Viewpoint

By Will Iverson

A Note about Java

This is Java month at MacTech Magazine. With a few notable exceptions, we’ve gone hog wild in this issue and put together some solid Java coverage. For those who prefer “serious code,” never fear. Next month, we’ll tackle the Internet from a C/C++ perspective.

Full Disclosure

This month’s issue features a Java IDE comparative review. Given my background, I chose to write the review myself. In my review, I strove to remain as impartial as possible when evaluating these environments, and I stand by the data. In the interests of full disclosure, my former employer was Symantec; my job was Macintosh Development Tools Evangelist, and I wrote Symantec’s Cafe Studio. Dave Mark, one of our regular columnists, is the Metrowerks Discover Java product manager. All of the companies have attempted to bribe me with countless T-shirts and other random tchatchkis (of which, the Natural Intelligence “Magic Boomerang” was probably, well, the most random).

All of these vaguely incestuous relationships aside, what is the final summation? Java development is still a nascent technology, but it has tremendous potential. Java can and will be ubiquitous - it has the proper combination of marketing, technology and spunk which will carry it through this teething period.

To be completely honest, I’m vaguely annoyed by the current Java situation. The potential of Java is so much more than the implementations currently available it’s depressing. You may note that I only briefly mention the performance of the various Java runtime implementations. Apple has announced that they will build Java directly into the operating system. Once this occurs, for all intents and purposes all of the custom implementations will disappear (with one possible exception - Metrowerks has announced that they will be including Microsoft’s ActiveX technology into their runtime. As it appears unlikely that Apple will include ActiveX in their implementation, those who are interested in this technology will have to license the Metrowerks runtime. As of this writing, Metrowerks was unable to give any details of their licensing and distribution plans for their Java runtime).

What Is Macintosh?

Why do we buy and use Macintosh computers? What is the differentiating factor between the PC and the Macintosh? What is at the soul of the Macintosh?

At MacHack and in the circles of friends and coders that I deal with, there has been quite a bit of concern over these issues. Not expressed as such, but as concerns over OpenDoc and Mac OS 8. “OpenDoc is neat,” they whisper, “but if I drag a dozen parts into a document and one part crashes, I lose the whole document.” “Mac OS 8 is interesting,” nervous voices say, “but the Blue Box still leaves most users with the same unstable shell and vulnerable data.” “The lack of finely grain preemptive threads leaves the Mac in the cold for robust Java implementations.” “Serious development on the Macintosh is impossible without protected memory.”

Where Are We Going?

People buy Macintoshes because they are the world’s easiest to use computer. PC people laugh at Macs because they are for “wimps.” In the past I would laugh and respond that “I just needed to get my work done.”

Let me illustrate my point in another way. My grandmother, a retired English teacher, doesn’t care about file allocation block size issues, asynchronous I/O, multi-homing, a decent virtual memory scheme, or any of the rest of that nonsense. Her PowerBook 520c was her first computer. She knew she wanted a Mac (I told her I wouldn’t help her at all if she bought a PC) and she wanted a laptop, so it was an easy choice.

She uses SeniorNet on AOL and is quite active both on- and off-line. She’s even met people and gone on trips with people she met on-line. I’ve basically never needed to help her. She’s asked a few questions, mostly recommendations for software and hardware, but I’ve virtually never needed to do major surgery on the machine. For me, and for millions of Macintosh users, this is why people buy Macs.

For the last couple of years, she’s been dutifully installing updates and patches she’s been downloading from AOL and other sources. She keeps up to date with the latest revs to the AOL software and to the Mac OS. The results? Her PowerBook has crashed quite a bit more often than it should. She doesn’t install as much new software as she used to, mostly because she’s “afraid it’ll make her machine not work right.”

What Happened?

Apple Computer is organized by technology group. You have an Open Transport group, a QuickDraw 3D group, a GX group, and so on. Basic political science tells us that scarce resources breed conflict, and this is telling with the paths Apple has taken. Whatever happened to the PowerBook with handwriting recognition? Killed due to competition with Newton. This scarce resource problem extends to competition for engineering resources. How many applications support the full AppleEvent suite? How many OpenDoc parts do we have for these data types? Most of these technologies support Drag and Drop (it makes a great demo), but few support the underlying infrastructure which makes this technology really work.

Each group builds a great demo running on a base 7.5 machine. Do we have real integration? Does each group do everything it can to minimize the impact of their technology on a user’s machine for support? No. This is why my System Folder is littered with a hundred and thirty-eight Extensions and Shared Libraries. “The quality assurance cycle for integration is too long.”

True Direction

From a marketing perspective, the Macintosh’s key differentiator is ease of use. With that in mind, from an engineering perspective, let’s think about what we need to do to regain the undisputed world title for the Macintosh as the world’s easiest to use computer. If all I was interested in was protected memory and preemptive multitasking, I’d buy Windows NT, but user interface and ease of use are more important to me.

I support the move announced at Macworld of moving away from the monolithic releases - I couldn’t justify the previous Mac OS 8 strategy to my grandmother. Now that a bad strategy has been jettisoned, a new one needs to fill the vacuum. Hopefully, by the time Thanksgiving rolls around and we are all sitting around the table, my grandmother can ask me what is going to happen and I’ll be able to give her a response that makes sense - to her, and to me. In the meantime, you can help decide the direction - send us your thoughts at letters@mactech.com