Welcome Guest. | Log In| Register | Membership Benefits

News

March 27, 2000

Printer ready
Printer ready
Developers Brace For Wireless Boom
continued...page 3 of 3

Related links:
  • Good Wireless Applications Are Short But Sweet

  • Where To Find Software Developers' Kits

  • Wireless Toolbox
  • And from our sister publications:
  • InternetWeek Old Perceptions Hard To Shake (2/21/00)

  • TechWeb Wireless Net Access Heats Up (3/3/00)

  • TechEncyclopedia
    Need a definition of a technology term? Look it up here:


    Send Us Your Feedback
    Users can only see a little information at a time and can't easily navigate or scroll around. There's no back button or "home" key to hit when the user wants to back out of the application. As a result, developers must anticipate what users need and give it to them in the form of simple menus and pick lists. "The real problem is how to make the user interface intuitive," says MapQuest.com's Hopkins.

    The solution lies in carefully selecting the elements of the application and the content you're going to provide, and then putting an extraordinary up-front effort into user interface design. "Flow charting and testing the navigation is the most important part of the application development," says KenTech's Nelson. "You really have to put yourself in the user's experience." Since developers typically work at workstations with large, high-resolution displays, they must rely on simulators to experience the screen as the user sees it.

    The next step is the coding of the application. This will be determined in large part by the device. IP-enabled cell phones equipped with a microbrowser typically support WAP and WML, but neither of these is a sure thing. "With WAP, the wireless industry is creating a parallel Internet. Increasingly, there are solutions that don't require WAP," says Herschel Shosteck's Zweig, referring to recent wireless announcements from Microsoft and Oracle that are not based on WAP. The ideal solution, she says, will incorporate ordinary Internet standards such as the Extensible Markup Language.

    Some patent issues that threaten to derail the entire WAP movement also have been raised. Egan dismisses the patent problems, but suggests that WAP itself is only an interim solution until the World Wide Web Consortium comes up with a better international wireless standard. In addition, the low bandwidth constraint for WAP is set to vanish by 2005, when wireless speeds are expected to reach 56 Kbps.

    In the meantime, developers will likely find themselves writing WML code for WAP devices. For developers familiar with HTML, WML coding will be a snap. WML is a tag-based language, but because there is so little that can be done on a cell phone display-no frames, audio, or color-the developer has fewer tags to bother with. All the major wireless phone manufacturers provide free software development kits (see story, p. 178).

    The low bandwidth, high cost of airtime, and limited memory of the user's device change the way developers structure the application. While HTML developers work with pages and Web sites, the WML developer works with decks and cards. A deck is a unit of screens that makes up a complete function. Each screen, representing one task within that function, is a card. The application sends the deck once and then displays each card as it's needed. This reduces the number of times the phone must access the WAP site.

    Handling user input presents another challenge. To create text, users must punch the number buttons one to three times, corresponding to the letter associated with it. Inputting the word hello, for example, requires hitting the number 4 twice, the 3 twice, the 5 three times for each L, and the 6 three times. You can't ask a user to do this very often. Instead, developers rely on menus of numbered items and pick lists. Where the user must input text, developers try to anticipate the likely response based on the first letter or two and pop up a list of appropriate words for the user to select.

    The technology will improve. Color phone displays will shortly reach the market. In time, bandwidth will increase, input options will improve, standards will evolve, and the multitude of devices will sort themselves out. But corporate IT doesn't have time to wait for all this to happen. Companies need to start planning their wireless applications today.

    return to page 1, 2


    Back to This Week's Issue
    Send Us Your Feedback
    Top of the Page