Live from Austin!

Rothko runs a coffee shop. Each week she needs to calculate how many hours each employee worked, given their clock-in and clock-out times. It turns out that this is a surprisingly painful task to perform using regular tools like calculators or spreadsheets. I built a Mac dashboard widget to make it easier . I’m putting it up for download because I can imagine other people might find it useful. (For example, anyone who has to fill out paper timesheets.)

To use it, enter times in the "start" and "end" fields for each day. You can use 12 or 24 hour clock. For 12 hour clock, if there’s an "a" in the field it’s a.m., if there’s a "p" it’s p.m. You only need to type the colon and number of minutes if there’s an actual number of minutes involved. (It’s all about reducing the number of keystrokes.)

Tab takes you on to the next field as usual, and automatically recalculates everything like a spreadsheet. The individual durations worked for each day are shown, and totaled into a final figure in hours and minutes at the bottom.

So for a standard work day, you’d type 9a [tab] 5p [tab] and it would enter 9:00 am start, 5:00 pm finish, calculate the duration as 8:00, and add it to the total.

The only other control is the Reset button, which blanks out the fields ready for the next calculation. Quick, simple, efficient, and hopefully correct…

Being laidback, easygoing types, the spouse and I often end up having long and tedious conversations about where to go eat, along the lines of:

“Are you hungry?”

“I guess so.”

“Let’s go out.”

“OK.”

“Any ideas?”

“Umm… don’t mind, really.”

“What kind of food do you feel like?”

“I don’t really have any strong preference.”

I used to have similar problems with Richard in Cambridge, and came up with an idea I called Oblique Restaurants.

It was basically a variation on Brian Eno and Peter Schmidt’s Oblique Strategies: a deck of cards with names and locations of local restaurants. I had the idea that notches in the cards could be used to allow you to filter out (for example) all vegetarian-suitable restaurants, or all Italian restaurants.

Then, when faced with a lack of restaurant-related inspiration, you could shuffle the deck and pick a card—but you would then be obligated to go to the restaurant on the card.

A while ago someone implemented Oblique Strategies as a widget for OS X. And now someone else has implemented Oblique Restaurants, except they call it Dine-O-Matic.

Now we just need Oblique Recipes for quick stuff to cook at home.

Regarding the Apple “copying Konfabulator ” controversy:

The idea of desktop widgets that pop up on the screen when you press a key is far from new. In 1984, Borland launched Sidekick 1.0 for MS-DOS. When you pressed a magic keystroke, a calculator, notepad and calendar would appear overlaid on your screen. Push the key again, and they vanished.

Also in 1984, Apple’s original Macintosh operating system had Desk Accessories, small tools which weren’t full applications, but which could be launched on top of your application windows.

It’s pretty clear to me that if you take Sidekick plus Desk Accessories, and use JavaScript to write the code, you have both Konfabulator and Tiger’s Dashboard. Sure, Konfabulator was a worthy innovation over the state of the art circa 1984, in that it made it easier for people to write desk accessories by using a scripting language instead of a compiler. However, Dashboard is a worthy innovation over Konfabulator.

I liked the idea of Konfabulator, but I didn’t want the widgets to get buried under my windows, or to take up screen real estate all the time. I had started designing my own desktop widgets system, in fact, which would appear and disappear somewhat like Dashboard. I’m sure that if I had started serious coding or had released my desktop widget system, I’d be pretty annoyed to see Apple release the same thing. As it is, I’m glad they’ve saved me the effort, though personally I wouldn’t have chosen JavaScript for the scripting engine…

This is the way the software industry has always been—you have to keep innovating, or someone else will do what you’ve done, and do it better. If the Konfabulator guys want to innovate and compete and try to jump ahead of Apple again, I have some ideas I’d be happy to share with them for free if they contact me. (I’d much rather pay them $25 than write and support the code myself.)