IN THE CHIPS

Embedded Systems Musings.

September 2007

Possibly figuring that the house is clean enough, iRobot has unveiled two new home robots that aren't robotic vacuum cleaners -- a gutter cleaning robot and, what seems to me, a mobile webcam robot.

The Looj Gutter Cleaning Robot is designed to simplify gutter cleaning. The Looj cleans an entire stretch of gutter from one location, reducing the number of times a ladder must be repositioned and climbed during gutter cleaning. The 2.25-inch high Looj drives under gutter straps propelled by a three-stage auger that dislodges and sweeps out dirt, leaves and other debris that can cause water damage, overspills, and ice dams.

On the other hand, the ConnectR Virtual Visiting Robot links to a home wireless network and provides VOIP-quality video and two-way audio communications with a remote person. The ConnectR has a top-mounted, full tilt-and-zoom video camera, speaker, microphone, and headlight to let users see, hear, and interact with people in the home as if there in person. It looks to be simple to operate via included ConnectR software, enabling remote users to "call" and drive the ConnectR around the home using a computer keyboard, mouse, or joystick. And like other iRobot home robots, the ConnectR has a charging base located in the home. When the ConnectR's battery runs low, the unit automatically returns to the home base to charge.

Posted by John Dorsey at 04:25 PM  Permalink |

What do Microsoft's Excel 2007 and I have in common? We're both math challenged, or so it seems.

In the case of Excel, Microsoft employee David Gainer reports in his blog that when users tried to get Excel 2007 to multiply some pairs of numbers and the result was 65,535, Excel would incorrectly display 100,000 as the answer. According to Gainer, Excel makes mistakes multiplying 77.1x850, 10.2x6425 and 20.4x3212.5. "Further testing showed a similar phenomenon with 65,536 as well," he said, adding that internally Excel calculates correctly but it is the display that messes stuff up. He went on to say that the problem is limited to six numbers from 65,534.99999999995 to 65,535, and six numbers from 65,535.99999999995 to 65,536.

In my case, I said in an editor's note entitled "2+2=What?" in today's Dr. Dobb's Report newsletter that "as when a 70% increase followed by a 60% decrease in statewide test scores is considered positive, even though the net effect on test scores is a 32% decrease." Well, if you actually compute that -- which I should have done before sending out the newsletter instead of after -- you get a 2% increase. As it turns out, the press release that I used as the source information had it backwards and, when I went back to the original document, it said "a 60% decrease followed by a 70% increase (resulting in a net decrease of 32%)". Thanks to readers Charles Linn and John Manis for quickly pointing this out.

Posted by Jon Erickson at 02:00 PM  Permalink |