News
Issue #2002 - 34 (September 2002)
(Updated Sep. 10, 2002)

OPINIONS

Java Falling Short of Mobile Promise

Source: MobileInfo acknowledges ZDNET as the original source of this opinion piece and offers it to mobile computing professional community with its own brief commentary.

Java on mobile platforms is not living up to the promise of letting programmers "write once, run anywhere", according to developers gathered in London last week. The Java programming language was built on the premise that a program written in it can run on any platform that has a Java Virtual Machine. This works to varying degrees on PCs and servers. But as more and more phones reach the market with Java virtual machines built in, it is becoming apparent that the cross-platform benefits of Java simply do not apply here in the same way, said the developers

At stake is the promise of mass-market economics for Java software on mobile phones and PDAs. Without standards, which would help assure that Java written for one phone would work on another, the promised economic benefits may never materialize. Jon Newth, founder of Kuju entertainment, which is now a part of games giant Eidos, said it is the consensus among games developers that Java will be "the next big thing" in mobile phones. "But it clearly has not reached the stage of write once, run anywhere," said Newth. Part of the problem, he added, is that Sun, which created Java and controls the language through the Java Community Process, is not in a position to enforce hardware standards with handset makers.

Brian Rodway, managing director of Affinity Studios, which counts among its clients Nokia and Crawfish Interactive, agrees. "If you go to Sun's Web site there are 56 devices that are J2ME compliant," said Rodway, referring to the Micro Edition version of Java that was developed for mobile phones and other low-powered devices. "But there are 24 different screen resolutions. Write once run anywhere just doesn't happen." A developer who wants to write a game for every device would have to do 24 versions just for the different screen sizes, he noted. "Then there are different sound capabilities, different colour depths and so on. We will never get mass market economics without standards." Even handset manufacturers appreciate the problem.

Paul Goode of Motorola, who chairs the Mobile Gaming Interoperability Forum, said the problem hitting a lot of developers is that they are used to a homogeneous platform with a common set of tools. "In the mobile industry we have managed to take that homogeneity and fragment it across servers, handsets, the networks, and the billing. And we now ask developers to write games." The industry is beginning to realize the extent of the problem. On November 7, 2002, the major handset makers will meet to present their roadmaps to each other, said Goode, adding that they hope to get Sun in the same room to lay out the roadmap for J2ME. 

Later this year a number of industry groups, including the SyncML Initiative, WAP Forum and MMS Consortium, are expected to merge with the Open Mobile Alliance, he added. "If we are to deliver the promise we have to ensure fast development time and deployment," Goode said. 

The current dissatisfaction is a story repeated across the industry, and it is likely to have a big effect on which handsets survive; only the most popular handsets are likely to get the best games. Developers estimate that porting a game to a second phone can cost 20 percent extra in development. Less popular handsets, therefore, even if they have bigger screens and better sound, may lose out to the lowest common denominator in games and applications.

Mobile games developer Morpheme's co-founder and managing director, Matt Spall, said the company has to choose very carefully which platforms it develops for. "We will concentrate on smaller screens because they have more market appeal," said Spall, adding that if there were even five or six standards covering the same number of different phone formats, the situation would improve. "If we can get 80 percent of the code to work across five or six handsets that cover half a dozen standards it won't be so bad." But, he said, it is simply not possible to develop Java games for 24 different screen sizes. Chris Wright of Digital Bridges agreed. "As a publisher (of mobile games), we see a fragmented market," said Wright. "We have to choose very carefully which devices we want to hit." The cost of cost of development, he said, rises fast with the number of platforms that are supported. "We won't hit 24 devices -- if we do so then it will be with very low end titles."

For some, however, standardization is not the key. Digital Bridges founder Kevin Bradshaw, who now heads up investment group Machines That Go Ping, said, "One thing about standards is that they tend to be two to three years behind the edge of innovation. It would be nice to see standards, but developers are smart, and can work around the problem." One thing going for mobile phones, said Bradshaw, is that the number of users is so high--"in the tens of millions" -- that it can be economical to develop better games for the more popular devices, so the manufacturers can use them as a differentiator. Others say that if Java does not fulfill its promise soon, there is no shortage of alternatives. Qualcomm recently unveiled the first major upgrade of its Brew technology, and phones are now appearing with Macromedia's Flash technology, which has the benefits of automatically scaling graphics and text to any size of screen. It is compact, they say, and well established--with many thousands of games already available. 

For more information: www.zdnet.com

MobileInfo Comments and Advisory:  This is a well-written piece from ZDNET summarizing state of affairs with Java as a portable language for handheld devices. Here are our succinct and summary comments orchestrating, in some cases, views of the industry players: 

1. Major problem is not with Java or J2ME but with proliferation of devices. Major device OS environments are now limited to Palm OS, Symbian and Pocket PC. We should have some diversity. Number of devices will gradually come down during the next 24 months. We can not sustain hundreds of device platforms. Screen size is the most important but only one of several differentiations in devices.  
2. We can not and should not expect same monolithic technology environment in mobile/wireless device market as we have in desktops and Windows. We question whether it is desirable in the larger interest of diverse application needs, industry's growth requirements and consumers. 
3. J2ME is by far the best development platform for interoperable handheld and smart phone applications with minimum delta development for supporting additional device platforms.
4. Howsoever imperfect Sun's developer community process is, it is more open than that of Windows. We do not live in a perfect world. We must accept tradeoffs.

Note: This news release may contain forward-looking statements. Readers should take appropriate caution in developing plans utilizing these products, services and technology architectures.  All trademarks used in this summary are the property of their respective owners.