Skip to Content
Uncategorized

A Self-Writing To-Do List

New online schedulers rely on natural-language processing to get you organized.
June 11, 2008

The problem with to-do lists and schedules is that you need to fill them out. Now, a new generation of free online schedulers promises to end that drudgery. These new Web applications use natural-language processing to interpret spoken commands and ordinary written sentences to build calendars and personal organizers.

Perhaps the simplest of the new generation of schedulers is Presdo, based in San Francisco, which launched in late April to help users collaborate to schedule meetings and other events. Borrowing from Google’s successful bag of tricks, Presdo’s home page is as simple as it gets: just a floating text box. Type in “have brunch with Margaret on Sunday,” and Presdo translates your command into data, bringing you to a page where you and your guests can check and tweak the details of your event.

By taking its cues from the ways that people naturally talk about time, the software frees users to be general about dates and times, says Presdo founder Eric Ly. Imprecise phrases like “next month,” which would be impossible to put on a calendar without picking a particular date and time, are allowed to stay fluid for as long as the user wants them to.

“There’s no widget in our system that looks anything like a calendar, and that was intentional,” says Ly. “We really wanted to make it very easy for people to express what they wanted in terms of time. We felt like the natural-language approach was going to be more flexible and expressive for users.” If you sign up as a regular user, Presdo will gather more information to help it guess automatically. For example, it will suggest restaurants near where you live via Google Maps, or it will remember Margaret’s e-mail address from your last event together.

But translating the vagaries of ordinary speech into data that a computer can understand is a tough technical problem. “One thing this made me acutely aware of is how weirdly people speak,” says Rael Dornfest, developer of IWantSandy, an online personal-assistant program based in Portland, Oregon, that uses simple text-based interactions to generate calendar items, to-do lists, and reminders. “There are little things that are sort of classic. When I say ‘next week,’ do I mean the week upcoming or the week after that? The problem is not about parsing. It’s that if you said it to 15 people, half would interpret it one way, and half the other way.”

Sandy–named after free-software advocate Tim O’Reilly’s real-life personal assistant–can intelligently read e-mails, text messages, and Twitter feeds. Dornfest calls Sandy’s algorithm “natural-language-ish processing”: it’s basically English, with a few keywords to help Sandy recognize common tasks. Telling her to “remind” or “remember” something generates an automatic e-mail or text-message reminder; adding “@todo” to your message places it on your to-do list.

By using ubiquitous communication tools like e-mail and text messaging to interact with Sandy, says Dornfest, users can get organized without stopping to think too hard about it. “A lot of the things Sandy takes down would never have made it into a calendar in your lifetime–it’s just too painful,” he says. “Most organizational systems break your flow. They try to make you do something else for a moment, and then you can go back to whatever you were doing in the first place.”

Another new program, reQall–developed by QTech, based in Hyderabad, India–pushes that idea even further by giving users a toll-free number they can call and leave messages at. Whatever your favorite communication medium–e-mail, Web, text messaging, or phone–odds are that reQall can parse it. Voice-recognition software, live human transcriptionists, and natural-language processing algorithms read your messages and use them to generate reminders that can be delivered by e-mail, text messages, or voice calls, customized for the user.

“If I say, ‘Remember to buy a watermelon tomorrow,’ I won’t see it today,” says QTech founder Sunil Vemuri, who got the idea for the program while a PhD student researching memory at MIT’s Media Lab. “The system will interpret the sentence and put it in the right place. It removes some of the cognitive burdens of trying to get the idea out and organize it.”

Neither Presdo, IWantSandy, nor reQall has an obvious business model. Their creators are contemplating charging fees for premium accounts in the future, but for now, all three applications are free of charge.

The sudden popularity of organizers that are just a text message away may be part of a larger trend. For two decades, software has been dominated by graphical user interfaces, which employ visual features like windows and icons to convey information. But clearly, Google isn’t the only company that’s banking on text entry. The command line is making a comeback–and increasingly, natural-language processing is bringing the ease and simplicity of text-based computing to the non-tech-savvy.

“There are going to be more and more applications which are less monolithic screens, and more dashing off quick missives,” says Dornfest. “We’ve just begun to scratch the surface here.”

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.

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.

How scientists traced a mysterious covid case back to six toilets

When wastewater surveillance turns into a hunt for a single infected individual, the ethics get tricky.

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 customer-service@technologyreview.com with a list of newsletters you’d like to receive.