What's more, I suck for wasting my time on that time waster. So no more of that. I have too much I need to do this week and in the rest of my life to spend another iota on his shit. If he's too wrapped up in himself to relate to me I need to move on to more productive endeavors.
Like my stupid ass recommender system paper. That thing's been dragging along like a poor unfortunate animal caught in my life's wheel well. I need to come to a full stop and put it out of its misery once and for all. I have six days in which to do this. I need a plan simple and sensible enough to bring me through the next six days with a paper to turn in at the other end.
How about this: I know I work better in the mornings so I'll reserve two hours every morning this workweek from 8 to 10am. Then I have to come into work, but if nothing is going on I can write at my desk. I cannot, however, read at my desk. Not the way I like to read, with hi-lighted papers sprawled all over the place. So the reading I'll have to do at home or on screen. I'll need a way to take notes, both here and at home. I kind of suck at that, so maybe a stack of cards is the best I'm going to do here. Hopefully it won't get too far out of control and I'll be able to use them as visual tools.
So that gets me to lunch. I guess depending on how I'm doing it might be best to eat lunch at my desk to keep working; or maybe I'll need a break so as not to burn out. I'll have to see how that goes. Methylphenidate will have to be at the beginning of each day of this plan. So let's say I get another couple hours in during the afternoon. I can go home and nap or chill for an hour and do another hour or two after dinner.
Add it all up:
(3 days x)
2 hours in the morning
1 hour at work before/during lunch
2 hours after lunch
2 hours after dinner
-2 hours Thursday for Dr Feldman
2 weekend days (~4 hours each)
19 + 8 = 27 hours
Let's say I can divide the paper into two parts:
1. recommenders overview
2. Jinni
However I slice it, recommenders is going to be shorter than Jinni, so I'll give it 9 hours. That means reading, parsing, writing, editing all have to happen within those 9 hours. That leaves 18 hours for the Jinni part, during which I'll have to figure out what algorithms make the most sense for it, and justify my choices. This means the bulk of the reading. I'll have to sift through all the literature I can get my hands on here. When I take into account how much harder it is to write when references have to support everything I think maybe the breakdown should be:
3 hours: recommender systems reading
3 hours: recommender systems writing
3 hours: recommender systems referencing
6 hours: Jinni reading
6 hours: Jinni writing
6 hours: Jinni referencing
How best to break this up over the next 6 days? Since I've already written a general overview of recommender systems and can basically polish that to the end of time and still not be very happy I'm going to say forget it for now. I need to focus on the meat of the matter: how Jinni works. It is, after all, what I've been avoiding dealing with all this time.
So reading on Jinni is the first task. I'll need all my clarity for it too, so I'll allocate the mornings to figuring this out. If it becomes too confusing I can switch to recommender stuff as a break. When do I absolutely have to be done with Jinni reading? I'm going to say by the weekend I have to have it under control. I have to have at least the movie genome figured out good and solid. I'm thinking this time around the goal is to have as clear an understanding as possible before starting to write. The other way certainly didn't work.
After this weekend I can reflect on my social and romantic lives and start worrying about Xmas shopping. For now, I'm putting on the blinders.
