As promised, I’ve digested another chapter of Scott Rosenberg’s Dreaming in Code, and will do my best to make my comments much briefer than in the past.

Before I discuss the chapter itself, let me emphasize again that Scott has posted an online version of the endnotes for the book, which you can find here. It contains hyperlink references to many of the books, newspaper articles, and reports that he mentions — sometimes just in passing, without any elaboration or discussion — throughout the book. There were some entries, like Vannevar Bush’s 1945 essay, “As We May Think,” that I thought might be too obscure for him to track down; I recall that it took a lot of work for me to track it down myself a few years ago, when I wanted to refer to it in a lecture. But Scott had this one, as well as publishing details and hyperlinks for virtually all of the other references, too. This adds enormously to the value of the book, in my opinion, though it does make one wish for an e-book version of Dreaming in Code — so that the text and the endnotes would be accessible within the same medium.

As for Chapter 2 of Dreaming in Code: it’s basically an explanation of what the Chandler development team was (and presumably still is) trying to accomplish. In short, they were trying to develop a Personal Information Manager (PIM) that would have as revolutionary an impact on people’s lives as the GUI-based Mac had on the lives of legions of miserable DOS users when it was first introduced 23 years ago. We don’t seem to use the “PIM” acronym very much these days, but for the Microsoft Windows users, we’re talking about Outlook. For Mac users, it’s a combination of iCal and Address, or NowUpToDate/NowContact (which I’ve been using since 1993, but which is soon to be replaced by NightHawk). No matter which one of these products you use, there’s a good chance that it organizes your schedule, maintains your to-do list, and keeps track of the names and addresses that we used to keep in a Rolodex or Filofax.

Aside from whatever complaints you may have about how these products accomplish these primary tasks, there’s something else that they generally do a miserable job of: storing, organizing, cross-referencing, and retrieving the hundreds — sometimes thousands — of unstructured, free-form notes, comments, references, bookmarks, video clips, photographic images, and so forth. Years ago, we Mac users were given a program called Stickies; I understand there are one or more Windows-based products of a similar nature. As a digital substitute for Post-It Notes, they’re okay; but it doesn’t take long to realize all the things that such programs don’t do, and it also doesn’t take long to become completely overwhelmed by the volume of notes that you’d really like to keep track of. I’m now using a program called SOHO Notes (formerly StickyBrain) to organize approximately 2,000 bits and scraps of information that I’ve collected over the years, as well as a program called NoteBook, from from a funky little software company called Circus Ponies. I’ve experimented with, investigated, and briefly used, roughly a dozen other PIM products over the past 20 years, and I’m always looking at new ones — such as the Mac-based GhostAction, which implements David Allen’s “Getting Things Done” method.

Ultimately, though, none of them are fully satisfactory — as evidenced by the fact that I’m using both SOHO Notes and NoteBook, since neither one is sufficient, alone, for the various things I need to store and organize and retrieve. So if the Chandler project manages to provide a really elegant solution to the free-form/unstructured PIM data management problem, it could be a tremendous success. Indeed, it might even disrupt the Outlook-dominated Microsoft Windows world, especially since (as I noted in an August 22, 2006 blog entry), there’s a whole generation of high-school and college students who have never used Microsoft Outlook, and who, when first introduced to it, complain that it doesn’t organize information in the way they really want to see it.

All of this, I suspect, you’ll find quite interesting as Rosenberg explains it … but to emphasize the difficulties in accomplishing such a task, Rosenberg finishes this chapter with a discussion of several prominent software-project failures that have been well-documented in recent years. If you’re unfamiliar with the FBI “Trilogy” project (which wasted $170 million of your taxpayer dollars), or the various unsuccessful attempts to upgrade and modernize the IRS computer systems, or the November 2004 crash of the entire UK pension system, or McDonald’s $170 million “Innovate” failure, or the FAA’s Advanced Automation System (AAS) fiasco, you’ll find some interesting reading here. And this is one area where the online, hyperlinked endnotes are particularly helpful; they can lead you to as much detailed documentation as you care to read about each of these cases.

By the way, an article in today’s New York Times (”Recasting the Word Processor for a Connected World,” by Michael Fitzgerald) quotes OSAF founder Mitch Kapor as saying that a version of Chandler will be released in April. If that turns out to be true, I assume there will be some publicity about it — and I’ll definitely be one of the first ones to try it out, in the hopes that it will be an improvement over all those other PIMs I’ve discarded, and the two PIMs I’m currently using.

And that’s all I have to say about Chapter 2. With any luck, I’ll find an hour in tomorrow’s busy schedule, and jot down some notes about Chapter 3. Stay tuned…

Reviews of earlier chapters of “Dreaming in Code”
Preface
Chapter 0: Software Time
Chapter 1: Doomed
Chapter 3: Prototypes and Python 

Endnotes