I purchased AI: A Modern Approach by Norvig and Russell in march 2008, and by December I'd read a pathetic 80 pages due to work and general cant be botheredness. So I decided to choose a deadline that had some sort of symbolic significance. I would have to finish the book before the end of the 2008. Yes, that meant over 1000 pages of textbook material before the year was out.

I calculated it would take 40 pages a day; skimming was not allowed nor was moving ahead without a solid understanding of the material. I knew that the end of the book would be a cushion, since there were a number of pages on philosophical issues that were pretty familiar for me.

The result was successful and I learned a few things similar to what Yvain has posted. One of them is leave yourself outs, don't be too ambitious, because the damage of a failure is persistent beyond a particular case. If you don't have a good idea of how demanding the objective will be, err on the side of laziness, or do a smaller test run. Choose small short term objectives, because if you don't calibrate well and end up having to strain, it won't be for long and you'll get a better idea what you can do. In my case I went on 5 day trip and had to drag the book along and steal time to read, not good. But thanks to the experience I know that a 1000+ page text book per month is not realistic for me.