Skip to Content

Trash Your Desktop

Mitch Kapor’s new, more intuitive computer interface puts all the information we need to manage our digital lives at our fingertips, no matter what form it’s in.

“Thwump” sounds happen in boxing matches, not offices. So when a loud thwump bounces off the exposed-wood ceiling in an office in San Francisco’s once trendy south-of-Market district, every head turns. Programmer Jed Burgess is flat on his back next to a blue fitness ball. Burgess gets up, pulls his socks off for traction, and manages to balance atop the ball. Applause breaks out. Then the office returns to quiet discussions of software architecture punctuated by the clicking of keyboards. Welcome to Mitch Kapor’s Open Source Applications Foundation.

Kapor himself, famous as the founder of Lotus Development and one of the software industry’s chief malcontents, is away from the commotion. But if his foundation succeeds, it’ll make a thwump the entire software business will hear. The organization’s 13 programmers are hard at work on a piece of software that could change the way we manage our digital lives, curing the headaches of having to juggle the dozens of types of information stored on personal computers by a variety of applications-and, Kapor hopes, making computer users happier and more productive in the process.

Code-named Chandler, after the mystery writer (because, Kapor says, what they’re creating was something of a mystery even to them when the venture launched two years ago), the software promises to put all related e-mail messages, spreadsheets, appointment records, addresses, blog entries, word-processing documents, digital photos, and what-have-you in one place at one time: no more opening program after program looking for the items related to a specific topic. It takes the core functions of Microsoft Outlook, the Palm Desktop, and other personal information management programs and integrates them with the rest of your PC and the Internet. All the information you need to complete a given task or project is grouped on-screen, organized around the one function-e-mail-Kapor sees as the central conduit of our electronic lives.

A New Agenda

Kapor hopes Chandler will draw droves of converts but says he knows how fickle the software business can be. The seminal Lotus 1-2-3 spreadsheet was the must-have application that did more than any other to launch the personal-computer revolution. But Kapor, who founded Lotus in 1982, left the company five years later to lead On Technology, which had less success. He quit the software world altogether in 1990, when he cofounded the Electronic Frontier Foundation, a digital-civil-liberties group that filed, and won, some of the earliest cases involving privacy protection and free speech online. After that, Kapor became a philanthropist and investor, hitting it big with founding investments in Real Networks, the Seattle-based streaming-media giant, and UUNET (now part of MCI), which runs the largest privately owned chunk of the Internet’s backbone network.

Now he’s back to his first love, software design. He thinks most of the “productivity” programs available to workers and consumers today are too complicated and inflexible. Case in point: Kapor and his wife Freada Kapor Klein, who leads a small sexual-harassment consulting firm, wanted to use the calendar-sharing feature of Outlook to coordinate their schedules and those of their assistants. But to do so they had to install and administer Microsoft Exchange, a heavy-duty server program for corporate messaging and collaboration Kapor calls expensive and hard to maintain. “It’s total overkill and it’s horrible,” he says.

That experience was on Kapor’s mind as he considered reviving the ideas behind Agenda, a database and information organizer that was his Lotus swan song. Agenda automatically stored free-form database entries-such as “Call Alice on Friday about the Australia trip”-under multiple categories, such as phone calls, Alice, Australia, and Friday. It then recalled the entries at the appropriate times-for example, when the user reviewed Friday’s to-do list. Even though Agenda ran on Microsoft’s original DOS operating system, requiring users to learn many typed commands, devotees raved about the program’s ability to sort related data from disparate sources. But the program never sold well, and Lotus abandoned it after Kapor left.

Kapor thinks Agenda was merely ahead of its time. And because so many of the ideas that keep our lives and businesses humming are now shared over the Internet, he believes that any program that revives some of the principles behind Agenda should be, first and foremost, built around the Internet’s killer app: e-mail. Despite the Internet revolution and the tremendous amount of money and energy invested in creating software for it, the main interface to your computer-the desktop-looks much as it did the first time you used a computer that featured graphical icons, even if it was a Macintosh in 1984. But with Chandler, Kapor envisions e-mail as the main interface with computers, with entities like calendars, contact managers, instant messaging services, and file folders grouped around it.

“People spend enormous amounts of time in their e-mail; we’re totally e-mail-centric. It’s the hot ticket in productivity applications,” Kapor says.

Secret Agent Man

Chandler puts the user’s convenience above all else-which means the way e-mail and other entities are grouped is changeable, depending on the tasks at hand and how users want their information arranged. The foundation’s programmers are calling the groups of files “contexts,” since the point is to let users easily access related items, and to control what types of items appear.

The “to-do” screen, for example, could be a context, with e-mail mixed in with related task items. So if you’re planning a party, Chandler might put a calendar with key dates on it (when to pick up a cake, say), the invitation form, RSVPs, a task list, and even a budget on-screen at once. When a guest’s e-mail request for veggie hors d’oeuvres arrives, arranging for them would automatically be added to your to-do list. Contexts will mean Chandler can reorganize the screen every time the user shifts between projects, as if she were replacing her desk with a new one. That’s a far cry from today’s software, which forces people to dig through applications and file folders to find things, and to print them out if they want to see everything in one place. And while Chandler will offer preset contexts, Kapor expects other open-source programmers to build them, too. If someone develops a better way to run spreadsheet analyses, a user can simply pull out existing contexts and replace them. (Try that with Outlook.)

Driving some of Chandler’s flexibility will be a technology with a checkered past: software agents. These are small pieces of code typically designed to perform individual tasks, such as beeping when an e-mail message arrives. Attempts to build more sophisticated agents, such as Microsoft’s much-loathed “Clippy,” an animated paper clip that purports to help people use Microsoft Office programs, have faltered. That’s where Chandler volunteer Andy Hertzfeld comes in. The exuberant, boisterous programmer, whose Mac OS remains perhaps the most user-friendly program ever, thinks agents done well could reshape how people use software. “On the network, there’s a whole world that’s constantly churning out there,” he says. “So can we allow end users to express their desires automatically and then track them?” For instance, Hertzfeld asks, why shouldn’t your computer have an agent that will perform mundane tasks such as making hotel reservations when it finds a room for the right price, or update your address when it sees that a friend’s contact list is out of date?

Hertzfeld can’t resist plays on words-he’s working on “postal agents” for e-mail, “travel agents” for booking trips, and “secret agents” to handle software encryption. But his purpose is not to be whimsical. “Mitch is afraid I’ll make it too cutesy,” he says. “I have to make things Mitch-friendly.”

From Blue Sky to Nuts and Bolts

As Chandler’s lead designer, Kapor has fun dreaming up new things. But he’s also the boss. “You can’t coordinate a project of this scope in an ad hoc structure,” he says. In one of the programming team’s regular Thursday morning meetings, Kapor lays down the law. “We should aspire to a rare and unprecedented level of honesty” about schedule slips and bugs in code, he tells the team. Kapor’s voice is slightly hoarse, a vestige of his pre-software days as a DJ, and he has a balky back, so he’s often more comfortable in meetings when he stands. He commiserates with the staff about the difficulty of meeting targets, saying “things never take shorter, they always take longer.We’re not going to change human nature here. But let’s have a reality-based schedule, if we can bring ourselves to do it. We’re not VC-backed, so we have more of an opportunity to do things differently.”

The chance to work with Kapor on a groundbreaking product has attracted exceptional software people. Besides Montulli, a number of other former Netscapers are involved. John Anderson, who has to figure out how to build what Kapor dreams up, is one of Silicon Valley’s best contract programmers. E-mail architect Kaitlin Duck Sherwood has spent most of her life around computer messaging: her parents worked on PLATO, a seminal 1970s communications project at the University of Illinois. Hertzfeld seems to speak for them all when he says, “The purpose of coming in to work every day is to improve life for the user.”

But that won’t be easy. Though Kapor has put serious resources behind the foundation, Chandler is by no means a sure bet. The project has moved slowly since its kickoff in the summer of 2001, with more vision than code to its credit so far. Critics say Kapor is better at concepts than execution, noting the failure of several of his products after 1-2-3, including Agenda. And Kapor’s attempt to reorganize functions like e-mail, calendaring, and contact management has some labeling Chandler an “Outlook killer” and questioning the wisdom of taking on Microsoft. Kapor, however, says it would be “psychotically suicidal” to challenge Microsoft commercially, and he thinks it’s far fetched to talk about dislodging Outlook from its market dominance anytime soon. Still, since he does expect Chandler to have mass appeal, he says if that forces Microsoft to rethink its approach to applications, great.

Kapor says the skeptics are also missing the point when assessing the group’s progress. After all, it was only in August, once the idea of contexts coalesced, that Kapor declared the end of the “blue sky” phase of the project, directing programmers away from developing models and demos to actually writing code for the program’s major pieces. Kapor now believes that late 2004 is a realistic ship date for the first full version of Chandler.

And while some doubt, too, that open-source efforts can produce programs that rival commercial software, the spread of products like Linux suggests otherwise. Indeed, Douglass Wilson, chief technology officer for Lotus (now a division of IBM), says that open-source methods can build very high-quality software “by virtue of having lots of eyes.” He says the key is getting the right eyes. “What makes open source go is the community. If a technology spawns a community of interestthen you have both a very powerful creative force and a market force.”

It’s too early to say whether Chandler will develop such a community. Besides the technical challenges and the user issues, it’s hard to picture exactly what Chandler will look like, and until its basic framework exists, developers outside the foundation’s San Francisco office can’t write code for it.

So will Chandler succeed? Kapor pauses. “Uh, uhhh, yeah.” And suddenly Kapor the open-source advocate morphs into Kapor the CEO, and asserts, “It’s like any other startup. When you’re doing something new and different, there are always risks. But I’m increasingly confident that it’s going to work.” He chuckles. “But maybe you caught me on a good day.”

Keep Reading

Most Popular

Large language models can do jaw-dropping things. But nobody knows exactly why.

And that's a problem. Figuring it out is one of the biggest scientific puzzles of our time and a crucial step towards controlling more powerful future models.

OpenAI teases an amazing new generative video model called Sora

The firm is sharing Sora with a small group of safety testers but the rest of us will have to wait to learn more.

The problem with plug-in hybrids? Their drivers.

Plug-in hybrids are often sold as a transition to EVs, but new data from Europe shows we’re still underestimating the emissions they produce.

Google DeepMind’s new generative model makes Super Mario–like games from scratch

Genie learns how to control games by watching hours and hours of video. It could help train next-gen robots too.

Stay connected

Illustration by Rose Wong

Get the latest updates from
MIT Technology Review

Discover special offers, top stories, upcoming events, and more.

Thank you for submitting your email!

Explore more newsletters

It looks like something went wrong.

We’re having trouble saving your preferences. Try refreshing this page and updating them one more time. If you continue to get this message, reach out to us at with a list of newsletters you’d like to receive.