What makes some software “interesting,” as opposed to merely usable? For one, software seems interesting when it allows you to see or consider information in new ways. With utilities like DevonThink Professional, for the Macintosh, or dtSearch, for PCs, you can run “semantic searches” for e-mails or files on your own computer and find passages whose themes are related even if they don’t share specific key words. (When you are doing key-word searches of files on your own PC, nothing beats the old standby X1, which costs $74.95 and up from X1.com or comes in a slightly different free version from Yahoo, at desktop.yahoo.com.)
Interesting software also tends to be extendable, or customizable. Personalized touches can be as simple as refining a set of bookmarks for a Web browser, or as complex as assigning long command sequences to a few keystrokes with the very handy PC program ActiveWords.
And, as I am shown by previews of two extremely interesting new programs, software can even be intriguing on an aesthetic level, by virtue of the creative choices that go into its design.
The two programs I have in mind are OneNote, from Microsoft, and Chandler, from the Open Source Applications Foundation, in San Francisco. In all the conspicuous ways, the programs are each other’s opposites. One is from the world’s biggest and most successful software company; the other, from a shoestring operation funded mainly by one man—Mitchell Kapor, the founder of Lotus. OneNote sells for about $100 on its own and is offered, like Word and Outlook, as part of the “suite” of Office products that is generally regarded as Microsoft’s cash cow. Chandler will be free, as will several related utilities and programming tools now being developed by Kapor’s organization. OneNote runs only on PCs or tablet computers with Windows software. Chandler is being designed to run on PCs, Macs, Linux machines, and about any other plausible system. The first release of OneNote went on sale three years ago, and a significantly improved version, OneNote 2007, will be available late this year. Chandler has been in the works since 2001, but won’t be ready for mainstream use for at least another year or two. At the moment, it exists in a limited version that mainly serves a “dog food” function—that is, its own designers use it for their daily tasks as a forced exposure to its strengths and limitations, a practice known as “eating your own dog food.”
But the programs share a fundamental goal, which is to keep lowering the barrier between the way computers work and the way people naturally think. The tech world has come miles in this direction since the days when you had to learn obscure commands to make a computer do anything at all. The barrier OneNote aims to surmount is one created by Microsoft’s own success in establishing Word and Outlook, plus the overall Windows file system, as the dominant standards for writing, calendar keeping, and workplace communication.
By Microsoft logic, specific programs are “right” for specific purposes: Word if you are writing something down, Outlook’s Tasks list if you have a to-do item. Chris Pratley, OneNote’s chief designer, saw it differently. “The key insight I had,” he wrote in 2004 in his OneNote Blog, “ … was that [the program] had to let you capture the thought or piece of info as you had it, without forcing you to deal with any software goo up front.” It is a sign of the charming breeziness of his blog that he refers to his company’s mainstay offerings as “goo.” (Disclosure: I worked on the Word design team at Microsoft in 1999, and when I did, Pratley was a friend and supervisor.)
“Properly” capturing and recording in the Office system the information coming at you all day long, Pratley wrote, can be more trouble than it’s worth. For a new phone number, you have to switch to Outlook’s Contacts view and fill out a form. If you jot down something in Word, you have to create and name a new file to store it, eventually littering your hard drive with zillions of confusingly named files.
The response to this problem, via OneNote, is a kind of all-purpose magic clipboard, or “notebook.” It runs while you are doing other work, and whenever you want you can pull into it information from another program. It will store whole Web pages or selected passages; Web links; e-mail messages; text or graphic files, either embedded like attached files in an e-mail or with their contents shown; video or audio clips; handwritten entries or hand-marked documents from tablet computers; and other forms of data I’m forgetting now. You can sort or name this information as it comes in, if you want—with special notebooks for phone numbers, research about Ecuador, meeting notes, etc.—or you can just dump it all in one place and find it later on. You do this subsequent finding in a variety of ways: through labels, tags, or other identifying information you have supplied yourself; through notebooks or folders you have created for special topics; or through the new indexed search feature in OneNote 2007, which retrieves relevant entries almost as soon as you type what you’re looking for.
This new release is integrated with other Office programs in a way that actually helps the user, as opposed to just advancing the corporate brand. You can write down an errand to do or person to call—and, with a keystroke or two, convert it to an item on your calendar or to-do list. You don’t have to remember to save the file you’re working on—or worry about what name you gave it or what directory it’s stored in, since everything you add to OneNote is saved continuously and is retrievable through the very effective indexed search or through the labels you applied. You can easily delete clippings or notes you decide you don’t want anymore. I’ve started dumping all my interview transcripts into this program, rather than saving each one as a separate Word file. In OneNote, I can have a single page show my typed-up notes of an interview, plus Web links about the person I met, plus the audio file I made on a digital recorder during the meeting—which I can click on if I want to review some passage. A feature in OneNote’s new release addresses a woe of today’s mobile workforce, by automatically synchronizing the notes that are kept on your laptop and those on your desktop machine when they are networked to each other. This auto-sync feature reflects an assumption that users should not need to keep track of the program’s underlying file structure, which is different from the standard Windows structure.
There is a lot more to this program, which you can see demonstrated in a seventeen-minute blogcast by Darren Strange, the U.K. product manager, at tinyurl.com/mjwcd, or read about, along with questions of software aesthetics, in Pratley’s blog, at tinyurl.com/k5k78.
Chandler has the potential to be at least as compelling, but for now it’s mainly just potential. Five years ago, Mitchell Kapor announced that he was committing $5 million of his own money toward development of an open-source “personal information manager.” This was a software category Kapor had practically invented at Lotus in the 1980s with his program Agenda, which I raved about in these pages in 1992, just before Kapor’s successor, Jim Manzi, decided to kill it. Agenda was special to me and 100,000 or so other devotees because it let you enter information in one central place and then retrieve or structure it as appropriate. It was not special to Manzi: 100,000 paid users was not a mass market. Or, as Mitchell Kapor put it to me, there was little commercial demand from users for a sophisticated information manager because of “a lack of imagination that things could be better—there was no shining beacon of America across the seas to make them realize they didn’t have to be oppressed and imprisoned” by their mediocre software.
Kapor’s attempt to create that beacon has gone more slowly than he foresaw. Despite substantial follow-up grants from foundations and universities, the team developing Chandler has so far released only a partly functional calendar application. Scott Rosenberg, of Salon magazine, became an “embedded journalist” on the Chandler project from 2003 to 2005 in order to investigate why good software is so hard to make. (His book about Chandler and complex software design, Dreaming in Code, will be published in November.) “It is taking a long time, but anyone who writes off Chandler is being short-sighted,” he told me. They are on a quest.”
So what is worth waiting for from Chandler? The easiest way to illustrate its concept is through its plan for handling e-mail. From a computer’s perspective, all e-mail messages are similar to one another—and different from all other digital items, like Web pages or calendar entries or to-dos on a task list. What the computer can detect is the form and structure of the digital item: e-mails have headers and come in from remote senders, calendar appointments are entered in the calendar program, and so on. Thus Outlook and the few comparable programs store e-mail in one big in-box, with subsequent crude classifications—read versus unread, flagged versus unflagged, those assigned to special folders versus others.
For the human being sitting at the computer, however, e-mail messages are not all the same. “One e-mail could be an invitation, so it’s actually a calendar item,” Mimi Yin, a Chandler lead designer, told me. “A series of five e‑mails actually should be one short chat session. Another one is reference content, one is a funny URL, and another is a specific to-do assignment. ‘Flagging’ the e-mail just tells you to look at it again, but you can’t add semantics to say why.”
Rather than letting flagged or unread e-mail pile up, as it often does in Outlook or other mail systems, waiting for you to figure out what you meant to do with it, the completed version of Chandler will funnel nearly all incoming information to its “dashboard.” This will let you easily process incoming data based on its real content—e-mails requesting a response or appointment will look different from those just containing reference material—and then will present the data to you at the times you’ve asked to be reminded of it. Another reason I’m a sucker for Chandler’s approach is that it draws inspiration from the productivity expert David Allen’s “Getting Things Done” strategy, previously described in these pages.
Exactly how the dashboard and the rest of Chandler will work, when they do work, is laid out in an engrossing series of postings by Yin and her colleagues on their Web site. Tinyurl.com/sxdk5 is a good introduction, as is tinyurl.com/n3esu; each has links to many more. These programs are interesting, and so are the minds behind them.