The *real* reason you want a multiple monitor setup
Despite the fact that there is little evidence that using multiple monitors will make a programmer substantially more productive, many coders will subjectively claim that they can’t live without a second display. Why do people feel so strongly about the issue? And is it possible that the perception of efficiency is just as important as real efficiency?
The Scientific Angle
A few months ago, I experimented for a while with a dual-monitor setup. My main computer is a 14″ Thinkpad, and I connected to either a 24″ widescreen LCD (in my lab at U of T) or my 20″ widescreen at home. After a while, I found that I wasn’t really seeing the “obvious benefits” that some people rave about.
I had heard about studies that supposedly proved that you can be up to 50% more productive by adding a second display. In my post Multiple-Monitor Productivity: Fact or Fiction? I looked at these studies, and concluded that in some isolated tasks — like cutting and pasting, or working with a large spreadsheet — you can see a significant benefit if you add a second monitor. But for most programming tasks, the benefits are going to be minimal (but still there).
After another study was recently published by some researchers at the University of Utah, Jeff Atwood took the time to put together a summary of the studies of all the studies we could find — a “a one-stop-shop for research data supporting the idea that, yes, having more display space would in fact make you more productive”. In case you couldn’t tell, Jeff is a big fan of multi-monitor setups:
I have three monitors at home and at work. I’m what you might call a true believer. I’m always looking for ammunition for fellow developers to claim those second (and maybe even third) monitors that are rightfully theirs under the Programmer’s Bill of Rights.
The Subjective Claims
If you read the comments on Jeff’s article, you’ll see that, despite the lack of empirical evidence that programming tasks will significantly benefit from multiple monitors, many programmers are pretty attached to idea:
Leon Mergen: “People who claim there is no benefit (or little benefit) in programming with multiple monitors, obviously haven’t really expercienced it.”
SB: “Have you ever actually used (like, for many months/years) multiple LCDs? … I don’t know how a programmer could go from multi-LCD setup to single display & not claim some, even if minor, productivity dropoff.”
Brian: “I personally find that in my case having a second monitor is ALWAYS more convenient and increases productivity.”
This morning I finally ran across a paper which talks about these subjective benefits. Jonathan Grudin’s Partitioning Digital Worlds: Focal and Peripheral Awareness in Multiple Monitor Use has some interesting insights. Grudin interviewed 18 people who used multiple-monitor setups, and came to the conclusion that:
A second monitor improves efficiency in ways that are difficult to measure yet can have substantial subjective benefit.
One of his interesting observations was that it’s not just about the screen real-estate, it’s also about the partitioning (emphasis mine):
A strong demonstration that multiple monitors can be more about partitioning than about increasing space is provided by the two participants who dock their constantly synchronizing palmtop computers next to their desktop monitors. One keeps his calendar visible on the palmtop, the other keeps email visible. The increase in space provided by the palmtop display is not significant and there is no information on the palmtop that is not available to the desktop computer. The value is in having instant access to a resource in a known location in peripheral vision.
This the same conclusion that Jeff made after seeing the results of a small, informal multiple monitor productivity study: two monitors is better than one large monitor.
Another interesting finding in Grudin’s paper was just how much people hate to use the taskbar or Alt-Tab to switch windows:
Given the ease of minimizing and restoring windows, why bother with a second monitor? Repeatedly, people indicated that they considered it a relief not to have to use buttons, “escaping from the need to Alt-Tab.” The ability to avoid a few keystrokes is welcomed with great subjective enthusiasm, although it might be difficult to objectively measure an efficiency gain.
Perception vs. Reality
To me, this really captures what the argument’s all about. It’s not necessarily about actually being more productive — perceived productivity is just as important. It reminds me of Bruce Tognazzini’s famous finding on the relative speed of the mouse vs. the keyboard:
We’ve done a cool $50 million of R & D on the Apple Human Interface. We discovered, among other things, two pertinent facts:
- Test subjects consistently report that keyboarding is faster than mousing.
- The stopwatch consistently proves mousing is faster than keyboarding.
This contradiction between user-experience and reality apparently forms the basis for many user/developers’ belief that the keyboard is faster.
In my experience, many people who love multiple monitors are the same people who are obsessed with knowing every keyboard shortcut in their text editor, and who can’t live without mouse gestures in Firefox.
Don’t get me wrong. Even if the benefits are unproven, minimal, or even non-existent (as in the mouse vs. keyboard case) — it doesn’t really matter. The most important thing is that you, as a programmer, have the tools that you want to do your job. I definitely don’t question the productivity benefits of being happy.
6 Comments »
RSS feed for comments on this post. TrackBack URI
I think you’re heavily misinterpreting the Tog quote. I posted a response here:
http://www.codinghorror.com/blog/archives/001088.html
Jeff,
I don’t think I am misinterpreting the Tog quote. Your post makes it sound as if Tog reverses his position further down in the article. He does point out an exception to the rule, but the part that I quoted is still representative of the main point of the article.
[The remainder duplicates the comment I left on your blog]
You make it sound as if the later part of the essay completely contradicts the first paragraph that you quoted. But all he says is that Ctrl-C and Ctrl-V are an exception, because you can use both the keyboard and the mouse at the same time. That hardly means that “modern” keyboard shortcuts aren’t bound by his claim.
You’re right that there is a handful of frequently used shortcuts that are undoubtedly faster than using the mouse. And I agree, it’s not a cage-match between the keyboard and the mouse.
The link you posted in the comments is also a very valid criticism of the Tog quote — where are the results? What was the method? The quote, on its own, is probably past its best-before date.
But I don’t think the conclusion of the article is necessarily that “using the mouse is ALWAYS faster than using the keyboard.” The takeaway is that something that feels faster is not necessarily empirically faster. In my article, my point was that there is a different between a perceived productivity gain and a real one.
I think the potential gains in productivity due to multiple monitors will vary greatly from developer to developer. I’d argue that what you call “perceived” efficiency is more aptly called seamlessness since what matters is not the time it takes to accomplish a trivial task (mousing vs. keyboarding), but the degree of distraction caused by performing the trivial task.
It’s hard to measure a subjective quantity like the “degree of distraction”, but that doesn’t mean it has no effect on productivity. As a developer, you’ve no doubt experienced being “in the zone” while writing code, getting to that point is non-trivial and often involves matters out of your control (scheduled meetings, unexpected phone calls, noisy co-workers). I think for certain people (myself included), having to click through icons on the taskbar or alt tabbing when more than 2 or 3 windows are involved can also be an obstacle to getting “in the zone”.
Also, visual learners (http://en.wikipedia.org/wiki/Learning_styles) may be more likely to get a productivity boost from multiple monitors. I like to be able to see all related windows (code, docs, terminals) without having to cycle through any of them, virtual desktops are a godsend that allow me to do “group” different applications together, multiple monitors allow me to group more applications per virtual desktop. I don’t understand how windows developers deal with a single desktop, but different developers have different habits and workflows.
ben
Hi Ben,
Yeah, I think you’re right. It’s more than just pure “efficiency” that we care about. It’s not just about raw speed. I can imagine that for some people, multiple monitors can really help them get into the flow) state.
Good point about visual learners being more likely to see a productivity boost from multiple monitors. I hadn’t thought about that.
I think that Jeff is misinterpreting the Tog article, it is completely damning of keyboard only use and says that everyone who THINKS that shortcut keys use is often faster, are just plain wrong, it says mouse is the way to go for everything.
QUOTE
Command-Key Illusion. Since users do experience the illusion that keyboarding is faster, there is market pressure to supply them with “shortcuts.”—even when using “shortcuts” will actually slow them down.
/QUOTE
Yeah… That whole Tog article is patently ridiculous.
The only way that using the mouse only would be faster is if you haven’t ever used the app before, and the app made discovering keyboard shortcuts hard (eg: makes you hold down alt before showing access keys, not showing shortcut keys next to menu items, etc).
As for multiple monitors, I am a bit biased, as I have 4 24″ and 2 22″ monitors at home, and 3 monitors at work…
However, multiple monitors suffers much the same plague as keyboard only use, most apps don’t make use of multiple monitors and even behave badly when you try to use them with multiple monitors (eg: try opening two Excel sheets and putting one sheet on one monitor and one on another, I was finally able to get this to work by editing the registry… blech).
Just look at the Windows start bar, it only appears on ONE monitor! What if I only want to see a task bar on the current monitor that shows only apps that are on the current monitor? (Yeah, I know, install a 3rd party hack… blech).
Imagine if your tabbed web browser only had one tab bar, and showed tabs from EVERY browser window on that one tab bar?
Also, I can’t believe Windows still doesn’t have virtual desks built in. I found a virtual desk manager for Windows that works pretty well (NOT the powertoy…), but without support built into the OS, it will always have serious faults since it’s just a hack.
If you have to use only one monitor, virtual desks make it much less painful, and they enhance multiple monitor use as well.
@scythe:
The whole point of Tog’s article is that although it might seem counterintuitive, the mouse can actually be faster sometimes, even if it seems slower. Until you put a stopwatch on something, you can’t know how long it really takes.