Chandler: What went wrong

Lotus 1-2-3 developer tries to tackle modern productivity, talks about where he failed.

Bob Walsh

Bob Walsh is the co-moderator of the the popular Joel on Software Business of Software forum and a consultant to startups and microISVs. He writes a blog at 47hats.com, and is the author of two books, Micro-ISV: From Vision to Reality and Clear Blogging: How People Blogging Are Changing the World and How You Can Join Them.

Bob Walsh
4 min read

Six years in the making, the 1.0 version of Open Source Web and desktop info organizer Chandler finally arrived on Friday. It was not met with thunderous acclaim, nor did it get the kind of press its development cost of $8 million and tens of thousands of volunteer hours was supposed to generate.

Chandler consists of the Web-based Chandler Hub, and desktop clients for Windows, Mac and Linux. It is meant to be your everywhere digital notebook for organizing appointments, tasks and notes. Chandler lets you import and export calendars to just about any iCal-compliant application, such as Google Calendar and Apple iCal; it creates alarms and reminders; and it provides simple task management.

I can remember when Mitch Kapor--the founder of Lotus Development and Lotus 1-2-3, the man who gave reason to put a PC in every office on the planet--started this project. It was going to free the masses from domination by Microsoft Outlook and Exchange Server and triumphantly herald a new age of open-standard killer software.

Chandler is a capable to-do list manager, but its design and capabilities are dated.

Despite some innovations in brings to the category, it has not delivered on the promise. Here's where it fails:

    • Whether you're working in the Chandler web app or using a desktop client, you enter to dos, appointments and notes in a single entry field, then add detail. But the devil is in these details, and you will get tired very quickly of the slow interface for entering calendar items.

    • While Chandler lets you triage your items into Now, Later and Done buckets, this is rudimentary task management at best and will leave practitioners of David Allen's Getting Things Done wondering where the rest of the application is.

    • Chandler's big selling point--seamless Web app/desktop synchronization and backup--works as advertised (although it did not work for me out of the box), but a competitor, the free but commercial Evernote, offers superior features, like picture and sound recording, and Web-based OCR of images stored.
    • Chandler organizes your information into collections, but unlike say Google shared calendars, there's no way on the Chandler Hub to discover collections others want to publicize.

With any 1.0 app you can expect rough edges and unimplemented features: neither Wikipedia or Mozilla Firefox were much to talk about in their early days. The question about this open source project is whether Chandler 1.0 is the start of something great or the last gasp of a party no one wants to be at.

I talked with Kapor, who underwrote $5 million of the nearly $8 million that went into Project Chandler, He said, "It's been a long, long journey. So long that a whole book was written about a part of it. It's obvious to anyone who is familiar with the story and the history that it is one of the project that, to make an understatement, did not turn out the way it was originally planned."

"If you view it from the point of view as something that would replace Outlook and Exchange, it has completely and utterly failed."
--Mitch Kapor

Kapor pulled the plug on the free money and office space for the Open Source Application Foundation back in January. "Last January I reached the conclusion that I was really ready and needing to go on to other things. And the team on the project had a really strong desire to see it through to competition. The team shrank in size pretty dramatically then. I put a little more money into it in order to enable this transition to happen in 2008, part of which was they were going to ship 1.0," said Kapor.

"You asked if Chandler has lived up to the dream I had six years ago. And I think the fair answer to that is in part yes and in part no. What has actually been delivered delivers on part of what the original dream was. There's a cross-platform, fully open-source, innovative personal information manager with a very strong calendar. Those were among the original goals. Obviously it's taken dramatically longer than I thought it would, which I would attribute to not good judgment about how long it would take or the complexity of the project."

Now Project Chandler is on its own. "It's more of a conventional open source project in the sense that its momentum hereon is going to depend on the extent there is a community of volunteers who find it valuable enough to contribute to it and move it forward." Are there enough volunteers to do that? "The short answer is I don't know, but there are promising signs. The numbers of people involved, while modest, are nontrivial and growing. If you view it from the point of view as something that would replace Outlook and Exchange, it has completely and utterly failed. But from the point of view of having built something that tens of thousands of people are happily using, and were using before there was a 1.0, by those metrics, it's pretty promising."

And any advice to the new generation of Internet wannabe millionaires running around the valley? "Well yeah, actually. It's easy when you've been successful to lose calibration about what you can accomplish, about how hard it can be, about how long it will take. There's lots of people on the second time around that have dug themselves into one or another kind of hole. Part of what allowed Chandler to get traction and to get to 1.0 was when people on the project, myself included, where able to have more modest objectives, to have more realistic planning and to get into a more agile cycle of development, on the principle it is far better to deliver something than it is to have huge dreams and deliver nothing. That would be my advice."

Disclosure: Bob Walsh occasionally sells a copy of a Windows desktop task manager he wrote three years ago.