Manage Pixels, not Monitors

Multiple monitors are often significantly cheaper than one large one. You can also aim the two at you more accurately, so you don't have to change what distance you focus at as much. The productivity study was short-term; how about long-term? Certainly having things too wide means you spend more time mousing to those edges, but how about eye strain due to viewing angle?

>That optimal number, for the vast majority of people is about 2500x1400.

Programmers are most definitely not the vast majority of people, when your study covers "people" as a whole. Programmers handle huge amounts of data and tend to have documentation and/or testing at-hand frequently, which takes up what would otherwise be usable space. I also don't see any info on how many people were used for the study, though I've only seen the slides, and have to get going...

I have a problem with his conclusion after looking at the methodology slide:

This study compared 26-Inch MultiSync widescreen (MS LCD2690 at 1920x1200 resolution), 24-inch MultiSync widescreen (MS LCD2470WNX at 1920x1200 resolution) and 20-inch MultiSync traditional format (MS LCD2070NX at 1600x1200 resolution) monitors.

There was no 30" monitor involved in either of the studies, nor was there any monitor with a resolution above 1920x1200. The slide with a graph showing a drop-off from 26" to 30" is an estimate based on the 4 configurations they tested (the 3 listed above and dual 20" standard monitors).

As far as I can tell the findings of the study were:

1. The 24" and 26" widescreens outperformed the dual 20" standards on a text editing task. 2. The dual 20" standards outperformed the single widescreens on a spreadsheet task, but by a lesser margin. 3. The 26" widescreen significantly outperformed the 24" widescreen on the text editing task at the same resolution.

How do you conclude from this that 2500x1400 is the optimal resolution for the vast majority of people?

After a re-read, it's clear my transition was ambiguous. My point was: the whole study is pretty bogus. It's an excuse to get you to buy more, more expensive monitors. therefore here's my take: there's some optimal number of pixels you want-- not an optimal number of monitors, but a number of pixels.

Users of Microsoft Windows have had to put up with weak facilities for managing large numbers of windows for a long time -- I remember having 60-70 windows open on an AIX workstation back in the 90's and never feeling lost, but it's pretty easy to feel lost managing 6-7 windows on a Windows machine.

Multiple issues are involved, for instance, the way that window application processes are responsible for moving themselves, the way focus works, the way cut-and-paste works, etc.

(To be fair, Microsoft has been steadily improving matters, and Win7 is a step forward)

Anyhow, I find things easiest in Windows if I run with windows maximized much of the time, and a multi-monitor rig is a good way to do that and still be able to see more than one thing at a time.

Some Windows apps are very well designed to work w/ multi-monitor rigs too. Visual studio, for instance, makes it very easy to split the VS across monitors 1 and 2 and have the application you're running on or your web browser on monitor 3.

I am surprised about the conclusion that pixels matter more than monitors.

I've found that the (admittedly, somewhat arbitrary) boundaries of monitors let me organize my windows as "one active window per monitor". This keeps separation of functions of the monitors easy to keep track of subconciously.

The tri-monitor layout of editor-requirements/research-build/testing has worked wonders for my productivity. I'm not sure I could easy translate that to one ultra-dense monitor in quite the same way.