Beyond Minimal Design: Minimal Solutions

I was listening to a talk by Nicholas Negroponte regarding his One Laptop Per Child Project, and was struck by his description of stripping back the idea of a laptop and reformulating it into something that would be a cheap and an effective tool for learning in developing countries. A powerful vision and a brilliant man, no doubt about it. I was really interested in the notion of redesigning a common consumer product for a similar but different and more specific purpose, and the insights that provides on what that product is and how it is conventionally used. I was also thinking about the methodology those guys used to even be able to get into the right 'head space': principally stripping commercial perspectives of technology design and construction such as feature bloat and lock-in.

The rapid development and release of primitive versions of ideas followed by incremental improvement is the mantra of agile software engineering, and of some of the Godfathers of startups. This process is a methodology for distilling an idea for a product or service. The realities of implementation, technical minutiae, and related time pressures force the distillation. I would advocate stopping there. Keep the minimal version of the product, and exploit incremental improvement to improve the adaptive fit of the system for its application (however the users help you to define what that means). Specifically, I'm referring to the sometimes irresistible urge to add features that are not core to a problem's solution (what the hell is an AJAX?).

I like visiting big websites that have heaps of features, although I don't want to live there. I use them like shopping centers to meet a specific need. I may browse or window shop a little although I usually bail once my need is fulfilled. Simply, I have better ways to spend my time. I 'live' on minimal websites. These are the online properties that get the most so-called 'face-time' in my browser because they help me to get things done, even if that 'thing' is optimised procrastination. Google products for example let me do things. Beyond a minimal design, the service sites of choice are minimal in features. For example, when I'm collating some notes, docs services me better than notepad, and trims the gigabytes of features from MS Word that I have, and never will touch. I fill my digital world with an ecosystem of these minimal service solutions. Their simplicity comforts me, as a user I feel like I have some control and that the service is transparent, whether these perceptions are true or not doesn't matter.

A minimal product or service represents a perfect case for a technology startup. A well defined problem with a no frills solution. A solution that nails the problem, irrespective of whether the problem is for a low percentage of the web users with the problem. Realising ideas as tangible products and services, like anything, is a skill that improves with practice. You educate yourself, and train on manageable cases of increasing difficulty (ambition). Plans of glory are checked at the door, although success comes to those that work hard.

Around 2005 I was managing my bookmarks in an XML document which I would run through a simple XSLT to produce a webpage I could access at work and home. Serendipity revealed del.icio.us to me, which solved exactly this problem for me. I no longer value bookmarks (search is king), so bookmarking applications no longer appeal to me. A few months of pushing links to the database highlighted the point. I got a lot more value mining the corpus for relevant pages (yet another way for me to search), and using the 'just posted' page as a pseudo news aggregation site. I liked and used it because it was simple: it did little other than capture my bookmarks for later location-independent use. It was not part of some big machine which tried to sell me other services or blind me with features I didn't understand or want (we're all just dumb users after all, just in different contexts).

del.icio.us is a startup wannabes wet dream. It is a simple site, with an enormous user base, built and maintained my one guy, that was bought out for lots of cash. As a case study it is important to point out that 'the guy' did a whole lot of web-based problem solving before del.icio.us, most likely spending years refining his capability to realise and launch an ideas into viable produces and services. Popularity, and therefore the user base and buyout may have been lucky, but you can bet that his execution of the project was close to clinical. Other seminal one-man cases include Richard Cameron's citeulike for academic references (a killer application as far as I'm concerned), and Gabriel Jeffrey's grouphug (procrastination or research into the human condition, you choose).

The patten, like writing a research paper is: one problem, one solution, no frills. The solution is arrived at not by coincidence or luck, but systematically and as a result of training. I would advocate not just minimalism in the presentation-level design (an implementation detail), but rather the methodical and informed removal of choice. Specifically, I'm referring to Richard Gabriel's notions of "worse is better", and Barry Schwartz's take on "less is more". A persistent minimal solution is not one-size-fits all (case in point), but it is a viable perspective and a good start for startups.