Greg Landweber - A ResExcellence Interview

Greg Landweber - A ResExcellence Interview by Michael Coyle

In 1995, Mac users anticipated the release of Apple’s next leap in operating systems. Code named “Copland”, a preview CD containing a Macromedia presentation was released to thousands highlighting new features and the new user interface. Collectively drooling over the menus, progress bars, and 3D icons, we would have waited years for the Platinum look if it were not for the efforts of Greg Landweber.
Together with Ed Voas, Greg released the first public beta of Aaron a few days before MacHack '95. He returned from the coding convention to find his In-Box stuffed with over 100 emails about the new software. A few months later, version 1.0 was released on info-mac and a growing online Mac community clamored to download the small extension that turned the plain look of System 7 into flashy Copland.

While Landweber had released two other successful utilities a few years earlier, Greg’s Buttons and Greg’s Browser, it wasn’t until Aaron 1.0 in July of 95 that the Macintosh community took notice of his work in a big way.

It’s hard to believe that in 1980 at the age of ten, Greg got his start programming BASIC on his elementary school’s Radio Shack TRS-80! When the school upgraded to an Apple II+, Greg was soon staying after school teaching LOGO to other students. Shortly after his parents bought one of the first Apple IIe computers, Father and Son spent the summer together learning PASCAL.

If you were interested in learning a programming language, which one does Greg think would be a good place to start?

"I would recommend Perl. It is a beautiful programming language. It is easy to learn and it is very handy for writing CGIs for the web. I have heard very good things about Java, but I have not yet tried my hand at it."

If you are running MacOS X Beta, it comes with a fully function install of Perl. Open a terminal and type man perl for more info.

Before heading off to college, Greg bought a Mac SE and made a little money during summer breaks programming in BASIC while learning C in his spare time. But it wasn’t Computer Science that he declared as his major, Greg was interested in a career in mathematics that would eventually lead to the cutting edge of physics and String Theory! Does this somehow compliment Greg's interest in computing?

"I'm afraid my mathematical research is not directly related to computer science. However, I do know some people in the quantum computing community who believe that string theory might have some applications in quantum computation... although from time to time I have considered leaving academia to write software."

Long before Steve Jobs took back the reins at Apple, their lawyers were busy stopping people from distributing the Appearance Themes that were leaked from an early version of MacOS 8.5. Recently, they have cracked down on sites distributing Aqua Themes and Schemes. How is it that Aaron, which copied the Platinum Theme before its release, avoided the legal eagles in Cupertino?

"When I released Aaron it got Apple's attention, but it took them a while to decide on their response. My first contact with Apple came via two separate e-mail messages I received on the same day: the first from a manager telling me to stop work on Aaron, and the second from an evangelist offering to work with me to improve it. Fortunately, the evangelist won. Apple gave me their internal specifications for the "Apple Grayscale Interface" and hooked me up with Arlo Rose. They also hired my shareware partner Ed Voas, who is now tech lead for the high-level Carbon Toolbox."

Apple was only willing to permit Greg to get so close to emulating their future OS release. When he displayed a demo version of Aaron that had Popup Windows, Apple squashed it.

As far back as his first shareware releases, Greg realized users wanted more control over the look of their Mac desktop. As Aaron began to, assume more of a roll in the Macintosh interface, a new tool was needed. Started by Ed Voas, Kaleidoscope was built from the ground up to totally redesign the desktop. When Apple picked up Ed's resume and offered him a job in 1996, Greg continued development and Kaleidoscope 1.0 was released on November 1, 1996. It was an immediate success.

An updated Kaleidoscope 1.5 allowed users with a little skill to create their own desktop design. Today, the Scheme Archive allows people to download thousands of different schemes. By the time Kaleidoscope 2.0 was released, Arlo Rose had joined the team. With his arrival came more legal rumblings from Apple.

"Arlo had left Apple and had started working with me, and Apple had legal issues over the ownership of his interface designs. Apple's lawyers were also concerned that Arlo had leaked their Appearance source code, and it took us two months of expensive legal dealings to convince them he hadn't."

During the beta cycle of Kaleidoscope 2, Arlo included a beautiful scheme titled “Mekong”. Originally designed before he worked at Apple, it was never an Apple project. Other than a few friends at the Infinity Loop Apple campus, the company was never formally aware of its existence. Yet when a Kaleidoscope beta was leaked to the public, Apple stepped in to block the scheme's release claiming it was their intellectual property. [This has been previously covered at ResExcellence and should give every professional pause when it's time to sign an Intellectual Property waiver.]

With Kaleidoscope as a rock-solid product, Greg’s shareware offerings were a hit. In fact, I suspect his products are used on more computers than many boxed commercial applications. Greg spends a couple hours a day and most of the weekend working on his products and answering email. His partner Arlo handles design issues, builds the software installers and handles many of the legal and business aspects of selling shareware. Initial partner Ed Voas is hard at work on Carbon at Apple.

To put together a successful shareware business, do you start out with an organized plan, or just decide you want to give it a try?

"I went through the "let's give this a try" stage with Greg's Buttons. At the time I did not know what to expect, and released it as shareware just to see what would happen. As it turned out, Greg's Buttons was a small success. By the time we released Aaron, I knew it would be a hit, but I was completely overwhelmed by the response."

One item that will make any business shiver is the thought that it may one day be obsolete. Several years ago, shortly before the release of MacOS 8.5 and Appearance Themes, an Apple employee demonstrated a Scheme to Theme converter at an Apple Worldwide Developer Conference. This would have made it possible to use the existing Scheme tools to create Themes without the need for Kaleidoscope.

"The scheme converter was shown during the days of Kaleidoscope 1.x, before we released K2 with its new scheme format. Back then, schemes were much simpler, and converting them was not too difficult. The new scheme format is much more complex, and has features that cannot be directly translated into themes."

"When Arlo and I heard about the Scheme to Theme converter, we were upset. We had always expected that one day Apple would release a version of the Mac OS that would render Kaleidoscope obsolete. However, a converter would effectively usurp the whole Kaleidoscope community that we had worked so hard to build."

"At the time, Arlo made a comment that if Apple released a converter we would shut down the Kaleidoscope Scheme Archive. Many people mistakenly interpreted that as a threat. However, the real issue is that the KSA was intended to promote Kaleidoscope and was supported by Kaleidoscope registrations. If everyone was downloading schemes intending to convert them, then our Kaleidoscope income would dry up, and we would no longer be able to support the archive."

Fortunately, the point became moot when Apple predictably forbid the release of the converter; in fact, Apple has pulled back on the idea of Themes for the MacOS. They have never released an official theme, and in MacOS X, the ability is gone altogether.

Future plans for Kaleidoscope include enhanced menus and adding the transparency effects of Power Windows. Eventually, Greg would like to add mouse-over effects for buttons and window widgets as in MacOS X. But don’t expect a transparent Finder or document window. Greg says these are impossible under Classic Mac, but should be easy under MacOS X.

Since its introduction in the early nineties with the introduction of System 7, the Mac Operating System has been a moving target. It has successfully made the transition from the old 68k processor to the PowerPC chip, gone through two major version updates to get to MacOS 9, and now in the next and perhaps biggest transition, the move to MacOS X and the BSD/Darwin core.

"Right now I have mixed feelings about Mac OS X. I was not impressed by the Developer Previews; the interface felt cumbersome, and it wasted so much screen space. I just received the public beta in the mail, and from what friends have told me, it is much improved. Still, for someone who likes low level hacking Mac OS X is not an evolution of the classic Mac OS, but rather a whole new operating system. So, if I am to start writing software for a new OS, I need to decide whether that will be Mac OS X or some other promising OS such as Linux."

"I don't know much about Linux other than what Arlo has shown me at Eazel, but what I have seen looks fantastic."

"Mac OS X is simply too complicated for a novice to use. I know that Apple is working very hard to simplify it and hide the complexity. In the past, Apple has been very good at this, and I think they will probably pull it off."

Kaleidoscope, Power Windows, SmoothType, and Greg’s Browser - it reads like a “Who’s Who” of customizing software. No one knows what the future life of the Classic MacOS is, or where MacOS X will lead us. But if there is an aspect of the OS that can be made to look cooler, you can bet software written by Greg Landweber will have played a role in it.

Thanks to Tom Elliott for his help in preparing this interview.