Thanks for the correct advice, I don't expect to follow a linear curve. The 40 lbs took more than two years to put on, and I know that my linear curve is highly improbable.

However, as far as the 1,000 calories a day reduction in intake? That's not only possible, I found 1,000 calories to drop just because in New York, the law requires sellers of most food to post their calorie count. So, I took a look at my outrageous snacking habit:

Post-Commute Snack: A Jalepeno Cheese Bagel (420 Calories)
Office Snack: Cookies (300-500 Calories), Occasional can of soda (200-300 Calories)
Pre-Commute Snack: Box of Candy (200-300 Calories)
After Dinner Snack: Chips and Salsa (400-500 Calories)

Not to mention that in my office, we have unlimited free soda, candy and other snacks, as well as regular birthday cakes and people bring in home-baked banana bread, muffins and cookies all the time.

Fortunately, I really like string beans and baby cut carrots, as much as I like chips and salsa, so my snacking habit is satiated with that.

I don't expect a linear decline at all (I'm well aware that it's going to be a curve with flat spots, bumps and dips) but I also know that I need the overall path to generally conform to a downward trend. I do a lot of statistical analysis at work, and I've looked at other folks who have trended their own weight-loss and the success cases all have a similar shape, a steep drop, then a level off, increase in muscle mass and then a gradual taper to target.

The logging the food thing - I used a product called "Event Tracker" for the iPod Touch - it really works. I know calorie counting is not in fashion but for me that's working. I like the empirical nature of calories.