|
Related Articles: • Affordable Wireless LAN Using Airport |
With a name like WAP, they should have expected it: the Wireless Application Protocol has really been taking a pounding lately.
The attacks are a sudden twist of fate for the protocol that over the past nine months has been hyped as the foundation of the next phase of wireless communications -- specifically, enabling the wireless web.
Predictions differ, but there's some consensus that at least by the end of 2003, more users will access the Web from phones than from desktop PCs (Nokia, 1999). WAP and the Wireless Markup Language have been touted by handset manufacturers and carriers as the technology that will best port web content and applications to a 4-line, 20-character display.
But critics charge that WAP isn't up to the task. Their criticisms fall into at least five categories:
Even with these drawbacks, it's clear there's no stopping WAP, at least in the short term. Nokia, Motorola, and Siemens are already selling WAP phones in Europe, and they're expected to roll out in North America in the second half of 2000. Given that, the objections are best viewed as things for the WAP Forum to keep in mind as it develops WAP v. 2.0.
"The real point about WAP is not that it's not very good," Psion's chief technology officer Charles Davies told the Ninth International World Wide Web conference last week in Amsterdam, "but that it's going to be in 100 million devices."
The first charge, that WAP is an upgrade designed to sell more phones, more air time, and WAP gateways, is one that WAP backers don't duck very low to avoid. Planned obsolescence is a given in the automobile and PC industries, so why not for wireless phones?
For much of what WAP will initially do -- send text data from Web applications in the form of news headlines, stock quotes, and personal messages -- existing digital systems do a decent job. Some people think that moving to a system like WAP that asks more push-button interaction from the user may be a step in the wrong direction.
Tim Zenk, corporate communications director for Xypoint, says developers shouldn't be thinking up applications that force users to push buttons on a cell phone. The natural form of input on a phone is voice. He cites what he calls the 50-percent rule. "Every time you force a consumer to push a button on a cell phone, you lose 50 percent of them." Xypoint's primary business is in wireless location for E-911 services. It's pushing a combination of voice technology for input, voice synthesis for output, and SMS -- short messaging service.
SMS is already a popular service in Europe and Asia, but mostly among teenagers. The system was originally developed for alphanumeric pagers and supports sending text messages (no formatting) up to 160 characters. Several phone carriers (NTT DoCoMo's iMode in Japan and several GSM carriers in Europe) began including it as a feature in cell phones last year. Teens picked up on it as a cool way to send instant messages to each other -- but it hasn't migrated upstream to business and other adult users, who don't have the patience to key in 160-character messages on a 10-key pad.
 |
|
||||
|
|
Related:  Learning WML & WMLScript Sample: Chapter 3: WML Tasks and Events See also: WML and WMLScript Tips & Tricks |
|
|
|
|
|||||
| |||||
SMS may be all right, but why stop there? Gopher and WAIS were impressive in their day, but HTML buried them. Similarly, the people at Phone.com say we're ready to move to a next level that will support markup, interactivity, and security for transactions.
SMS has limited interactivity, according to Phone.com's Roger Snyder. In effect, you're fetching streams of text. WML (the wireless markup language) and its accompanying WMLScript (with some features similar to Javascript) allow forms and input validation (in other words, if it's asking you for a number, it won't accept a name).
WML also offers some display options like font and table control and the ability to display small graphics -- like a map of traffic congestion, generated on the fly. In fact, Snyder says the only argument not to upgrade is the cost of the handset. What's a few hundred dollars a year to upgrade to the latest model for technophiles who want to stay on the cutting edge?
|
Members of the World Wide Web Consortium have accused Phone.com and the WAP Forum it helped found with Nokia, Ericsson, and Motorola of not cooperating with the W3C. During a panel at last week's WWW9 conference, one panelist accused the WAP Forum of creating a separate "WML ghetto," when working within XML and XHTML might have created a more unified system.
Murray Maloney, a W3C member who helped push the SGML's agenda to the W3C to create XML, indignantly charged Phone.com with antisocial behavior. "If you want to work with the W3C, you need to work on it. ... The people who are working on WAP need to get with the program."
Maloney compared Phone.com's current position to the one held by Netscape in the mid-1990s, when it was forging past the W3C's more ponderous deliberation process and creating new HTML tags it said its users wanted, things like the blink tag.
"Now Netscape is what it is, it's part of AOL, and it doesn't have that kind of control anymore." Woe to the corporation that doesn't play ball with the W3C.
In general, some complainants at the conference seemed to resent that one particular technology for accessing content on the Web -- specifically, wireless phones -- seemed to be getting special attention because of media hype. But given the predictions that phones will be the way more than half of all web users access the web, is the special attention warranted?
Most web developers currently check their designs on several of the most popular platforms, to ensure that the pages designed on a Macintosh with a 21-inch screen will be legible through Internet Explorer 5 on a Windows 98 laptop as well as through Netscape 4.5 on a Linux box. Those who really want to ensure accessibility test their pages on somewhere between 15 and 20 browser-platform combinations, including Lynx and the W3C's Web Accessibility guidelines.
Early indications on designing content for WAP indicate that the situation is about to get a lot worse. Don Schuerholz, the manager of Phone.com's developer program, told a tutorial audience at WWW9 that they recommend web developers branch their code at the browser and create two interfaces, one for Nokia WAP phones, the other for Motorola. Why's that? "WML's still a little fuzzy on the browser implementation."
And that's in the second quarter of 2000. Imagine within a year when WAP phones with multiple screen sizes are coming from Ericsson, Samsung, Sony, Siemens, and more. Screen resolution and sizes vary. There could be up to 300 clients out there, said Ericsson Telecommunications' Egbert-Jan Sol, speaking at WWW9 last week. "We are going in the wrong direction."
Schuerholz and others put their faith in XSL (XML Style Sheets). Ideally, one should be able to code their data in XML and have XSLT format it for delivery based on the identification of the user agent.
But don't hold your breath. Style sheets for the Web have been around since 1996, and most sites still mark their code with formatting tags. Publishers and application developers who wait for a magic black box that will deliver XHTML to the desktop and WML to the handset may be waiting a while yet.
|
Even as WAP phones roll into stores to deliver light web content to wireless devices at late-1980s modem speeds (9600 bps to 14.4 Kbps), the hype has shifted to faster wireless networks that may not need to relay on gateways to convert data between web and WAP formats.
Nokia recently tested a packet-based wireless service, GPRS (General Packet Radio Service), for the first time out of the lab. Unlike circuit-based services, which rely on a single, direct connection for the length of the call, packet-based services work like the Internet, delivering data in small packets over an open channel to their end destination where they are reassembled.
GPRS backers were originally hyping the system's promised speeds, in the ISDN range around 115 Kbps. But just recently, according to Psion's Davies, they've dropped back from these claims and are instead pushing its "always on" feature as the prime selling feature. The reason: callers share a data channel with everyone in their cell. As with a cable modem or LAN, the more people online, the slower your stuff moves. Real speeds, Davies said, are more like 43 Kbps -- and that's with only two callers on a cell. Add 14 more and it drops to a crawl.
Even if GPRS doesn't deliver, G3 networks are being envisioned to deliver rich multimedia content to wireless devices. These third-generation networks may deliver speeds comparable to DSL or T1 lines. Combine that with greater processing power, and the need for WAP as web protocol lite fades away.
Perhaps the most damning charge is that incompatibilities between the security protocols for WAP and the Web require Phone.com's WAP gateway to first decrypt messages in one format and then re-encrypt them in the other. Secure messages that come in from the caller using WTLS (WAP transport layer security) must be decrypted, then re-encrypted in something like SSL before being passed onto the bank or credit card company or whoever's receiving it. That means that for a moment, the transaction is insecure. (That's the moment you paint the bulls-eye on.)
Phone.com's Schuerholz says the problem's being addressed in WAP version 2.0 under design, with the goal of allowing the merchant or bank's secure server to include WTLS encryption. Until then, he says, transactions can be called secure if the WAP gateway sits within the secure party's firewall.
Such gaps in forethought lead even some of the technology's strongest advocates to look on it as something of a stepchild. "I guess we all agree that WAP, the way it is today, sucks," said Häkan Mitts with Nokia's mobile phone research unit, speaking to a WWW9 audience.
But it may be that it only sucks in the eyes of technical researchers who had their sights set on something better. For the hundreds of millions of users expected to log onto WAP phones over the next few years, the experience of accessing their favorite web content and applications through a tiny green interface may be good enough to keep them fascinated ... at least until the next great thing comes along.
David Sims was the editorial director of the O'Reilly Network.
Discuss this article in the O'Reilly Network Wireless Forum.
Return to the Wireless DevCenter.
Copyright © 2009 O'Reilly Media, Inc.