One Point Five Hours

The conversation went something like this:

Me: I’m bored, and I have a writer’s block – I have so many things to blog about and absolutely nothing to blog about.

Friend: You know what you should do? You should blog about work.

And here it is – my first work-related blog post. But first, let me treat you to an image, the result of preparing for this blog post:

1.5 hours in the Afternoom

What It Is

What you see is a combined image with a screenshot of my desk at work. The white lines are my mouse movements, and the white dots are where the mouse stopped. The larger the dot, the longer the mouse stopped there. I used a program called IOGraph to generate the lines, and then I overlaid the lines over a simple screenshot of my desktop. For obvious reasons, I only share my screenshot, instead of sharing what actually goes on – most of which are code and terminal windows. I had also unmounted several network drives which were sitting on my desktop – again, for obvious reasons

This image was taken in the afternoon – from 12pm to 1.30 pm today. For exactly 1.5 hours, those were my mouse movements. Contrast this with another 1.5 hours, but taken in the morning – from 10am to 11.30 am:

10 am to 11.30 am

1.5 hours in the Morning

Contrasting the two images, one can tell quite a bit about my mouse movements, and in this day and age, quite a bit about my working habits. As this is a two-monitor set up (yes, I get a very sweet rig at work), I will refer to the left half of the window (the one with the Launcher Bar * I hate the Ubuntu launcher, for the record ) Left Monitor, and the right half the Right Monitor.

Behind the Mouse Stop Points

So if you notice on both pictures, on the Left Monitor, my mouse position is often paused and clustered around a certain area – particularly near the edge of the galaxy in the wallpaper. This is especially pronounced in the Morning version. It wouldn’t take a genius to guess that that’s where all my coding is done. In fact, that is where my Eclipse window sits almost permanently.

The mouse pointers on the Right Monitor are a bit more scattered, but they too have independent clusters, and this is more pronounced in the Afternoon screenshot (the first picture) – where there are a number of mouse-stoppage at the middle bottom of the Right Monitor. Those are IM windows.  I often gain focus of the window by clicking the chat area, and they’re usually at the bottom of the screen.

On the Right Monitor also sits on average, 3-6 terminal windows. One terminal window is specially dedicated to R – I don’t use an IDE for R, since I mainly use R for exploratory purposes only, and the actual processing is done in Python * Although increasingly I have come to use more and more of R for actual simulation and modelling tasks .  Another terminal window is dedicated to MySQL-client. There are some other terminal windows for other stuff, like a interactive Python console, and a terminal solely for monitoring purposes (top/tail/grep – etc). Of course, on the Right Monitor also sits my web browser – Chromium. And to stop sounding like I am advertising for these products (I support and endorse Python and R the most), I shall recreate the story of my day today.

The Story

So, let’s try to recreate a story based on the two images and known information, shall we? We start the day at 10am. I am working on a particular code on Eclipse – as evidenced by the tight clustering of mouse stop points – this means I stop and click on the various part of the code a bit. I’d say, I’m working on either a long function/class or several short ones. I sporadically reply to my IMs, but I had probably alt-tabbed the IM windows out more often than I have it floating in sight. This also means that there is a higher probability that my web browser is on top more often – means also that I am reading my mail more often, since I use the webmail interface for my office emails.

Also do note that in the Morning image, the big dots are a lot bigger than the big dots in the Afternoon  image. This means I stopped my mouse for longer periods in the morning than in the afternoon. Many reasons could attribute to this – I talked more to people; or I went for morning coffee; or I simply concentrated more on what I was doing. Because this is a personal blog, there will be a bias – I didn’t talk to anyone all morning, except for the cursory “good morning”. I am on a paleo diet, and I’m trying to reduce my dependence on coffee – so I didn’t get any beverages. So a reasonable conclusion would be I was concentrating on a particularly tricky bit of coding.

And guess what? I was! I was having trouble with a recursive function I had written – this tweet should give you a clue as to what the problem was * I DO have a love-hate relationship with lambda , but by about 11 am, my problems were solved (hah, I work fast)

Then I went for lunch * My lunch was meatzza: a pizza whose base is made of meat .

After lunch, I came back to work. I had more IMs from coworkers and friends, and more questions and more interactions with humans after lunch – this can be seen by the increased amount of  mouse stops at the bottom of the Right Monitor. This also implies that my IM window is on top more often than my terminal or web browser (since if it is hidden from view, I just alt-tab it out). Also I moved some of my terminal windows to the left screen – as I was working on R. I started various data exploratory works, and this is evidenced by the stop points that are not in common with the Morning picture.

Evidence of more IM messages can be compared by looking at the Empathy/Unity icon at the top. Here is the one in the morning:

1.5 hours Early Morning Empathy

And here’s the one in the afternoon:

1.5 hours Afternoon Empathy

It is blatantly obvious that I accessed the Unity/Empathy message button more often in the afternoon than in the morning.

I certainly feel this way too – I felt I was more productive in the morning than I was in the afternoon – too much distraction after lunch, I have no idea why.

So, that was my day. How was yours?

comments powered by Disqus