One of the hottest topics this past week has been the formation of Mitch Kapor‘s OSAF and its big project, Chandler, a kind of souped-up do-everything PIM. The term that’s being bandied about at the moment is “Outlook on steroids” but, as the product page says, Outlook is not the right comparison model here. The feature summary looks like a standard-issue email client until you hit the bullet points at the bottom:
- structure data how you like it, view it that way, change your mind at any time
- automatic recognition of names, places, dates, and etc.; automatic categorization of items
Those features go way beyond what most PIMs offer today, yet they come from a 15-year-old DOS program. Chandler’s true daddy is the PIM that Kapor delivered to the market in the 80’s: Lotus Agenda.
Prompted by others’ reminisence on haddock this time last year (“I’ve been playing with Lotus Agenda via dosemu, and it’s fucking fantastic,” said Nick. “That program damn near got me organised,” said Danny) I did some snooping and Googling, determined to find out what it was that made the program so useful. (I didn’t have a PC to play with until ’94, so Agenda passed me by completely)
What does Agenda do? With it, you can weed through a mountain of information and arrange it into categories. Suppose you keep track of phone calls, are writing a proposal, and need to maintain a daily calendar. If you type “Call Bob and tell him to send the proposal notes by Friday,” Agenda is clever enough to read this sentence and assign this item to the categories: phone calls, proposal, even Friday. When you ask the computer what you have to do Friday, it will remind you to call Bob and insist that you ask him about the proposal.
The automatic recognition of facts (dates, people, places etc.) within free text is obviously incredibly useful and not all that hard, so it’s odd that few PIMs have done it since. It’s the magic icing on top of Agenda’s most prominent feature, in fact the core of its design philosophy: the flexibility of data formatting and categorisation. If you want to just enter stuff in free, flat text, you do that. Agenda will help you sort it.
Lotus Agenda is the only available database in the market that allows the keying of data to precede the creation of database tables. It may appear dull, since more thought has been given on its internal design than on its physical appearance, but Agenda is an excellent tool for sorting piles of information into meaningful categories. With this program, users can keep track of their activities, writings, research, notes, expenses and even other programs. Agenda can accurately read dates in practically any wording, from ‘next week from Friday’ to ‘6/30/93,’ and can create a separate category for items that have not been classified under any particular category.
Much heavier detail about the structure of Agenda is available in this document from Agenda’s creators. The most notable part is their list of key design requirements for the program:
- The user must be able to easily enter, edit, and manipulate free textual items without concern for the underlying structure of the database.
- The user must not be required to specify the structure of the data in advance and must be able to modify the database structure as it evolves without losing data or reorganizing the database.
- The user must be able to define reports in idiosyncratic formats. Through these reports, the user must be able to create and modify both database structure and content.
Views, finally, are presentations of the information in
your items, arrayed and selected according to the categories you
specify. This may sound similar to what a normal data-base
program does. With Paradox, dBase IV, RBase, and so on you can
retrieve pieces of information, through a “query,” according to
the criteria you choose. (“Show me the last name, first name, and
phone number for all families whose addresses have a zip code
from 10001 to 10292.”) The difference is that Agenda eliminates
the need for queries. In most data-base programs, there is one
bed-rock chunk of data, the mother lode, from which you request
samplings from time to time. In its fundamental technology,
Agenda also has one mother-lode of data, but – in ways that are,
again, easier to appreciate on the screen – it creates the
illusion that the information exists in small, pre-customized
chunks. You can create an Agenda view called “New York City,”
comparable to the zip-code query above. Whenever you flip there,
with one key, it can show you all the dealings you’ve had with
anyone in New York.
(I love his hyphenation of “data-base”. It feels so quaint and different to the way we write to-day.)
The use of user-defined categories was a key part of Agenda, but you could save the task of categorisation until after your data had been entered. Furthermore, the task was made much easier by being able to define remarkably-capable sets of rules and triggers; see the section on “Automatic Assignment and Implicit Actions” in the designers’ overview. Most importantly, a piece of information could be tagged with any number of categories.
So the way Agenda worked was to let you enter your data (contacts, appointments, notes, ideas) as freely as you liked, then slice and dice with views. However, it could give you a formalised interface to your information, depending on context: Agenda 2.0 came with Planner, a sample, customised view suited to appointments and to-do lists.
One negative point that I continually come across is the idiosyncracy of Agenda’s interface. While not being too hard to learn, it was still different enough to put most people off, people who preferred to stick to classic interfaces such as Lotus’s other PIM, Organiser. This was its downfall. Rowe:
Victor Cruz, spokesman for Lotus Development Corp., says Lotus stopped developing Agenda after selling only 100,000 copies. They thought Agenda was too difficult to learn, so they bought a no-brainer program called the Threads Organizer from a company in the United Kingdom. Threads looks like a notebook and a day calendar, so it is obvious what it does. Agenda is more subtle. Lotus has sold 450,000 copies of Threads.
Jimmy Guterman speculates that the program fell foul of the subjective suitability of most freeform idea managers:
It’s unlikely that all of the people who bought (or whose companies bought them) Agenda used it, or used it as suggested–not everyone’s mind works like Kapor’s. Anyone who has taken a single course in perception or neurobiology knows that every person’s brain interprets and organizes information differently. There are basic similarities (i.e., we all use the occipital lobe for visual information), but our neurons are as unique as our fingerprints. It’s easy to be skeptical when a company claims to have a program that “organizes your computer like your mind.” A recent PIM, “The Brain,” made such a claim, but it only worked like the developer’s brain and appears to have flopped in the marketplace.
The feature that appears to be most relevant to Agenda’s usefulness, and most lacking in today’s applications, is its use of views and categorisation to slice your information in as many ways that you need. From Guterman’s 1998 interview with Mitch Kapor:
But Kapor realizes that, as millennium approaches, none of the currently popular PIMs match the original vision Agenda. “Oh, we’ve had some evolution. PIMs have evolved a lot. They’ve gotten better at handling contacts and appointments. They’ve become very sophisticated. But the one thing that was the greatest thing about Agenda and why it still has a cadre of followers is the one thing that hasn’t been incorporated into PIMs: multifiling.”
“Today,” Kapor observes, “the PIMs are very Web-influenced, they have connectivity features and all, but they’re stuck in the old mindset. They’re focused on managing contacts and calendars. Agenda was all about managing ideas. Maybe that means Agenda isn’t really a PIM. But then again, the term ‘PIM’ was invented by Connell Ryan, Agenda’s marketing manager, at the time of the product’s first release. He invented that category name, but in retrospect the category didn’t describe what Agenda was.”
I’ve certainly been continually astonished by the lack of these relatively basic features in popular applications. The most obvious one is email: I have yet to find a decent personal-level email system which will let me file the same mail in more than one folder, or allow me to store and reuse views across my mailboxes. I certainly can’t get anywhere near Agenda’s rule and action capabilities without getting into my mail server and writing code. As my friend Manar Hussain said to me, years ago: “Your email is probably the most important database you have, so how come you can do so little with it?”
It also got me thinking about something we take for granted in the software world: continual feature evolution. We tend to think of software functionality as being on a linear good-bad scale. Good tech evolves and thrives, bad tech dies. Yet this is one case where some obviously good technology had to sit in the dustbin of history for many years before being revived; it’s lucky it’s being revived at all (and it still may not be, given Chandler’s current non-existence). If Apple hadn’t rescued NeXT from oblivion, what would have happened to brilliant ideas like Display Postscript?
It’ll be many months before Chandler is anywhere near useful, but I’ll be keeping a close eye on it. It sounds like this thing is easily the closest to my dream PIM, and anything that has the faintest hope of getting a shloch like me organised deserves the red carpet treatment.
(Coming up next: A take on the whole OSAF/Open Source anti-competition argument, and a brief overview of some of the other idea managers that have arrived since Agenda. I’d particularly welcome suggestions for the latter.)