Technology Futures

A Mach 10 (ten times the speed of sound) test of a scramjet took place in Australia, where a rocket carried the scramjet into near space, then ignited the scramjet to return to earth. Scramjets are special jet engines that work at very high speeds and at high altitudes, and research has continued on them for decades with mixed success. A successful and reliable scramjet design would allow travel between London and Sydney, Australia in as little as two hours.

FiOS is Verizon's fiber offering, which they are now rolling out in limited areas. According to this article from someone who signed up for FiOS TV and then dropped it, there may be some real limitations (note: scroll down the page a bit to the "FiOS TV" title).

The problem is that FiOS partitions the bandwidth available to a home into three separate portions: TV, telephone, and Internet. But to do on-demand movies and video, Verizon has to use the Internet/broadband partition, not the TV partition. So as the note describes, you may see much of your promised bandwidth eaten up for a movie.

What's the problem you ask? Suppose you are working from home, using your FiOS broadband partition to connect to the company network for VoIP and company systems. It is summertime, and your kids are home. It's 10 AM, the kids are bored, and they order a two hour on-demand movie. For the next two hours, you may see your available bandwidth drop substantially. As the author noted, this may or may not actually be an issue, but Verizon would not provide any details.

In a open services, Active Optical Network (AON), vendors like PacketFront run all services in a single large network connection to your home, and on-demand video does not affect your Internet access because each service is managed end to end with guaranteed Quality of Service (QoS). It's a different and better design that does not have the limitations of the FiOS PON (Passive Optical Network) design. Disclaimer: Design Nine often recommends PacketFront systems for community broadband networks.

This limitation of FiOS and PON systems is a real and serious economic development issue. As more and more job and work opportunities allow people to work part time or full time from home, communities need broadband systems that fully support business use of the network. Communities that think FiOS or AT&T's Lightspeed services are enough may be disappointed as economic development in their region begins to stall out when workers and businesspeople discover these networks don't support business class services very well.

In the biggest change to personal computers since the arrival of an affordable 5 1/4" hard drive about 25 years ago, solid state "hard drives" will be appearing soon in laptops. The solid state storage devices have no moving parts, use much less energy, weigh less, are faster, and are more durable.

Laptops will shrink dramatically in size and weight. Weight will go down directly from the reduced weight of the replacement solid state drives, but also indirectly; spinning hard drives are the biggest drain on batteries in a laptop. The replacement solid state memory will allow both smaller batteries and much longer useful work time on laptops.

Laptops will be thinner, lighter, smaller, and faster, and won't cost much more because of rapidly dropping prices. In less than four years, you probably won't be able to buy a laptop with a rotating hard drive, and "hard drives" will quickly become quaint relics. In fact, within five to seven years, we probably won't even have hard drives on most (but not all) desktop machines.

From an environmental perspective, computing is rapidly moving from being big wasters of energy (CRTs and disk drives use a lot of energy even when the computer is otherwise idle) to being relatively green; the combination of low power LCD flat panel screens and solid state memory will cause dramatic drops in the energy footprint of computers.

At least part of the reason that health care costs so much is due to arcane software and poor information systems design. I had to have a routine blood test this morning, and went to the hospital to do it so I could take care of it early in the day.

The whole process kicked off at a desk in the lobby where I had to fill out a slip of paper that was carried by hand to the patient registration folks. After a twenty-five minute wait, I finally got to talk to a registration person. This is where everything went seriously off the rails. I've been to the hospital before for this sort of thing, and they had a record of me in the computer. But the poor woman registering me clicked away with her mouse and typed for six full minutes. I thought I was on line at at airline registration desk--the only other place where I see long periods of typing going on for simple tasks.

The registration process generated at least 18 sheets of paper that I could see, and I suspect several more were generated at printers whirring in the background. Part of the time was spent checking data like date of birth, addresss, and phone number, which they already had but had to ask again, in a tedious process repeated daily in tens of thousands of doctor's office and medical facilities around the country. And of course, there was the requisite photocopying of my insurance card. I suspect that every day, we probably use a forest's worth of trees making copies of insurance cards.

The actual process of taking a blood sample, including being greeted by the nurse, rolling up the sleeve, printing labels for the samples, and putting on a band-aid, was under three minutes.

Total elapsed time: 40 minutes.

What should and could happen? This business of photocopying the insurance card is insane. The hospital has computers. The insurance company has computers. We have this thing called the Internet. Insurance companies and health care providers could establish an open standard to verify insurance coverage electronically, with much higher reliablity than a worn piece of pasteboard carried around in a wallet.

Today, you can buy a thumb drive with 2 gig of memory on it for under fifty bucks. This would hold my entire medical history, from birth, including a good sample of important X-rays. It could also hold some personal information like address, phone number, and next of kin. Instead of endlessly and repeatedly manually filling out forms and entering data, we could carry one of these, with something like thumbprint biometric secure access. Instead of many minutes of typing, I would plug it into a little keypad at the registration desk, enter a PIN number and a thumbprint, and everything needed to would transferred (and only what was needed, under the patient's control). Quick, easy, and more accurate than hand data entry.

This is not a difficult technical problem. I suspect it has more to do with the health care establishment's fear of giving more control to patients. But ultimately, we pay, one way or another. We should be in control, not underpaid and overworked clerks (who are really controlled by IT staff who have failed miserably to be of any real service to their clients and customers).