape.

What Could Go Wrong?

Any policy decision will have negative impacts as well as positive ones, but this set of policies could have huge implications for our government, our citizens, and our businesses.

First, a new set of complex regulations and compliance enforcement policies will create a large cost burden on businesses and taxpayers; the former to comply with regulations, the latter to fund the personnel and systems required to enforce the policies. You might argue that businesses should be in compliance in any event, but having to prove compliance to the government will create expenses beyond the internal implementation cost.

Second, a global traffic observation system will be costly and may raise significant privacy concerns. Will the government be watching every movie you download from a foreign server, or reading every e-mail you send to international friends and co-workers? Beyond this, it will create costs for network operators and will require people and systems that will perform the monitoring and alerting functions.

Third, a global traffic management system will potentially threaten valid traffic between networks and could allow the government to intentionally cut off Internet traffic to any or all other countries, so this is a serious threat to our freedom. If your personal or corporate communication was blocked, and this communication was valid, what would you have to do to get it un-blocked? What rights would you have to freely send and receive data, and what obligations would the government have to protect those rights and not infringe upon them?

Fourth, requiring devices to be inspected by the government prior to being sold will create delays in time to market and additional expenses that smaller companies might not be able to bear. This will advantage large companies with established revenue streams and more human resources, and thus will probably retard innovation.

Finally, a new large and powerful bureaucratic organization will also create new problems. It will have to be funded and staffed, initially. After a while, as all bureaucracies do, it will attempt to solidify and expand its position in society by increasing its authority and budgetary requirements. It will create friction with existing bureaucracies which will result in wasted energy within the government, and potential fallout to the citizens and businesses as they try to decipher and navigate the new requirements. Consider the impacts encountered when businesses and individuals were trying to understand the impact of complying with the Affordable Care Act when it was introduced.

Summary

At the end of the day, I believe we need to have some kind of national cybersecurity policy defined and implemented. Whatever we do, it will be complex, expensive, and impactful. Additionally, it will create risks to our freedoms that will have to be carefully checked and managed. However, continuing along our current path will clearly incur greater threats in the future – we simply have too much at stake to allow continued exposure and exploitation of our governmental and commercial information systems.

The SpaceX Plan to Colonize Mars

At this week's International Astronautical Congress, SpaceX CEO Elon Musk outlined a bold and well thought out approach for the colonization of Mars based on three key principles:

1. Reusing all system components to minimize costs
2. Fueling outbound transports in orbit to maximize payload capacity
3. Enabling fuel production on Mars to eliminate the need to carry return fuel

Reusable booster rockets will carry interplanetary transport vehicles and fuel tankers into Earth orbit and return to land back at the launch site. These massive boosters, powered by 42 of SpaceX's new Raptor engines, have four times the payload capacity of the Saturn V moon rocket. Reusing these boosters not only cuts cost, but drastically reduces the time required to deliver additional payloads.

The transport vehicles will carry people and cargo back and forth between Earth and Mars, and the tankers will be used to fuel the transports in Earth orbit. This approach will allow the transports to be launched without fuel on board, maximizing the amount of people and supplies each ship can carry to Mars.

The transport ship and tanker will be based on the same fuselage design to keep costs to a minimum. Both of these ships can return to Earth using aerobraking and propulsive landing capability so they can be reused on subsequent missions, another key factor in managing costs. The transport ship will also be capable of landing on Mars and launching from there back to Earth. Since Mars' gravity is only 38% of Earth's, the transports will not require a booster rocket to return home.

In order to get back to Earth, fuel must be produced on Mars. To enable this, the Raptor engine runs on methane and oxygen, both of which can be produced on Mars using the carbon dioxide that makes up most of the atmosphere and the water ice that can be found on and under the ground. Power generated by solar panels or nuclear reactors can be used to melt the ice and split the hydrogen from the oxygen via electrolysis. Some of the oxygen can be used for rocket oxidizer, and the rest can be used for life support. The hydrogen can then be combined with the carbon dioxide in a reaction to create methane rocket fuel and water for life support.

Musk's plan is based on the (approximately) 2-year cycles of Earth-Mars orbital alignment when getting to Mars will only take 90 to 120 days. He envisions a fleet of transports that will gather in Earth orbit and all leave for Mars together at these times. This will allow large amounts or cargo and large numbers of colonists to arrive at once, increasing the odds of survival for the colonists since the loss of any one ship will have less overall impact on the mission.

All of this sounds amazing, yet what makes it plausible is that SpaceX is already a successful, fully-integrated space systems provider, and they have already developed many of the technologies required for this plan to succeed. However, there are still some major obstacles to clear:

• SpaceX will need a lot of additional cash to complete this plan. Musk was clear that he will need funding help from both the private and public sector to get the system built.
• The booster and transport/tanker vehicles are still being designed, and there are major technical challenges to be worked out along the way. For example, the design calls for composite fuel tanks to keep weight down, but it may be difficult to make this material work at this scale and with the temperatures and pressures required as Lockheed learned during their failed attempt to build the SSTO Venture Star space transport back in 2001.
• How the colonists will be housed, fed, and protected from radiation on the long journey to Mars is still being worked out. Musk believes the radiation risk is small and can be mitigated by keeping the ship pointed away from the sun during the transit, using the engine, fuel tanks, water stores, and cargo to shield the passengers.
• How the proper landing sites will be selected and prepared prior to the first colonists' arrival is still to be determined. Building enough habitat for hundreds or thousands of colonists is a large task, and making sure that fuel production and food production facilities are already in place is critical to success. If you land too far from the ice, or if one of the transports crashes on the fuel factory, the mission is doomed.
• Exactly how the colony will be sustained in terms of food, water, air, and housing is still not clear. We'll need to figure out how to build and maintain airtight and radiation-proof structures using as much local material as possible. We'll also need to learn how to grow food on Mars.
• Having a lot of people on Mars will require some form of local government. Just like on Earth, there will be disagreements, crimes, and other problems between people that will have to be dealt with. You will need police, courts, and jails.
• Since there will be men and women, there will be babies. Education facilities will be required at some point.
• What will be the basis of the colonists' economy? Will they be employees of SpaceX? Will they get (or even need) a salary? Will there be taxes? How will goods and services be equitably distributed?

Musk indicated a belief that building a self-sustaining colony on Mars would be a 40 to 100 year process, and he plans to start sending the first ships by 2018. These would initially be the smaller unmanned Dragon capsules, followed by the large colonial transports in 2023. Assuming he can get the funding, given the estimated costs and life expectancy of the various ships, he believes he can deliver people and cargo to Mars for less than $140,000 per ton. This is an amazing achievement when compared to the cost of current approaches for interplanetary travel.

Finally, we should consider the human element of this endeavor. Life on Mars will be difficult, the work will be hard, and the risk of death will be omnipresent. In many ways, the Martian colonists will be like those who left Europe for the New World in the 17th century. Many will die in shipwrecks, from work accidents, from illnesses, and other things that we don't often concern ourselves with in the developed world. On top of all that they will have to live in an enclosed habitat all the time; going outside will require a space suit. Given this, will people even want to go to Mars? If they go will they want to stay? What will be their motivation? Musk's goal of making humanity a multi-planet species is admirable, but previous colonization events have been driven by people's internal desires to find a better life for themselves and their family. Will Mars hold this kind of promise for enough people to create a viable community, or will it end up as just another lonely outpost for a few dedicated explorers and scientists? Only time will tell, and I look forward to seeing how this develops over the next few years.

Sour Apples: iPhone 7 Disappoints

The new iPhone 7 is a disappointment, not for what it is but for what it could have been.

Let me explain.

On Wednesday, Apple announced their new hardware, specifically updated watches and phones. There was a lot of anticipation in advance of the event, with intense speculation that Apple would remove the analog headphone jack from the phone.

The headphone jack is certainly gone from the iPhone 7. However, what's also missing is any real innovation. There are many small refinements including better cameras, improved CPU performance, improved wireless network performance, and improved battery life, but these are things most consumers take for granted in annual tech refresh cycles. Other changes like new case colors, stereo speakers, and the new home button design seem more focused on style than substance.

The headphone jack removal is very polarizing for Apple consumers, because it interferes with common usage patterns that have persisted for almost a decade. Many people listen and recharge at the same time, and now you'll need an extra dongle for that. Many people plug their phones into their cars and other audio gear using the headphone jack, and now you'll need an extra dongle for that, too. And did we mention that this dongle doesn't offer another Lightning port for charging? So if you don't have Bluetooth audio in your car, forget about playing your tunes and arriving with a fully-charged phone.

Why didn't Apple add wireless charging? This is a feature found on many high-end Android phones and the Apple Watch. It would have nullified the complaint that with the new iPhone people can't simultaneously charge their phone and listen to music or make phone calls. It would have eliminated the need for a 'charge and listen' dongle that people don't want to carry around with them since they already have to carry a charger and/or they have one on their desk or nightstand. The option to use Bluetooth headphones is there, but these also have limitations mostly in the form of battery life; I have been on many conference calls that exceed the 4-hour capability of Apple's new AirPods, so would I have to buy two sets of them just in case they conked out mid-call?

Why didn't Apple introduce an OLED screen? This offers better contrast and lower power consumption than LCD, and has also been a feature on higher-end Android competitors for years. Because it doesn't required a backlight panel, it would have let the iPhone get even thinner or the extra space could have been used to increase battery capacity.

Finally, why didn't Apple ditch the Lightning jack altogether and introduce WiGig capability? This super fast 60GHz radio would have let the iPhone support HD and UHD video to desktop monitors and big screen TVs. It would have also allowed the device be totally sealed against dirt and water with no external connectors. More than that, it would have allowed the iPhone to become a single device that could use multiple screens and peripherals to serve as a phone, tablet, desktop, and entertainment device. Imagine pairing your iPhone with a wireless game controller and big screen TV to play video games in high-definition, or pairing it a desktop monitor, keyboard, and trackpad to use office applications and browse the web.

These are the kinds of changes Apple's customers are looking for, but at this point we can only hope for next year. Unfortunately for Apple, the Android platform is more competitive than ever. The recently released and highly-polished Nougat OS and new high-end devices from Samsung and other manufacturers are appealing even to Apple die hards. It seems like Apple's visionary image is tarnishing, and they are transforming into just another mega company trying to protect margins on existing products instead of taking bold risks on new capabilities that drive the industry forward.

p.s.: I didn't even mention that there was not a single announcement related to their personal computer business, even though the Mac lineup has not seen a major change in several years. I'm writing this on a three-year-old Macbook Air, and Apple doesn't have a single product that is compelling me to upgrade. I'll probably write about this in a future article.

Three Wishes for the iPhone 7

There's no doubt that the iPhone is one of the most popular consumer devices ever made, but to some extent it may be a victim of its own success. We now have all the screen sizes we need, from 4 inches to 5.5 inches, and all of them have pixels so small that adding more won't make the image any sharper to the human eye. The processors are fast enough for any application you might want to run, and the cellular, Wi-Fi, and Bluetooth radios deliver enough connectivity options and bandwidth. All iPhone models have fingerprint recognition and Apple Pay capability, and battery capacity is good enough for most people to get through their entire day without concern.

So what new features can Apple put into the next iteration of the iPhone to keep the platform fresh and compel current users to upgrade? Certainly another bump in processor speed or memory capacity isn't going to be very interesting, although it is probably inevitable. If I could ask for three enhancements, these are what they would be.

Wish #1: WiGig

Apple should add a new radio capability to the iPhone 7 called WiGig. Also known as 802.11ad, this technology uses the 60 GHz frequency band to transmit data at up to 7Gb/s. It won't replace your LTE or Wi-Fi connection, because these high frequency signals can't pass through walls, but it will provide several valuable capabilities.

The most immediately useful of these is the ability to watch movies streamed right from your phone to your TV in full fidelity with no messy cables or adapters. Imagine, taking your iPhone into the family room, opening up the Netflix app, and playing a 4K movie to your big screen TV without juggling remote controls. Then imagine going over to a friend's house and watching that same movie on his TV. You can try this today with AirPlay over Wi-Fi, but the video may be choppy and 4K resolution is not supported so it's 1080P at best. This might be the set top box killer we've all been waiting for.

It will also allow video games to display on the TV while using the iPhone as a controller. As an option, a Bluetooth game controller can be coupled to the iPhone to allow a more conventional game console experience where the iPhone replaces the console completely.

WiGig can allow the iPhone to drive a tablet display when you need more screen real estate; for example, while reading a magazine or book. The user can interact with the tablet while their iPhone is sitting on their desk or nestled in their pocket. Apple could create a 'dumbed down' version of the iPad for this, or just add this capability into new iPads.

Finally, and perhaps most importantly, it will allow the iPhone to be used with a high-resolution desktop monitor. When coupled with a Bluetooth mouse and keyboard, this could allow your iPhone to be your only computer. Apps could be extended to support a mouse, just as they have been extended to support a pencil on the iPad Pro. You would never have to sync your iPhone with your PC because all of the data would stay on the iPhone. Imagine traveling without a laptop computer, just using available WiGig terminals in hotels, offices, and other workspaces when you need a more traditional PC interface.

Wish #2: Visual Unlock

While the fingerprint reader has been a useful addition to the iPhone, I think we all have experienced the frustration of it not working from time to time, as well as the awkwardness of using it when your hands are full, dirty, covered in gloves, or just not conveniently positioned.

Apple should use the front-facing camera to enable the iPhone to be unlocked visually, by examining your facial features and getting enough of a sample with small movements and eye blinks to know that it's not being fed a static image.

Wish #3: Remove the Connectors

It's time to eliminate the anachronism of electrical sockets. The iPhone's Lightning connector and the 3.5mm audio jack should be removed, because there's just no need for them anymore.

Getting rid of these metal sockets will allow more freedom in physical design and, with minor changes to cover the volume, mute, sleep, and home buttons, allow the phone to survive serious encounters with water and dirt. Imagine not worrying about taking your iPhone to the beach or pool. Imagine being able to clean your phone by sticking it under running water.

The audio jack is made redundant by Bluetooth. I'm sure this will make some audiophiles choke, but for the vast majority of people Bluetooth is a better option because it is free of cables and provides comparable audio quality at the range of a normal headphone cable.

The Lightning connector can be eliminated by adding inductive charging to the iPhone; just lay the phone on top of a charger to top up the battery. Besides charging, the only other reason for Lightning is fast data transfer, but WiGig has that covered.

I'm sure this suggestion will elicit gasps from those who believe that this will upset users with legacy Lightning devices, but it's easy to imagine a WiGig to Lightning/USB adapter that can keep them out of the scrap heap.

Wrapping Up

It's time for some serious innovation in the iPhone family, and at this point that means Apple needs some extreme focus on key features that will really make these devices better and more useful for the majority of their customer base.

Distracted Driving: Let's Slay the Monster We Created

Almost every day we see it: Someone in one of the cars around you moving down the road at high speed with their attention focused on their smartphone instead of the traffic around them. It's a global phenomenon, and there's no sign of it decreasing in frequency despite all the laws against it. According to distraction.gov, the official US government website for distracted driving, 3,179 people were killed and 431,000 were injured in motor vehicle crashes involving distracted drivers in the USA during 2014. The World Health Organization predicts that by 2030, road traffic injuries will be the fifth leading cause of death globally, surpassing AIDS, diabetes, violence, and all forms of cancer. Yet with all that carnage, the site goes on to say: "The best way to end distracted driving is to educate all Americans about the danger it poses."

As a technologist, this is appalling to me. This problem exists because of the smartphone and mobile network technology we have created, and has grown because of the improvements we have made to that technology specifically so that all of us can be connected at all times, everywhere. Yet our best solution is to try to convince people that they shouldn't do the very thing that the technology was designed to enable!

I can't accept that, and I think we need to devise and implement technological solutions for this technological problem. Certainly people can and will continue to eat, drink, groom themselves, and engage in other distracting activities while driving, but I think we all agree that very few of these have the same distractive capacity as messages and other notifications that are actively pushed to your device at random times during the day causing it to flash, shake, and make all kinds of noises to get your attention.

The solution seems obvious: Simply stop the driver's phone from creating or allowing distractions while the vehicle is moving. Unfortunately, a logical examination of that approach leads to another set of problems that are not easily addressed:

• How do you know that the user is driving and not just a passenger in a car, bus, or train?
• What applications are allowed? Should the driver be able to have a voice conversation over Bluetooth? Should they be able to have a voice interaction with Siri, Alexa, or Cortana? What about Google Maps? Pandora radio? How do you allow some interactions and prevent others?
• How do you disable the driver's phone yet allow the passengers' phones to continue normal operation? People have multiple cars, multiple devices, and people also rent and share cars, so this is not straightforward.
• How do you deal with people who actively try to bypass or override the control mechanisms?

This used to be a problem related solely to text messaging, and perhaps could have been addressed through controlling SMS delivery in the mobile operator networks, but now it's about mobile applications over the Internet. Since you cannot disable Internet service without making the device completely useless, the solution must lie with the device manufacturers and the automobile manufacturers. I don't know what the answer is, but I believe it will involve a few specific technological components. 

First, every mobile device must be able to transition into a Driver Mode that limits its capabilities. For example, screen input may be disabled and only voice interaction permitted. Applications will have to be qualified to operate in this mode, and if not they will be disabled by the device operating system.

Second, every automobile must be able to provide information to the device that would allow it to engage and disengage Driver Mode at appropriate times, such as when the car is put into or taken out of a drive gear, or when the car is in motion or stopped.

The automobile could also assist or take over engagement with the driver in more intelligent ways, perhaps via technologies such as Apple's CarPlay or Google's Android Auto. It's certainly better to present visual information on the car's dashboard or infotainment display than require the driver to look at their phone's screen, and it's also better to present audio information over the car's speakers than those on the phone. In today's world where every smartphone can use Bluetooth and screen sharing, this should not be a big leap.

Finally, there will have to be some solution for dealing with multiple devices in the car. One approach may be to have each device's location precisely analyzed to determine which one is nearest the driver's position. Another may be for the car to detect all devices and force one to be placed into Driver Mode with the notion that passengers would object to having their phone disabled by the driver.

There are a few technical solutions available now, such as CellControl, but these are mainly for teen drivers and require parents to install and configure hardware in the car and applications on their child's phone. Most adults would never voluntarily install this kind of control, and could choose to opt out of it whenever they wanted. 

I think this problem is serious enough that phone and car manufacturers should get together and agree to a solution that works, and ensure that the technology is standardized, documented, and freely licensed. A subset of drivers will try to circumvent the solution, but the majority of drivers will probably appreciate the improved interaction and integration between their phones and cars and never look back.

If you have any thoughts on this important problem, please share them!

2010 MacBook Air: Lose weight, and nothing else

Now that I've had a few weeks (with one slight interruption) to get to work with the MacBook Air, I thought I'd share my findings.

I bought the top-end 13" Air with the intent of replacing my early-2008 15" MacBook Pro as my sole personal computer at home and work. Here are the specs:

• Pro: 2.6Ghz Core2 Duo, 4GB RAM, GeForce 8600M GPU, 7200RPM HDD
• Air: 2.13GHz Core2 Duo, 4GB RAM, GeForce 320M GPU, 256GB SSD
  

One of my biggest concerns was performance — I wondered if the Air could live up to the level of the Pro. Here's what I found out:

The screen is physically smaller, but the resolution is the same (1440x900) so the actual real estate is unchanged. The GPU is comparable, and both 2D and 3D graphics are snappy. The mini display port interface on the Air is superior to the DVI port on the Pro since it can support higher resolution displays and can carry both audio and video.

The CPU is slightly slower but the memory is faster; the new Air has a 1066MHz bus compared to the 800 MHz bus on my '08 Pro. So far this more than makes up for the CPU speed reduction in my estimation; response time in all applications (even graphically-intensive apps like iPhoto) seems to be the same or better.

There's no comparing disk performance — the Air blows the Pro away due to its use of solid state drive technology. Since the Air doesn't contain a 'disk' at all, but instead stores all of its data in memory chips, the access time for reading files is lightning fast and writing files is also faster. Even better, the Air has a 56GB storage advantage over the 200GB drive in the Pro.

The Air's audio is better, with a new under-keyboard speaker system that sounds fuller and less tinny than the speakers on the Pro.

The battery on the Air is delivering about 6-8 hours of normal use. This is double the amount of time I would get on the Pro.

At 2.9 pounds, the Air is 2.5 pounds lighter than the Pro. Considering the Air's smaller power adapter, the total carry package is about 50% lighter.

Peripheral connectivity options favor the Pro but, with two USB ports and a built-in SD card slot, the Air is more than adequate. However, the lack of a FireWire port on the Air may be a problem for cinematographers.

So, all in all:

• computation and graphics capability are about the same
• audio is a little better
  
• data storage and retrieval is much better
• battery performance is twice as good
• carry weight is halved
• lacking a FireWire port (not an issue for me)

At $1799, the Air turned out to be a good value for me and I've totally transitioned to it as my only personal computer.

Customer Service: One Reason Why Apple is Awesome

Last weekend I bought a new MacBook Air. Rationally, I did it because I wanted to reduce the weight of my travel bag; actually, I did it because it looks cool and I wanted a notebook with solid state storage.

After spending a day with it and getting it all set up for my office and home networks, installing software, and copying files from my older MacBook Pro, I decided to take it in to the office. As I was walking from my car into the building, the strap on my bag somehow unfastened and the bag and all its contents (the new Air and also my iPad) crashed onto the concrete floor.

I unzipped the bag to look inside; the iPad seemed OK, but the Air had a big dent in the corner of the case. Opening it revealed damage to both the top lid that contains the display and the bottom that contains the keyboard and logic board. Miraculously, the thing still worked and there was no damage to the screen. I probably could have lived with it, but the bent metal was keeping the top from closing completely on that side and I was afraid that the display would eventually get damaged from flexing.

So I made an appointment with the Apple Genius Bar at my local Apple Store in Tampa, and took the machine in. I was expecting that the lid and bottom case would have to be replaced and that it would cost me a fair amount of money.

The employee there looked at it, confirmed my view of what had to be replaced, and said that they would fix it — for free!

I was floored. I know that it is going to cost them at least a few hundred dollars in time and materials for this repair. In addition to this, they thought it would take 5 days to get the parts since the machine was so new, but three days later they completed the repair.

So now I'm on the way to pick up the computer and definitely thinking that I am going to be a loyal Apple customer for a long time. Way to go, Apple!

Why the Volt is a Great Idea

General Motors has had its share of bad ideas over the years, but I think they have a really good one in the Volt.

The Volt is a new electric car that will be released this fall in the USA at a price of $41,000. After government tax credits, that price will be effectively reduced to $33,500.

What's interesting about the Volt is its practicality. First of all, from the outside you would never know it was anything special. It looks like a normal car. It has four doors and a roomy hatch. It has room for four people and their luggage. It has a radio, air conditioning, GPS, and OnStar service. You can fill it with gas, just like an ICE (internal combustion engine) or hybrid car. You can also plug it in to charge the battery, just like a pure electric car.

Yet compared to ICE cars, hybrid cars like the Prius, and pure electric cars like the Tesla, the Volt is very unique. Unlike an ICE car, the Volt is driven by electric power, not a gas engine. Unlike a hybrid car, an electric motor drives the wheels at all times; there is no complex transmission that connects the gas motor to the electric motor. Unlike a pure electric car the Volt has a gas engine in addition to a battery.

The battery in a Volt has more power than the battery in a hybrid, but less power than one in a pure electric car. When fully charged it can carry the car for about 40 miles of normal driving. After that, the gas motor starts up and generates the power to drive the electric motor another 300 miles per tankful.

This resolves the key dilemmas of hybrid and pure electric cars. Hybrid cars run mostly on gas so even on short trips gas is consumed and pollution is created; in the end they are only slightly more efficient than normal ICE cars. Pure electric cars run solely on battery, but when the battery is drained it takes hours to recharge, making the car impractical for long trips and causing a persistent fear of unplanned detours and the possibility of being stranded. But the Volt allows the driver to go as long as they want, stopping for just a few minutes every 300 miles to refill the gas tank.

This key capability will be what allows people to finally embrace an electric car as a practical vehicle. Though most trips and daily commutes are fairly short, we all make long trips from time to time and most of us can't afford to have a different car for those longer trips.

I drive about 40 miles to work and back each day, so on weekdays I will use almost no gasoline. Since the price of electricity per mile is radically lower than the price of gas per mile, I will save a lot of money. I will also generate a lot less pollution. Electricity from the grid has the added benefit of being generated from multiple different fuels (coal, natural gas, oil, nuclear) or even renewable sources like wind and solar power. Much of this power can be generated domestically, eliminating dependencies on foreign energy sources.

This is why I can see the Volt becoming a very popular car, and the first really practical electric car to ever be mass produced. Kudos to the engineers at GM for coming up with such a brilliantly simple concept, and kudos to the management for embracing the concept and getting it to market.

'Net Neutrality, Take Two...

Two companies, that is. Google and Verizon, the behemoth of Internet content and the leviathan of US mobile and fiber optic broadband.

Today they announced a proposed "legislative framework" around 'net neutrality. If you're not sure what this all about, here's a brief overview:

The Internet is a collection of networks that carry data between computers. This data is transported in the form of packets, where one logical communication, a video for example, is broken down into a series of packets that are carried from the sender to the receiver. This is a core feature of the Internet, and makes it very resilient because packets can be sent along many paths to get to their destination, so if in the middle of a transfer a cable gets cut somewhere the packets can immediately be re-routed over a different connection to get to the destination.

Now all of this is fine until there are more packets waiting to be sent than can be carried over any one segment of the network. When this happens, packets are delayed, like cars in a traffic jam, and the result is a slower download or choppy video. This can and does happen when there is not enough bandwidth or when bandwidth is shared across a number of users. To solve this problem companies can add more bandwidth by adding cables and equipment to the network; mobile companies might add more cell towers and backhaul circuits.

Another approach to solving this problem is to prioritize traffic, so that some packets get sent with a higher priority than others. Imagine a highway with bus or carpool lanes; while most cars sit in rush hour traffic, carpools and buses zip by at full speed. The network can be just like that, with certain bandwidth reserved for specific types of packets.

Where this gets interesting is that many of the companies that provide Internet connections and carry packets also provide data services. Consider your cable TV company — you can buy Internet service from them but they also provide video and voice services. What would happen if they protected the quality of their video and voice services by reducing the quality of Netflix and Skype? Would that be fair to these other companies?

This is the core argument for 'net neutrality; companies that carry Internet packets should not be allowed to prioritize some traffic at the cost of other traffic. If you prioritize, you are not neutral because you will make that prioritized service better than the non-prioritized service.

The concern is that without rules that require network companies to treat all packets equally, they will eventually resort to prioritizing their services and throttling those of competitors. The cable company wants you to use their video on demand, not Netflix or Hulu. The phone company wants you to use their voice service, not Skype or Fring. Thus, in the end, these alternative services won't get a fair chance because the network companies will choke them out and stifle competition.

I agree with a general policy that would prevent traffic prioritization, and think there ought to be a clear delineation between companies that transport packets and companies that provide data services like email, video, voice calling, and social networking. Without this, the Internet innovation engine will stall and we will return to the bad old days when a few big companies controlled all the information and communication services available to us. Even though there is no longer a telecommunications monopoly, over 97% of the US population gets mobile service from just four companies (Verizon, AT&T, Sprint, and T-Mobile) and a very large portion gets fixed line service from Verizon, AT&T, and a small number of major cable companies, so a small number of large companies have an awful lot of power.

However, in this announcement Google and Verizon take it upon themselves to define a framework where some traffic can be prioritized to allow broadband service providers to offer "differentiated services" whose packets could be prioritized over others. I think this loophole would be exercised to the maximum extent possible by the consumer fixed and mobile broadband companies in order to preserve their revenues at the expense of competitors.

What do you think?

The Ethics of Automated Weapons

I regularly follow developments in technology of all sorts, including in the computing, aerospace, and military fields. One thing that's becoming quite clear to me is that applications of automated killing are rapidly proliferating. As someone who knows a thing or two about computers and the software that runs them, I am more than a little concerned.

In the US, virtually all branches of our military are devising, developing, and deploying unmanned machines that are capable of taking lives. These include aircraft, ships, and ground vehicles.



These systems had humble beginnings focused on support functions like reconnaissance and mobility; aircraft with sensors and cameras that could observe targets and vehicles that could carry gear across rough terrain. They were like remote controlled toys on steroids, reducing the risk to our troops and increasing their combat effectiveness.



As the machines became more robust and reliable, their functional envelope was extended to contain offensive capabilities. It is now common for these machines to carry air to ground missiles, bombs, cannons, and machine guns. We regularly hear about unmanned aerial vehicle (UAV) strikes on ground targets in Iraq, Pakistan, and Afghanistan.

Today, each of these vehicles requires direct human supervision. UAVs have pilots that fly via remote control from thousands of miles away. I sometimes wonder about the morality of killing through a video screen and then going home for dinner with the family. Does this lead to a lower threshold for deciding to kill? Does it remove some of the moral repercussions of killing? Will the reduced risk to our citizens make us more apt to select war as a means to further our national interests?

Even more challenging, the future direction is clearly towards autonomous or semi-autonomous killing machines. There are many valid arguments for this strategy. The time delay inherent in remote communication reduces the effectiveness of current remote controlled vehicles; UAV pilots report that there is up to a two second time delay between their control inputs and the actual change in the aircraft's attitude, and the images being displayed back are similarly delayed. This results in a reduced ability to lock on to fast-moving targets and avoid threats that could destroy the UAV. If the UAV could be directed to a target and then allowed to operate on its own, it could make rapid course corrections that would increase its ability to successfully carry out the attack and avoid or escape from threats. This would be a semi-autonomous UAV, because a human operator still designated the target and decided to attack it. Fully-autonomous vehicles would simply be directed to a patrol area and could decide on their own when to engage a target or flee from a threat.

Additionally, autonomous machines could be programmed to coordinate attacks in real time with other machines in the area; for example a fast moving jet carrying bombs working in tandem with a helicopter hovering behind a treeline, and the helicopter popping up just in time to point a laser designator on a target to guide the jet's bombs thus minimizing the risk to both vehicles and increasing the probability of a successful attack.

The notion of autonomous killing machines raises three key concerns:

1. What if the machine fails to select appropriate targets? In the best case, a valid target may escape and cause more allied casualties later. In the worst case, allies or civilians may be killed. In the realm of asymmetrical warfare, it is almost impossible for humans to make clear choices and avoid mistakes in these situations; my estimation is that even our best computers are not up to this task.
   
2. Will this completely eliminate the ethical burden of making war? With today's UAVs, a person must still, ultimately, decide to 'pull the trigger' and take responsibility for the action. If we remove ourselves from that decision loop, will we still take moral responsibility for the actions of the machines we built and deployed? And if we don't, what will that do to the fabric of society?
   
3. Who will control these machines? One thing I've learned from history is that, generally speaking, a large enough group of people will enforce moral behavior. An army won't attack a civilian population; some amoral subgroup might try, but the majority will rapidly bring them under control. In a world where a few people could control a large number of these machines, however, what would prevent a potential slaughter? And could an army of these machines be used to control the population at large in any country, even our own?

These are serious concerns, and there are probably many more I haven't considered here. Yet it seems clear that this technological development track will continue because it is a natural evolution of our current capabilities and there are reasonable arguments for it. I'm sure that I would have much stronger feelings about this if I lived in a country where these weapons are likely to be deployed, and can only tremble at the thought of being on the wrong end of a gun controlled by a soulless machine.

SDXC Raises Interesting Possibilities

I recently read this article on the new SDXC memory card standard. In early 2010, SDXC devices will store up to 64GB of data and transfer data at 52MB/s, but the industry expects capacities up to 2TB and transfer speeds up to 300MB/s within a few years.

This kind of capacity and performance in this form factor raises some interesting possibilities. Perhaps we'll keep all of our personal files on our SD cards and only keep system files on our computers, making it easier to upgrade systems or use multiple computers during the course of the day or week. All of the new MacBooks announced this week are adding an SD card slot, and many Windows PCs already offer this feature.

This will also make it easier to move data between computers and mobile devices like smart phones and media players.

I also wonder about the possibilities for high performance applications. Imagine a small RAID device with one or more USB 3 interfaces and multiple SDXC slots. Striped and mirrored, an array of 16 cards could provide 16TB of storage with aggregate transfer speeds of up to 2400MB/s. When one of the cards starts to 'wear out' — as flash cards will do over time — you just pop it out and replace it with a fresh one. No need to make backups, and no downtime from disk failures.

There are more powerful flash memory technologies available; FusionIO, for example, already offers high performance PCI Express flash cards for servers. However, the mass market aspect of SD cards means the price per GB will be lower than enterprise-class hardware. This makes it attractive for low-end and mid-range applications, and maybe even for enterprise systems looking to take advange of commodity technology.

What's Google Up To?

So you've probably heard of Chrome, Google's new web browser released this week.

Of course, anything Google does is going to get a lot of attention in the tech community and this was no exception. First there was the 'leaked' comic strip for geeks that illustrated — literally — the high points of the technology. Then there was the Walt Mossberg review (he got an advance copy). In addition, there's an article in Wired magazine (they were clued in a month in advance). It was all over every tech blog before it was even released.

Cool features? Google claims:

• Better reliability due to sandboxing web apps; each tab is a different OS process
• Extreme Javascript performance, 10x better than anything out there
• Innovative tab model
  
• A smarter address bar (with search integration)
  
• Better awareness of phishing sites
• Application shortcuts that make web apps look kinda like installed apps

And I think that last feature is really a clue as to why Google did this. This browser is made for web apps. Anyone that's had issues using Gmail, like pauses, slowdowns, and weird exits knows that much of this is due to the performance and stability of the browser and the behavior of other applications running in other tabs on the browser. By isolating the execution of the code in each tab and using the operating system's resiliency features that keep processes from interfering with each other, this browser will make people trust web applications more and give web apps legitimacy that previously was reserved for locally-installed programs.

Taken together, this added reliability and a radically faster Javascript engine are a game changer for web 2.0. See, Javascript is what makes web pages come alive, enabling programmers to dynamically modify text and images and create all kinds of graphical goodness like sliders, accordions, fades, and other cool effects. So the real goal is for Google to enable web apps (including Google's web apps, of course) to be first class citizens in the application universe.

Have they done it with Chrome? We'll see. Some concerns are that we may see Google pushing non-standard Javascript and CSS features out through Chrome, and using these to make their applications work better. This causes confusion for developers because they'll have another browser to target with its own unique quirks. Users may see apps that lose features or, in the worst case, don't operate at all on some browsers. However, it may drive innovation and get the other players to move faster in delivery of new features; Microsoft, for example, has been lagging in its support for CSS standards and Chrome may light a fire under them to get this rectified.

Google might have been agreeable to implementing these features in Mozilla's Firefox browser, but the Firefox folks probably would have resisted the massive changes because they would have broken many of the extensions that Mozilla-aligned developers have created over the last few years. So now we will see further fragmentation in the browser market and most probably a loss of share for Mozilla's Firefox.

Finally, it should be noted that Chrome only runs on Windows for now, but Google has said that versions for OS/X and Linux are in the works. Given the relative desktop market share, no one would argue Google's choice to release a Windows version first. What they really need to do is convince a large percentage of Windows users to use Chrome instead of IE, or force IE to support Chrome-like features in response to this threat. Either way, Google wins because it opens up a channel for more advanced web applications on the PC.

New House!

We snagged a condo in downtown Tampa on the bay. Here's a picture of it (the tall building in the center).

It's a great place, just across the channel from the downtown area, the arena, and a large restaurant and entertainment venue. We've already been to a few concerts and I'm planning to see a few hockey games when the season starts.

One neat feature is that the cruise ship terminal is right on that channel, so the ships park just a few hundred feet away from my front window. It's really amazing to see those huge ships gliding by in the afternoon when they leave port and head out into the Gulf of Mexico.

New Job!

As of August 1, 2008 I am working at Syniverse Technologies, as their VP of Strategic Development.

What this means is that I'll be leading the development of new products as well as strategic enhancements to current products.

Syniverse is in the business of connecting network operators to each other, and we do this on many levels. The three main categories are signaling interworking (SS7 protocols) for things like roaming registration and call delivery, roaming clearing and settlement, and message transport for things like SMS and MMS. We work with hundreds of large and small network operators around the world.

This is an interesting area to be in now, as we move towards completing the migration to an IP network backbone. Syniverse will be enabling many new service offerings by bridging different companies' applications and networks together.

So far things are going very well, and I've been extremely busy learning about the products and services we offer as well as getting to know the people here. I'm just getting to the point where I can take a breath (and start blogging again!).

iPhone: The Agony and the Ecstasy

So I decided that today was the day to get a new phone — a shiny, happy iPhone 3G.

Little did I know that, at about the same time, the gremlins decided that today was the day to screw with me...

The local AT&T store was still out of stock, but we're lucky enough to have two Apple stores in Orlando. I called the Apple store at the closest mall to me while driving there to check availability and wait time. I waited on the phone for over 30 minutes, but at least I got a prompt every thirty seconds or so telling me how many calls were ahead of me in the queue. After getting to a person, I found out that they had plenty of phones in stock but the customer queue was about 4 or 5 hours long! By the time the call was done I was in front of the mall.

I called the other Apple store at another mall about 5 miles away. They said they had plenty of inventory but almost no line. Go figure! So I headed over there like... well, let's just say 'expeditiously'.

I walked through the mall, entered the queue, and waited for 45 minutes to get into the store escorted by a guy in a pale blue iPhone 3G tee shirt and flip flops. This would be my 'activation assistant'. Then the real fun began.

After telling the pale blue tee shirt guy that I wanted a black 16GB iPhone and the 900 minute calling plan, he gets the device and proceeds to scan in the serial number and SIM card information printed on the box. Then the activation process starts. He collects my existing phone number and my SSN last 4, and verifies my address from my driver's license. All of this was collected in the little mobile POS device. POS is an acronym for point of sale, although another possibility soon came to mind.

Of course, there was something on my account that thwarted the normal activation process. The handheld POS came back with an error, and so Mr. blue tee shirt called AT&T support. The guy on the phone got my account set up with the new SIM and device information, and changed the rate plan. OK, so a little snag but we recovered well. Or so I thought...

Then Mr. blue tee shirt tried to ring up the phone so I could pay for it. Whoops! The device rings up at $499. The price is supposed to be $299. Why the difference? According to the POS terminal, I don't qualify for the 'upgrade price' on the iPhone and have to pay full price, which is $200 more, and I still have to sign a 2 year contract. 

Well, of course, I say that this is not acceptable. You see, I had heard about this pricing issue and went over to my AT&T store to verify that I qualified for the upgrade price. So why the change of tune now? According to the Apple guy, it's all AT&T's fault since their system is the one that says yea or nay to the upgrade based upon my account status. So he calls AT&T again to track it down.

15 minutes later, he's still on hold. Thinking he got mis-routed, he tries again.

35 minutes later, his call is answered and the dialog begins. After a while, it's determined that when the first agent activated the iPhone he made a mistake and went too far by assigning the iPhone service plan. When the Apple system called the AT&T system, it looked like my account already had an iPhone 3G on it, thus I didn't qualify for a 'second' upgrade.

Mr. blue tee shirt and the AT&T rep worked the problem for about 3 hours, and there was no possible way that they could find to activate the account for the upgrade price. There was no override capability for the AT&T system, nor was there a way for the Apple store to adjust the iPhone price.

Well, because the iPhone was activated, my working Blackberry worked no more. So now I had no phone and the prospects were dim.  I suggested that they try to add the iPhone as a second line, cancel the first line and move the phone number over. This looked like it would work, but no, it couldn't be that easy. See, the Apple POS system insisted that the iPhone was already sold and couldn't be sold again.

In the end, the Apple folks gave me a new iPhone to get around this problem. They had to pull the SIM from the first one and put it in the second one to make it work. Unfortunately, they ran out of the black phones an hour before and so I had to take a white one. They gave me a voucher written on the back of a business card to exchange the white one for a new black one when more were in stock.

In the end, I spent over 5 hours and 45 minutes getting the iPhone working; 45 minutes in line and 5 hours pulling hair through the sales and activation process. Apple is going to have to dump two perfectly good 16GB iPhones and probably resell them as refurbished. I have to go back to the store and exchange the phone again (hopefully this time they'll just swap the SIM).

I have a working phone, and it's nice, but all of this leaves me wondering why these systems are so brittle and so restrictive. They've had a long, long time to plan and test. What's the excuse?

Update on Solar Power

Digging into this subject some more, I came across Bright Source Energy, a private company developing solar power. Google is one of the investors, as well as several other energy companies and venture capital companies. 

Bright Source builds and operates 'power tower' solar plants that use a field of flat mirrors to focus sunlight on a collector mounted on a tower. You can watch a short video on their technology here. They claim the lowest cost per watt of any solar power generation technology.

I also found out that there are quite a few solar power stations already operating commercially (at a profit!) in California's Mojave Desert, collectively producing over 300 megawatts since the late 1980s. Here's what one looks like:

View Larger Map

These use an older technology that require parabolic mirror 'troughs' which are more expensive to create and don't generate as much heat as the power tower approach.

Bright Source estimates that covering 2% of the Mojave with their stations can generate all the power needed by the entire state of California. If they're correct, 20% coverage could generate all of the power needed by the entire country. Remember, this is useless desert land, so there's no environmental impact. There are places all over the southwestern USA like this.

This BusinessWeek article has more information on the subject.

Power for the Future?

With the cost of gasoline nearing $5 per gallon in the USA and increasing political instability in the regions that provide most of the world's crude oil, many people are concerned about the long-term viability of our oil-driven industrial society. Some extreme thinkers are worried that we (or our children) will be destined to live in a world that looks a lot like a "Mad Max" movie, but I don't think that's going to happen.

To keep our economy humming we need power. Power for cars, trains, planes, tractors,light bulbs, microwave ovens, computers, televisions, radios, and factories. Right now most of the worlds power comes from fossil fuels, and some comes from nuclear fuels. Both of these fuel sources are limited (there's only so much oil and uranium in the ground) and also have troublesome environmental issues from their production through their consumption.

So when the oil runs out or gets too expensive, what's going to happen? Will we collapse into a dark age of anarchy? I don't think so. Thankfully, we have alternative sources of energy that can provide all the power we need for the future — the sun and the oceans.

Solar power generation is becoming more viable every day. It's now estimated that the entire energy requirement of the USA could be met by building about 100 reasonably-sized solar power stations in the desert southwest. All day long, energy can be generated by keeping lots of mirrors reflecting the sun's light towards thermal towers, concentrating heat to drive turbines that generate electrical power. The excess heat is so great that it can be stored and used to generate power during the night — it's actually more efficient to store energy as raw heat than to convert it to electricity and store it in batteries.

Some of this electricity will be sent to facilities that take seawater and create hydrogen through the process of electrolysis; passing electricity through water causes the hydrogen and oxygen to split. The hydrogen can be collected and used for power where electricity isn't appropriate; in jet engines, long-range road vehicles, and industrial manufacturing processes that require a lot of heat. The hydrogen can be bottled and trucked or sent through pipelines to get to its point of consumption.

Cars and trucks of the future will probably be driven by electrical motors, because they are simpler, quieter, and more efficient than the internal combustion motors we use today. The power for these may be provided by a battery pack or a fuel cell; in the latter case the vehicle will contain a hydrogen fuel tank that can be refilled at hydrogen filling stations.

In the end, we'll have all of our needs met, create less pollution, cause less political strife, and yet still have private cars and single-family homes and all of the luxuries of life we're used to today.  And we'll power it all from the sun and the sea.

The End of the Telephone?

The telephone as we know it has been evolving slowly since its invention in the late 1800s, but many of the basic concepts haven't changed much. The standard 'home telephone' is still an analog device that works over a copper wire circuit. There are, however, some game changing developments that may be signaling the end of the traditional public switched telephone network.

The first of these is mobile telephony. Introduced to the masses in the 1980s, the first cellular phone networks bore many similarities to the analog fixed network but have since evolved into a totally digital medium. Mobile phones introduced the concept that one calls a person and not a place; no matter where I travel, you can call me at a single number. Where traditional phone networks were organized around a strict hierarchy of fixed locations, mobile networks are more dynamic; the endpoints (mobile terminals) can rearrange themselves in real time and the network routing infrastructure must compensate by sending signals over different frequencies, radio antennas, and even to other peered networks.

Soon after cellular telephony was introduced, the Internet began its ascendancy into popular consciousness. At first used for simple things like file transfers, electronic mail, and web pages, the Internet has grown into a reliable real-time data transfer platform for almost any kind of information. With increasing bandwidth and decreasing latency, the Internet can now support real time audio and even video transport. This enables applications like Skype, iChat, and other communication applications to emulate and even interface with fixed and mobile telephone networks. Due to the openness and flexibility of the Internet, these applications can provide additional functionality that telephones cannot, such as real-time presence notifications and the ability to transfer messages, sounds, images, and data files. This opens up many new possibilities for human interaction.

Now we are witnessing the fusion of mobility and the Internet. 3G radio networks can transport data at over a megabit per second to and from mobile handsets, and those handsets have enough memory and processing power to render all kinds of media in real time. Many people, especially younger people, are using mobile phones exclusively, ditching the idea that they need to have a land line in their home.

I wonder how long it will be until the PSTN as we know it is relegated to an anachronism, supported as a backward-compatibility feature for old phones and other equipment like fax machines?

Here are my thoughts about what the future might look like:

• The network will become less visible to people. Right now people have to know arcane dialing codes to make phone calls. In the future,  people will have one or more network addresses that look like email addresses. Instead of dialing 407-555-3223 you will contact joe.difonzo@att.net.
• People may have multiple network addresses. This would allow them to separate personal and business communications, for example.
• A single network address will support all types of communication, from instant messages to e-mail to voice and video calls. The person initiating contact will simply select who they want to contact and what type of contact they want. For example, I may choose to contact John Smith at his personal identity and send a video message, or Mary Jones at her business identity and start a live voice communication.
• Integrated presence management (seen today in instant messaging applications) will let people share their current status with friends and colleagues; this will let my wife see, for example, that I'm currently in a meeting and avoid her calling me only to listen to a ring tone for 30 seconds before she can leave a voice message.
  
• Applications like address books and calendars will be server-based, enabling access from multiple devices. This will allow, for example, my secretary to add calendar appointments while I'm on an airplane. The mobile device will stay synchronized with the server any time it's connected to the network.
• Voice will be transported over IP and not the current circuit-switched network. The mobile network as we know it will become more homogeneous by eliminating the circuit-switched network interface.
• Personal applications will become location-aware, and provide new functions that are sensitive to the location of the user. For example, reminding them that they need to pick up dry cleaning (if it's on their to-do list) when they are in the vicinity of the shop.

I don't know how long it will take to see this transition complete, but it seems that the pace of change is accelerating and we may be at the tipping point sooner than we think.

What do you think?

Out in the Mountains

My wife and I took a trip this week to the Rockies, and I interviewed with a tech firm in Denver.

This is an excuse, I guess, for my lack of posts this past week.

While out here I've had a chance to clear my head a little bit and relax. We went to Rocky Mountain National Park north of Boulder, explored downtown Denver, and visited the resorts at Vail and Breckenridge which were quite busy even though the ski season is over.

The weather was great and the scenery was impressive. If you haven't been out to the Denver area, I recommend taking a trip.

Better, or just different?

While I've been setting up my new Macbook Pro, I've been thinking about why computers are so freaking hard for most people to understand.

The Mac, of course, is supposed to be the paragon of human interface goodness. As I lifted the lid and booted the machine for the first time, choirs of angels were supposed to be singing and dazzling rays of light were supposed to be washing over my pudgy face. Here I was, finally liberated from the dreaded Windows PC that held me in shackles for all of these years. Hallelujah!

However, being completely objective — which is hard to do with a shiny new (and expensive) piece of hardware sitting in front of you — I can't say that using OS X is much easier than using Windows. As I've been attaching the machine to my network, adding printers, configuring preferences, and performing other mundane tasks associated with getting a new computer set up, I've encountered lots of stuff that puzzled me and certainly would be undecipherable to the average person.

Don't get me wrong, I'm not a hater. The overall Mac experience is much, much better than any PC I've worked on. The reliability and security of UNIX-based OS X is a huge improvement over Windows, the amount of software that you get with the machine is incredible, and the total package is tight and pretty.

However, I've come across many cryptic dialog boxes, I've needed to try things multiple times, I've had to repair keychains, and I've spent a fair amount of time Googling to validate problems and find solutions. Once you learn some tricks the answers seem obvious, but until then you could spend a lot of time scratching your head.

I think a major source of this problem is that a computer, as Alan Turing taught us, is an 'anything machine'. Unlike a calculator, a car, or an iPod which has a few basic capabilities, computers can be extended with programs to do thousands of different things. Because of this, no two computers are alike; someone bought the same model Macbook I did on the same day last week, but I guarantee that by now our machines have lots of different software in them. Additionally, because computers can interface to so many things, the differences multiply by the number of printers, network routers, external storage systems, cameras, portable media players, and other types of peripherals that they have to deal with. The probability of any two computers being completely alike within a week of purchase is basically zero.

All of this different software and all of these different peripherals lead to complexity and inconsistency. There are just too many options, too many choices, too many potential conflicts. There's no rule book that says, for example, if you're building a network router, here are the five items that should be user-configurable. In fact, it's just the opposite; in an attempt to gain market share companies try to make their products different, try to add new features, and as a result end up putting a hundred knobs on something that would work almost as well with two. Look at the millions of ways you can configure Microsoft Office! Does anyone really need every one of those options? Of course not, but when you want that one specific option you're going to complain if it's not there. (I'm sure that somewhere there is someone who wants to have page numbers only on every 7th page!)

So we have millions of dialog boxes, each designed with its own sense of a task flow and vocabulary that is relevant to the program or peripheral for which it was designed but not obvious to the unacquainted. As a result, we have billions of users getting confused and potentially screwing up, costing time and money. Yet overall the balance pays off, so people put up with the crap and try to deal with it as best they can. Sometimes they have to call Geek Squad, or a relative who can fix it for them. Sometimes they just live with the problem — I know people who are OK with restarting their computer multiple times a day because it locks up or drops the Internet connection!

I'm going on, I know, so I'll wrap this up by saying that Apple can't really solve this fundamental problem. No single company can. If things ever do get easier, it's going to take some more advanced logic that will hide all of these complexities from the user, not just present them in a prettier dialog box. Something that can analyze all of the dependencies and potential issues and automatically make the best decision based on the intent of the user rather than their direct input. This is a problem that's worth solving!