"Once there I’d get my Chai tea, open up my laptop and set my alarm for one hour, I worked diligently for at least one hour each day, I could go longer but at a minimum I had to put in one hour of thesis work each day."
I like this rule. I think that having a whole day off doesn't make as much sense for knowledge workers as it did for factory or farm workers. I think working an eight hour day is extremely exhausting for a programmer, unless we're really enjoying what we're doing. On the other hand, problems tend to simmer in our heads on weekends and sometimes are forgotten by the time Monday rolls around. For me, a little work each day beats trying to work a full day five days a week.
Edit: Oh, and if I didn't have to commute 45 minutes each way, 8 hours might be easier. That's a whole other issue, though.
Yeah, I find it draining when working too long on the same thing and get distracted. I can (and do) work for long periods of time, but I feel it after and it makes me less efficient overall. I think working for many short bursts throughout the day lets me get a lot more done (and this is how I work on my own projects, eg during weekends: work for 1 to 4 hours (depending on the task), 1-3 hours off, 1-4 hours on etc and at the end of the day, I'll have racked up 5 or 6 hours of good, highly productive, efficient work. If I do 8 hours straight, I'm guessing only about 2 are actually efficient. When I was in uni, I often leave my assignments until the last day (BUT think about them before hand) and then sit down for maybe four blocks of 1.5 hours and get them done in 6 hours of work - compared to people trying to sit down for full days and spending a lot lot longer on them..
I agree - commuting means I'm tired before I even start. (Being a night person means I'm tired when I get up for the day job in the morning too, but thats another matter).
The full day off is important to me, but probably a lot less if I didn't get drained from doing 8 hour days.
PS: forgive my bad comment structure, its Friday, long and busy week, I'm pretty drained right now.. see above..
Historically I’ve approached big projects with the nonstrategy of “let them overwhelm you in your head, then put them off till moment when the fear of not finishing exceeds the fear of starting.”
This is well put, and exactly how I handled all of college. It worked OK there because project size was generally reasonably bounded.
The "fear of starting" is pretty insidious. A project often looks hugely different on either side of "started working" line -- it can be a really quick change from "I'm not really sure how this is going to work" to "I'm in a groove, this is going great!"
I had this exact conversation with a girl one time. She was the standard "work and work and work and work and work and you'll do well" kind of person, and I was more like you.
We were both in the lab, and she was working on finishing her assignment up for our class later in the day. I was reading Reddit and working on a side project at the time, and she asked how my assignment went. I replied that I wasn't yet finished, and she seemed horrified. I told her I had only about half an hour's worth of work yet, and I just simply didn't feel like doing it. My project was more interesting. She asked why I didn't just finish it and then I could "screw around" with the remainder of the time, because then I'd know I was done, and I couldn't really give her a better answer.
I started twenty minutes before it was due, and finished the work in ten.
That may be true, but that is not a strategy that would apply to writing a PhD thesis. That is just such a great amount of work, and with no real end-date.
Even then, the fear of not finishing on time will very likely come too late with such a big project.
I think the best advice in that case is just to think about how you eat an elefant: one bite at a time.
I've found that I only have about 2-3 hours of "genius" (for me) level work in me per day. That's when things are clear, and I see new connections (inspirations). If I try to push beyond that, I start to go backwards, and/or I am exhausted the next day.
Of course, there's lots of other work to do in a day that doesn't require "genius" level.
If you sit down at your specified time and are drawing a blank, get a new document/piece of paper and freewrite. Write ANYTHING, even if you just write “I have no idea what to say” over and over. Ignore spelling and grammer. Write stream of consciousness. Just get something on the page. Eventually something will click and you will start to come up with ideas that are more useful. But staring at a blank page is intimidating. You can always delete/throw out what you wrote later but you have to start somewhere.
I'm not sure there's a programming equivalent though. Playing on the REPL?
this is similar: when finishing work for the day, leave some small simple thing undone. then complete it at the beginning of the next day, to draw yourself back into the project.
I will often do this when I am drawing a blank. Write psuedo-code, or code that is how you would like it to be, if your language / macros / libraries supported it. Then you can worry later about making your language support what you just wrote. This eliminates a whole set of mental blocks to writing code.
I don't write anything longer than a page or two of script, but sometimes I'll walk away from the computer and scribe psuedocode onto a few sheets of paper, because I'm talented enough at becoming perplexed to be able to do it in that short a script.
I read Stephen King's book on writing earlier this year. There's similar advice in there as well (and this author references the fact that most writers work in a similar manner).
What I've found is that I've recently slipped into a habit of doing this same exact thing, unintentionally. Every night after dinner, I start up the laptop and shoot to get X, Y and Z done before going to sleep. (...on my programming project...)
I like this rule. I think that having a whole day off doesn't make as much sense for knowledge workers as it did for factory or farm workers. I think working an eight hour day is extremely exhausting for a programmer, unless we're really enjoying what we're doing. On the other hand, problems tend to simmer in our heads on weekends and sometimes are forgotten by the time Monday rolls around. For me, a little work each day beats trying to work a full day five days a week.
Edit: Oh, and if I didn't have to commute 45 minutes each way, 8 hours might be easier. That's a whole other issue, though.