...or; Matt's wrong & I'm irrational


Nearly a day without access to my feeds and a couple of days (mostly) without net connectivity have made me realize that I'm irrational. It also made me think of a post Simon Willison where he's quoting Matt Mullenweg, about unplugging from the net to be more efficient. Matt's wrong.

Spend 10 minutes collecting everything you need to work on a problem, and unplug the internet for 2 hours. You’ll finish in 30 minutes.

When the above quote came across in my feed reader a few days ago I smiled a bit, thought how true and imagined for a second what I could get done if I unplugged from the net a couple of hours each day. Then I continued to go through the seemingly unlimited number of unread items in Google Reader, a number made much higher than normal since I hadn't read any the previous day (due to the recent update to Reader rendering it unusable in Opera for a day).

Having caught up with the world I happily continued with my daily routine, coding away on what will become the next revision of SparkCMS. At one point I needed to check something in the PostgreSQL documentation, but couldn't reach the site. A quick check revealed that it was our internet connection that was down.

But before I continue, here's a mostly unrelated story about something that took place a few years ago. At the time I was working in a office that had fifteen other employees...

Cleaning day

One day it was decided that the carpet in our offices be replaced by a plastic floor surface of some kind, requiring that all furniture and office equipment be moved out of the way. Our manager determined that this was a good opportunity to do some general cleaning as well, and a day was set aside to clean cabinets, tidying drawers, sorting stuff in the archives and generally just throwing away old clutter.

Cleaning day came and within a couple of hours we'd probably sorted and/or thrown away hundreds of kilos of paper and/or trash. Then the entire neighborhood suffered a power failure. Except from our UPS powered server, everything went black. Now, as cleaning cabinets, drawers and archives really doesn't require any power (this was early summer so we had enough lighting from the windows and nobody really used any kind of electricity powered utilities), you'd probably expect that this actually increased the productivity many times. After all there would be no more interrupting phone calls or emails with tasks that simply had to be done right now. But surprisingly, everything actually stopped completely up. People were mostly drinking coffee (while it was hot), talking, some going outside to do personal errands. Some tried to find what happened with the power and when it would come back, others worried that they'd have to go all day without coffee. But as far as I can recollect - no one was doing any tidying. This (a very unproductive nothing work related) went on for a couple of hours until the power came back, at which time everybody returned to the house cleaning.

Years later I'm still amused and intrigued by this collective breakdown in reasoning. I always attribute it to the general irrational nature of humans, with an office workers added dependency on electricity to power everything from computers to paper shredders. Regardless of how irrational it may sound, it just was this simple; we were office workers in our offices, without electricity we simply were unable to work. Just as any other office worker elsewhere would be without electricity.

Disruptive downtime

In a very roundabout way this leads me to what I intended to write about in this post; my reaction to the internet connection going down. First I went through the normal steps identifying the problem and trying to get it working again, then realizing there's nothing I could do about it I called our ISP to have them fix it. At this point comes what, in hindsight, made the think about the above story. Nothing I was working on really required an active net connection, if I'd been totally depended on the Postgres reference I was about to look up when the connection first failed, I could easily have used the local copy of the documentation I keep for situations just like this. But instead of continuing coding on the CMS, doing completely other things such as system backups, general maintenance and other non-productive tasks such as reorganizing my media library.

The thing I find to be really scary about all of this, is that at no point did I consider if the non-functional net connection actually effected what I was working on. Nor did I consider how I could work differently to avoid being disrupted by it. As an internet worker without an internet connection, I just automatically stopped working on anything internet related. Everything with a similar, flawed and irrational logic as that found in the above story.

An excuse to make a graph

Then came Matt's post (or more correctly Simon's post with Matt's quote) which got me thinking about what I spend time on when connected and if it affects my productivity. More specific I wondered if my biggest "time waster", reading a couple of hundred feeds each day, had any correlation with my productivity.

The graph is pretty simple, green bars are the "items read" from Google Reader while the black line represents a (highly subjective) evaluation of my productivity (based both on hard facts as dev.logs and more unquantifiable stuff such the general impression I have of the work I've done in a day). These numbers doesn't really prove anything when compared, nor do they provide anything I can base a decisive conclusion on, I can however make a nice looking graph based on them. To compensate for the shortcomings and enhance the quality of the answer the graph gives, I've collected and plotted data from a series of 42 days!

a bar and line graph

There are four areas that significantly stands out. Although the fluctuations aren't necessarily fully explained by it, some specific stuff happened at the same time these are;

  1. Three days of travel where I tried not to work that much
  2. Google Reader upgraded, it stopped working using Opera
  3. T&B Hansen fucked up and our net connection was down
  4. A relaxing long weekend where I tried to take some time off

Although this cannot in any way be described as being a scientific study, and as such the value of all of this really being an enormous null, I consider it safe to say that my feed reading habits doesn't negatively effect my productivity. Though it's interesting to see that the "read items trend graph" rather accurately echoes the productivity graph. The scary parts still are the sudden (and again irrational) drops in productivity that occurred both when Google Reader was unusable and when our net connection was down.

However you decide to look at it, the answer still comes out as 42. Now, does that answer your question? (What was the question again?)

Couldn't generate page, however a cached copy was found and served on the 1st of January 1970 @ 00:00 in 0.031 seconds...

Errors occurred while generating this page... (More info availible)