Out of Control? Night Thoughts of the User Over Your Shoulder
Journalism Rule #1 is not to write a story about how you didn’t get the story. Yet here’s a review about how I couldn’t write the review. Of course I’ll tell you my opinion (don’t I always?); I think In Control 2.0, which I raved about thirteen months ago back in TidBITS-191, is better than the new version 3.0.
Can a new version of a great program, which loses none of its predecessor’s functionality, nevertheless be less great? I feel it can be less great if it imposes new features that impede its original excellence. If a program does everything it did before, and I don’t have to use the new features I don’t like, is it fair to call it worse? Am I conservative? Or sentimental?
So, this is not a review of In Control 3.0. I already reviewed In Control. If you didn’t read my review, download and read it; if it makes your mouth water, then buy In Control – it does everything it used to do, at a great price. The point of this article is to raise a philosophical problem: how does, and how should, software evolve? I don’t know the answer; I merely focus on In Control in order to illustrate the question.
In Control 2.0 combines an outliner and a database: the left "column" is an outline (topics and sub-topics), but other "columns" can contain keywords or other information in parallel to any outline topic, and you can then sort or match on information in any column, to see a restricted subset or reordering of the outline topics. It’s as though you had a lot of note cards that you can retrieve in interesting combinations, and which are themselves hierarchically arranged. Add superb import/export and printing, and a helpful interface, plus the ability to link any outline item to a file on disk, and you have an inspired (and inspiring) combination of simplicity, flexibility, elegance, and power.
In Control 3.0 (IC) adds vastly extended calendar and date-book facilities. The addition is clearly important to Attain, which used to bill IC as a "To-Do List Manager," but now calls it a "Planner and Organizer" for "Personal Information Management," and emphasizes the calendar capabilities in the manuals and example documents. Attain seems to think of IC as neither outliner nor database, but as a "time manager," a computer version of the looseleaf planners you read ads for in airline magazines.
Okay, I admit it: I’m a culture snob. IC 2.0 was a powerful tool for organising and navigating ideas and information, which is what I do, so naturally I think it worthwhile. "Time management" is something paper-pushing, white-collar, anal-retentive, MBA drones do, right? I know, I know: totally unfair. But you don’t deny, do you, that a Mac program can have an ethos? I can’t help it: I don’t like IC’s new ethos.
Besides, the new features are not mere additions: they’re intrusive. On my home computer, 2.0 takes just five seconds to start up, and fifteen seconds to open a largish existing document; 3.0 takes twenty seconds to start up, and twenty-five seconds to open an empty existing document. On my work computer (a Classic II), response to typing is so slow in 3.0 that I lose five of every fifteen or so characters, rendering the program useless. In 2.0, the bar of buttons across the top of the window can be completely dismissed, to see more actual text in the window; in 3.0 it can only partially be dismissed. In 2.0, new documents have only the basic outline column; in 3.0, new documents have three additional columns, and two of these ("start" and "end") cannot be deleted.
What’s more, the new calendar features, though souped-up, do not seem well thought through: I immediately banged against limitations. The three calendrical views – month(s), week(s), and day-book – interact pretty well with each other and the outline, but there’s no way to move directly from an outline item to the day-book for that day. The calendar day boxes are illegible on a small screen, because if there’s a time attached to an item, the text wraps only to the right of the time, not all the way back to the left edge of the box.
The calendar can be used to trigger "reminder" alerts, but the attention-getting version of these (a big splashy window) is available only if IC is running with the calendar file open. If it isn’t, all you get is a beep and a blinking Apple menu icon – unless you pre-set IC to launch at alert time, which works only if you have enough RAM free at that moment, and which (as mentioned above) takes twenty-five seconds (during which you can’t continue whatever you were doing).
The month and week views can have "banners" (text labels crossing several calendar days, to represent clumps of time like "vacation" or ongoing projects like "write chapter one"); but the day-book, though it lists banners that run across that day, doesn’t tell you what day of the banner it is or how many days it has left to run. You can combine outline and calendar into one view; if you double-click a day in the calendar, the outline is supposed to focus on everything for that day, but it totally ignores banners running through that day.
When I informed Attain of my views, I ended up in an email exchange with one of the program’s authors, Alan Albert, who was friendly and receptive. He argued in reply to this point that other leading calendar products have worse banners; and this may be. But although he doesn’t intend this point to justify IC’s shortcomings, I don’t see its relevance. The measure of a feature’s worth isn’t the implementation of the same feature in a rival product: it’s the ability of that feature to accommodate real-life usage.
And that’s the problem, isn’t it? When I first received In Control 2.0, the program told me half an hour out of the box that it was going to improve my life. When I first received 3.0, it told me half an hour out of the box that it wasn’t. I didn’t have to look for any shortcomings – they were obvious.
Alan Albert also pointed out that "from the Day view, you can display the Start and End columns to see what day the banner starts and ends, or include a Calendar in the Day view, to see at a glance both a single day or a longer period of time." The truth of this statement depends on the meaning of "you can." Perhaps he can; I don’t have the screen real estate. As I wrote him in reply, 2.0 felt to me like it was written by thinking people concerned with helping the user to help herself; 3.0 feels like it was written by people who had fast computers, large amounts of RAM, and enormous screens, and contempt for anyone who didn’t.
But that’s unfair too. The truth is the opposite: Attain does listen. In fact, that’s partly why 3.0 is as it is. Alan Albert commented: "We don’t see ourselves in the position of being able (or wanting) to dictate to our customers what features they ‘should’ have. Instead, we try to listen and respond. This accounts for the majority of the new features we’ve added to In Control. (We took this same approach when developing FileMaker, and it appears to be one that works.)"
But what does "works" mean? Probably, it means "sells." That’s what software developers are in the business of doing, after all. But, oh my friends and oh my foes, how I wish it weren’t! Of course I want software companies to listen to suggestions (especially mine!); but, steeped as I am in the ideals of Plato’s Socrates, I want them to decide what to do based not upon the wishes of the majority, but on considerations of what’s best. I don’t want my Mac to be full of lowest common denominators: I want it to be full of greatness. I want software developers to be wise, detached, superior, trustworthy, to aim at Quality (in a Plato/Pirsig sense). And I have a bad feeling that eventually the reality is going to let me down, every time.
Attain Corporation — 617/776-2711 — 617/776-1626 (fax)