I recently watched a talk by Ben Orenstein about bringing your Vim proficiency to the next level. A quick poll of the audience revealed that many (if not most) of the Vim users present self-identified at a low-intermediate skill level with the editor, and Ben went on to discuss several habits that prevent many developers from advancing further:
- Navigation with the arrow keys/mouse
- Slow file selection (e.g. using visual file trees)
- Scrolling within files by holding “down”
- Editing text using incremental character movements and visual mode instead of text objects and verbs
These suboptimal strategies all seem closely related to me in that each represents a form of visual movement (Progress) towards the end goal where a faster search-based Jump option is available.
To take the scrolling example, we seem predisposed to this physically analogous “walking” through the file even when we know the exact sequence of characters we want and could execute a Vim string match must faster. I think many would argue that the process of navigating to our target location here is more or less the same, with the former strategy simply being a slower version of the latter since it doesn’t take advantage of a speed enhancement tool (Vim’s character matching). But these strategies are undertaken with fundamentally different states of mind.
The first strategy is an example of a Progress-driven approach. You are undertaking an exploratory endeavor that allows your brain to process the topology of the document and the context in which you will ultimately find yourself, and to put all of this contextual information along a timeline of events (you were here, then you did this, and that’s how you got here).
The second strategy is an example of an information-driven Jump approach. You incorporate knowledge of your environment (in this case the document) and the tools at hand (Vim) to circumvent the progression altogether. The result is landing in the same place without witnessing the transition state.
Ok so what, who cares?
The more I thought about this idea of a Progress vs Jump mindset, the more its relevance seemed to expand out of Vim into other daily work-flows. Take a Google search for example. If you know what you are looking for generally it is the first search result (and in many cases I know for a fact it will be). Yet how often do I use the “I’m feeling lucky” button? Never. It’s not like using that button has never occurred to me, but something in my psychology prefers navigating to the first result through the results page. At the risk of over-extrapolating, consider how distrustful we are of snap judgements (see Malcom Gladwell’s “Blink”) and how often you peruse the entire menu of your favorite restaurant despite already knowing your order.
My point isn’t that the Jump mindset is always better, but rather that it is necessary if productivity is the goal. And we appear to be naturally inclined against it. So what makes us unconsciously shy away from Jump strategies?
- Distrust in tools. I don’t use the search bar in the OSX Finder; instead I navigate folder by folder (if I find myself in the Finder at all). Why? I’ve lost faith in the Finder search bar’s ability to find what I’m looking for, since it never seems to work. Jump strategies require a lot of trust in the tools enabling the jump, and the pervasiveness of crappy tools has created a culture of distrust with respect to software at large.
- Uncertain information. There is a certain amount of contextual information you need to have in your working memory for a Jump strategy to work: You need prior exposure to the context and confidence in your recollection of it. There is a comforting surety in the Progress approach that trumps the productivity benefits of Jump in unconscious thought almost every time due to a lack of confidence in your own memory (the chicken wraps are the lunch special on Tuesdays right? Or was it fish tacos…better check).
- Preference for deferred effort. A Jump strategy requires more thought up front (e.g. figuring out line numbers in a relative Vim command vs. highlighting in visual mode). We tend to favor the easier immediate path even if the end result will be more work.
- Concern about personal limitation. Maybe this one is just me, but for some reason I intuitively feel like there is an upper bound on the number of hotkeys and shortcuts I can remember. I think this is related to general inertia in the learning process; we tend to shy away from things we don’t currently know in favor of the things we do, so learning alternative ways of doing things requires conscious effort.
When I’m trying to maximize productivity I’ve been making a conscious effort to identify those four hurdles, and recognize if I’m making Progress decisions when a Jump option is available—I’ve sped up a good number of my daily processes already.