One of the reasons I left systems development several years ago was because of the necessary long hours of work you have to put in. There was a project I was involved in about 10 years ago where I lived in the office, literally. The programmers would come in around 9 or 10 and I'd be sleeping till noon. I'd wake up, have lunch, do some work, take a bath, change clothes in the work area, and work till dinner. After dinner, I'd still continue to do code, stopping at around 1 or 2 in the morning. After which I'd do some maintenance work, and go to bed around 4am. It was a six month stint. And I just felt burned out.
It has been more than ten years since I last stayed in the office for more than an overnight session. This time I started early Tuesday morning, and ended afternoon of Wednesday, spending 38 hours with less than 2 hours of sleep. I was luckier than some. Two guys left the office at 10pm Wednesday evening, one of them spending 38 hours and the other guy (my boss) 44 hours.
The day started inauspiciously enough. I got out of bed around 2:30, and got out of the house just after 3:00am. I went to Bayan to catch a taxi cab from there. Bought some breakfast at the 7-11 and took the cab with the attendant driver. (Most of the other cab drivers were sleeping.) We took C-5, and the driver was running at a sedate 60kph up to Libis, and sped a little bit after the Ortigas flyover, running at 70kph. He was safety conscious, not being tempted to drive at 100kph. The funny thing was that we almost met an accident. While approaching the foot of the Vargas flyover and approaching the bridge crossing the Pasig River, a bag of cement flew from the truck we were following. We were several car lengths away so the cab driver was able to avoid it. And he tried to flag the truck driver warning him that his cargo was flying off the truck.
I arrived at the office at 3:45am, while my boss had arrived earlier at 3:30am. The rest of the day I want to forget. It was all a blur, with the initial plans going out the window because of unexpected events delaying the debugging and testing. Almost the whole development team was thrown into the effort for deploying to the production (live) server. One of the team members had to go home just before midnight Tuesday. Some of the crew were sent home at around 2:00am, Wednesday. Four of us stayed for the clean-up and actual deployment. The programmer who went home early assisted as soon as he got to the office. At around 3:00pm, I wasn't sure if I could be of any more assistance, so I decided to go home.
During the whole time I was able to get some naps of about 30 minutes each. Twice I slept on the couch, once while sitting on my desk (didn't bother to lie on the couch), and in the morning just after breakfast I lay on the floor of my boss's office for another 45 minutes. When I got home, I rested a bit, then me and my family ate out. I slept like a log and couldn't get myself out of bed the following day, coming in late on Thursday.
NOTE:
After several years of being outside of the IT industry, my current job takes me to the cutting edge of systems development: programming for the internet. The needs follow internet time: it is now, now, now. Unlike traditional business environments, where the project needed to be done yesterday (hence the requirements are outdated once approved), internet development is for the here and now. There is an immediate need which did not exist yesterday. It has to be done today. The requirements change on a day-to-day basis, with additional functionality being delivered for each iteration. An iteration lasts about a week or two and the deliverable for each iteration is a working program. In this sense, development is evolutionary. Functionality is added as the system progresses. At any point in time, the delivered iteration is a working program.
--andoy
6 May 2005
No comments:
Post a Comment