Who Needs PCs?

Friday Oct 27th 2006 by Steve Andriole
Share:

Given communications technologies and trends, our IT/Biz Alignment columnist writes, it makes sense to invest much more in the server than the client.

There’s not much one can’t find, analyze, or purchase on the Web.

The past 10 years has seen the evolution of the Web from a passive repository of information to a proactive pusher of user-generated content and an enabler of personal and professional transactions. Educational curricula, music, films, surveys, customer service portals, travel planning, job placement boards and personal matchmaking services are all on the Web.

For some, the Web is so deeply woven into the fabric of their lives that it’s impossible for them to imagine a disconnected world. (I think I may be becoming one of these vagrants living in as much in digital as physical space. This year, for example, I am working on giving up paper.)

Let’s argue that the Internet is the ultimate virtual server and all that anyone needs to access its content and transaction capabilities is a very thin, throw-away client. The argument obviously is that we should focus much more on the virtual server than on the device used to access it.

In fact, given communications technologies and trends, it makes sense to invest in the server much more than the client. (There’s also the digital divide issue: the cheaper the access device, the more people can participate in the digital revolution.)

Read More IT/Biz Alignment Columns
Taking It To The Board

The Next Wave of Commoditization

Who's In Control?

Ignorance Is Anything But Blissful

What's Your Technology Center of Gravity?

Let’s look at several trends that point to why this approach makes sense. But now I believe that in addition to helping companies compute more cost-effectively, thin clients can help everyone exploit the Web -– regardless of their lot in life.

First, network access is essentially complete: we use desktops, laptops, personal digital assistants (PDAs), thin clients, and a host of multi-functional converged devices, such as integrated pagers, cell phones and PDAs to access local area networks, wide area networks, virtual private networks and metropolitan networks. The networks work, damn it. Are they perfectly secure and 100% reliable? Not quite, but we’re getting there. And for those who worry about the collapse of the Internet, there’s enough redundancy and reconstitutability in the technology to make us sleep well -– if not perfectly -– at night.

Small, cheap, reliable devices that rely on always-on networks make sense. Shifting computing power from desktops and laptops to professionally managed servers makes sense. Moving storage from local drives to remote storage area networks makes sense. Fat clients should lose some weight -– a lot of weight -- as we bulk up our already able, under-utilized servers.

One way to approach this is to begin with what the ideal converged device might look like and then strip it down to make it as thin and cost-effective as possible -– while still allowing it to be functional. Some of the characteristics of today’s “fat clients” include:

  • Small (pocket-able; about the size of a larger PDA on the market today) with as large a screen as possible within the given form factor
  • Touch-screen
  • Expandable memory
  • GSM or CDMA cellular phone service with broadband (EV-DO or EDGE)
  • Address/phone book
  • Mp3 playback (via broadband or memory)
  • Video playback (via broadband or memory)
  • Web browsing with full Java support
  • Popular OS allowing a wide-range of programs
  • Camera with video capabilities
  • GPS with full mapping capabilities
  • Bluetooth (file transfer, keyboard, headset, etc compatibility)
  • Wi-Fi

    The cost for such devices is in the $700-$900 range. Over time, we can certainly expect the cost to drop, but the ongoing maintenance and replacement costs for such devices will remain substantial.

    What if there was another way to exploit all of that content and transaction processing capability? What if we could develop devices so thin and cheap that everyone could afford one? Here are some of the characteristics they might have:

  • Even smaller form factor (comfortably pocket-able; about the size of a flip mobile phone), as large a screen as possible or preferable in that form factor
  • Touch screen
  • GSM or CDMA mobile phone service with Internet with broadband
  • Web browsing with full Java (or any open standard) support that makes email client, word processor, audio/video playback, etc. available without installing applications, which can be accessed directly from the Web

    The technology is here to make these devices in these form-factors, so long as the remain “open.” Wide-area wireless network technologies, such as WiMAX, have the potential to drastically reduce the price of such devices. With WiMAX, companies (or municipal governments – as they are now beginning to do) could blanket entire cities with Wi-Fi-like broadband Internet service.

    This would not only enhance the “always-on” nature of devices, but could potentially render the entire mobile phone industry obsolete due to VoIP services. Devices would only need the Internet.

  • Over time the cost for ultra thin devices will be less than $100 (they might even be free as new pricing models emerge for Web-based transaction processing). Thin Web clients will become throw-aways eliminating completely the break-and-fix/replacement cycle that plagues so many IT shops and frustrates so many not-so-tech-savvy users. Pricing trends will also make the devices affordable to just about everyone.

    There’s great appeal in stepping away from managing any aspect of the communications infrastructure or content management on fat devices that require substantial care and feeding. When the industry first started thinking about thin clients -– even before Larry Ellison’s appearance on Oprah Winfrey’s show over a decade ago –- everyone understood the network and transaction processing implications of thin client architecture.

    In those days there was a lot of uncertainty about just how to power the transactions that a 24/7 network would deliver. But more recently, architectures have developed that suggest just how a thin client/fat host might work. New service oriented architectures (SOA) will make it possible for transaction power -– and flexibility -– to reside on distributed servers capable of communicating and fabricating transactions at a moment’s notice. SOA combined with AJAX (asynchronous JavaScript and XML) will make it possible for consumers to use incredibly skinny devices to accomplish all sorts of Web-based activities.

    What this all means in practice is that our ability to extend distributed computing is growing dramatically and that new architectures will make it possible to imagine all sorts of seamless, instant communications from all sorts of devices -– including ultra thin ones. What a wonderful world it will be.

    So the next time you think you need a PC, look at what thin clients can do for you. I suspect that you may not need to lug all that capacity around -– that you can get pretty much what you want for a whole lot less money, pounds and hassle.

    Share:
    Home
    Mobile Site | Full Site
    Copyright 2017 © QuinStreet Inc. All Rights Reserved