Thursday, February 17, 2011

The future of computational devices?

Imagine, for a moment, the computer you're reading this post on.

What type of computer is it? Is it a traditional desktop? A notebook or a netbook? What about a tablet or a smart phone?

Your options on what you use to access information are continually growing, even now they are several times greater then they were just a few years past.

If you are on a traditional computer, say a desktop, what kind of specifications might it have?

A modern 2010-era computer, sold for a reasonable price, might have a set of specifications like this:

* Dual-core processor
* 500 GB Hard drive Storage
* 4 GB of System Memory
* 512 Dedicated Graphics card with 3D acceleration
* Multi-channel sound system

What sized box is your tower? Is it a larger, standard ATX-sized unit, or maybe one of the small form factors?

Whatever the size, I want you to imagine taking that desktop and shrinking it....continually smaller and imagine a computer with similar specifications, but with a form factor the size of your phone.

Sound crazy? Well, consider my own smart phone, a Nokia N900, with the following specifications:

* 600 MHz ARM Cortex-A8 CPU
* 256 MB System Memory
* 32 GB Storage
* PowerVR SGX 530 GPU supporting OpenGL ES 2.0
* Stereo sound system

Not too bad. In fact, as little as decade ago, those specs would probably have been fairly impressive in that desktop your on right now, wouldn't they?

Is it really that crazy that the technology in smart phones could approach the level of desktops? I don't think so.

Consider laptops. Not that long ago, people who chose laptops for the portability advantages they offered were forced to sacrifice the performance of desktop. This is no longer true, as laptops have reached complete parity with desktops in terms of the specifications and abilities.

Those of us today who continue to choose desktops mostly do it for form factor reasons, for example my high definition 22 inch display, full keyboard with number pad and mouse. Of course, these things can additionally be added to a laptop. Other uses for desktops over laptops might include, like myself, use as a DVR (more easily permanently connected to my TV and cable box) or the ability to have multiple disc drives and the like.

Nevertheless, choosing a desktop today is more about form factor and preference then specifications.

In fact, I dare say that while smart phones, net books and tablets continue to make leaps and bounds each year in the amount of power they offer, the traditional computing paradigm of desktops and laptops seem to have plateaued.

For example, why don't we commonly go our local computer stores and see 8 GHZ processors and computers with 48 gigabytes of memory? Are we finally seeing a plateau of Moore's law? Or is the slowdown more for marketing and business purposes?

In fact, one of the problems with sticking more and more transistors on a chip is that the damn things get too bloody hot. Who needs an Infinity-GHZ processor when you need to burn thousands of watts of power just to keep it cool?

Why, even the modest Athlon chips in my two previous laptops could get into the very uncomfortable (and dangerous) 80-90 degrees centigrade range. Had they kept with the numbering convention, I'm sure the slogan for the the Pentium 5 would have been, "Now, you can cook toast on it too!". On the other hand, the Athlon X2 250 processor in my desktop rarely gets above 30C, nor does the Intel Core Duo in my laptop.

But the fact of the matter is that we don't need never increasing clock rates and increases in memory to be happy. In fact, I remember reading an article several years back (that I unfortunately can't source) suggesting that the major chip manufactures such as Intel and AMD would soon stop trying to increase their clock speeds and instead focus on the chips they got: basically, trying to shrink them down and make them more power efficient. This a good thing, not just for your power bill, but for the environment too.

It seems that we are living this reality: Processors aren't getting faster, but they are getting cheaper, smaller, more efficient and multi-cored. We need this more then we need more gigahertz, because there is clearly a limit of diminishing returns. We don't need faster computers because we don't have applications (unless you are in the server or HPC market) that can use them. At least, not yet. Even my desktop with a modest 2GB of Ram runs circles around many computers of better specifications, DVR'ing, web browsing and play games at the same time. Of course, I use a far superior operating system then most :).

So what does that mean for the future of such devices? If laptops and desktops continue their plateau, and the smaller form factor devices such as smart phones continue their rise, will we eventually reach a point where they are all at parity?

It wouldn't surprise me. Likewise, it also wouldn't surprise me if the day comes when your entire computer system fits in your hands, and that's the only computer you need.

For example, imagine a smart phone 10 years from now. We'll consider this our speculative "super-device". It can be connected to a GSM or CDMA network, likely has wi-fi and cellular data capabilities, camera and GPS, plus also a large touch screen and optionally a physical keyboard. It can make calls, play the newest high-end games, browse the web, has storage in the hundreds of gigabytes, extremely fast data transfer and processing rates, and more.

What are the disadvantages of this device? Well, no body wants to stare at web pages on a small screen forever, nor do they want to type up their reports on a keyboard only a few centimeters big.

But wait! Picture another device, in the form factor of a laptop, with a large screen, full keyboard, optical drive and card reader, larger battery perhaps, etc. Except that this device is just a "shell", it has beauty but no brains. No processor, motherboard or memory of it's own. Instead, slide your smart phone into a receptacle and suddenly you can have an entire computer system ready to rock. Able to type reports, see movies and web pages on a larger screen, even play the latest visually stunning computer games.

But why stop there? Don't need a keyboard? Just provide a large touch screen dock, sans keyboard, for your smart phone with receptacle and suddenly you've got a fully functional tablet (or e-reader). Add a keyboard with no optical drive and you've got a net book.

Need a larger screen for those high definition movies/games, or want to use a printer? Just provide a small dock which is nothing but ports, for monitors, printers, keyboards, even DVR connections if you want, and there is your desktop.


The receptacle could also be integrated into cars, essentially taking over as the entire entertainment and communication system of the vehicle.

I fully feel as though this is the natural evolution of where technology is heading. But is it a good idea? What are some of the pros and cons of such a design?

Right now, I have three "computers" that I use on a daily basis. My desktop, my laptop and my smart phone. Each has it's own place in my technological arsenal. My desktop of course serves as my main "home" PC: it does my DVR'ing, plays games, lives as my music and media server, browses websites, check my personal email, Skype conference with my family and more. My laptop is mostly work oriented, it has all my work schedules on it, current projects, contact information, work email, etc. But I also occasionally use it when I travel for web browsing, watching movies, etc. My smart phone, while of course admirably fulfilling it's capacity as my only phone, also handles all my personal schedule, memos and todo's, plays games and browses the web, at 5MP doubles as my primary picture and video camera, and is a full Sat-Nav GPS device with voice guided directions.

I'd be lying if I said the thought of all those devices being combined into one, but each with it's own profile what I wanted to do at the time, wasn't appealing to me. It's easy to get into a state of 'digital fatigue' when you are surrounded by too much technology and want to simplify things, only to feel your current technology is unable to fulfill your needs in some form or another. Even I find myself wanting a tablet, net book, or second laptop, even though I can pretty easily convince myself that I don't really need them. And on top of that, I still have game consoles, several televisions, DVD devices, and so on.

But there is danger as well. Phones are of course designed to be robust, they have to be, being jostled around all day after all. There are significant dangers in putting all your eggs in one digital basket: what happens when your phone gets destroyed, damaged or even just lost?

This could have some pretty bad consequences. But there are other problems as well, for example, Vendor lock-in. Just because you buy your device from Vendor A, you shouldn't have to buy your shells from Vendor A. For such a system to work, the dock and protocols should be entirely open and implementable by all.

The idea of an "all in one" device capable of doubling as any computing device we have today excites me a great deal, though there are pitfalls that I seriously hope we can avoid in order to realize such a device.

There is one pitfall we might not be able to overcome: upgrade-ability. A properly built desktop can be upgraded endlessly, to the point where it is an entirely new computer. Laptops are also upgradeable, but to a significant less degree: the hard drive, memory, battery and optical drives are often changeable but good luck trying to upgrade the screen, motherboard or video card. Unfortunately, as the form factor gets smaller, the ability to upgrade decreases proportionally. Good luck trying to change the memory in that smart phone, or adding an optical drive to that net book.

To make our speculative super device, we want to keep two principles in the back of our mind at all times: longevity and recyclability. We've already made the assumption that the specifications of all devices types would largely plateau out, become equal. But I'm not saying that at this point technology growth would stop, merely that the growth of the three major form factors (desktop, laptop and smart phone) would all grow at the same rate. There will still be advances as people develop new technologies and find uses for them. So technology *will* advance, albeit and hopefully at a more sustainable pace.

I think these devices would need to have a long life span, technology sufficient to last as long as possible. And, when you are finally ready to get a new device, we need programs in place to reuse, resell or recycle the old ones, possibly even taking off from the price of a new device.

Could a device/system like this ever become mainstream? Companies such as Motorola are already taking the first step with their Atrix phone (though I've heard rumors the laptop dock is only available with certain plans...which doesn't bode well). Just imagine if a company, say Apple, announced tomorrow that they had a new iPhone that, with the right dock, could also be your iPad, MacBook and iMac? Would not flocks of people swarm out to buy it? I think so. And the other major vendors, Dell, HP, etc would all follow while Microsoft would probably try to slap Windows on everything. Unfortunately, it might not be in the best interest of these companies to work together, which would create a hell for consumers.

Ideally, I would like to see everything left as open as possible. I could go on for a good length of time on how I believe in the decoupling of hardware and software, but we shall save that for another post.

The only way I would like to see this happen is if people are in control of their own devices. For example, as a strong proponent of free and open source software, I'd want to be able to run my own operating system on my device, and still have my hardware work and interact with other devices. We can place extra security and encryption on the devices (biometrics, perhaps), to help prevent the devices from being compromised if lost.

The phone component needs to optional. We can add a SIM card slot onto the device, and hopefully, carriers and manufactures will allow you to hook up to their networks seamlessly. The phone itself would be little more then an optionally installable application on the device. Hopefully carriers would remove those ridiculous data caps on their networks...but I know that is likely little more then a dream.

What about dedicated uses of the technology? Like I said, my desktop doubles as my DVR, and my ultimate device that I envision will hardly be able to record television shows for me if it's in my pocket on the other side of town.

This could be where device "reuse" comes in. In any case, there are likely to varying types of devices with different hardware capabilities. So it's not that crazy that I could use an older one, or cheaper one properly configured for DVR use while my main device stays with me.

We may still end up with multiple devices, but the fact is that the flexibility and configurability of the devices would all them to act as any other device, which would ultimately reduce the number of simultaneous devices we need at once. And with things such as longevity built into the device, they would need to be replaced less often, while the form factor can no longer improve.

I think such a technology has great potential. It's reasonable to implement, and could revolutionize the way we interact with our devices. But is has pitfalls as well, aspects we need to carefully avoid and implement properly if want to be successful. Nevertheless, I believe it is likely where we are to be headed, hopefully it'll be more of a blessing then a curse.

Do you agree? Feel free to share your thoughts and feelings in the comments, and have a great day!

2 comments:

  1. LOL, "Now you can cook toast too!"

    All my devices already "talk" to one another. My phone talks to my desktop, my laptop, my GPS device.

    Currently, I pay lots of $$$ to lots of companies for each device I have, if there was only "1" device and a bunch of "shells" (presumably) cheaper, then the manufacturers will make less $, so it doesn't seem in their best interest to move this way.

    I am excited for the day my devices talk to my car/mini-van :-)

    ReplyDelete
  2. It's true that it might not be in the interest of manufactures, but they might not have a choice. The point is that all it takes is one popular company to release a single phone/handheld on par with todays desktops.

    Smaller/Larger companies will see the potential market here and began producing the third party shell components, others will follow suit, and the market will take off.

    You don't even need large market share to be successful. Just look at Apple as an example. Get the right brand, and all that stuff comes later.

    Thanks for the comment!

    ReplyDelete