A design philosophy for 2015
A friend asked me what my design philosophy was and how it had changed over the years. He was asking about the design and flow of applications and websites. Actually, he actually asked about visual design specifically, which I found interesting in itself.
I paused for a moment when he asked, which is unusual for me as I’m normally quick to answer questions about myself. It’s a big question, in that he was looking for both my current opinion and how my perspective has changed. It was more than that though, as he was also begging the question of “What is design?”
He clarified his question by adding that he wanted to understand what I thought was good design and good design practices, which led to a discussion with many examples and clarifications. We talked about how websites like Reddit.com could have great content yet such a visually appalling, awkward, and dense experience. He initially wondered if visual design was the crux of my philosophy, but I told him it was more than that. Aesthetic can further a site with a better or more focused experience, but the visuals do not stand on their own.
We talked a bit about I what I thought design was terms of the overall experience, including information design, content flow, information modelling, and aesthetic. I think his perception hinted that I’ve talked a lot more about the visual aspects of design in the past and less about the other factors. I likely have too, as I am most inspired by things that look great, and that inspiration is something I talk about. Things that flow and function well, while still inspiring, are less obvious and are things I speak about less passionately.
We talked through more examples, each of which I found helped to remind me what I had learned and where my mind had explored. I tend to forget how long a journey it has been, that other people haven’t seen what I have, and that I haven’t seen what they have either.
It was a great question. It made me think about design from a different perspective, considering where I’ve been and how I’ve travelled. It also made me admit that I had not been writing about design, something I should be doing much more of.
So what have I learned in 2014?
Focus and content is something that was missing on the web, in web applications, and in software in general. Luckily we have seen this change in the last few years. I believe focus to be the most important shift in the design of content and applications in general, as well as a narrow, more vertical scope.
Applications and content improve as unnecessary things are removed. As things are pared away, applications become more tightly focused on whatever their core truth is. Focused writing applications are one example of this, or narrow publications like Serious Eats, or simplified presentation mechanisms like the presentation of content in Medium.com. Single serving sites are another example of narrower approaches (doing one thing well, often focusing presentation and limiting the scope of what they provide). Focus, on all axes, is paramount to better experiences.
Grammar and standards are something I’ve talked a lot about over the years, but recently I have become more comfortable with my position. Both grammar and content standards are a fundamental part of content design and experience. This covers things like post title formats (e.g., sentence case, tabloid case, AP case, etc.), citation formats, callout conventions, and so on. There are many documented standards that we should look to first when designing content and applications that use metaphors from written and printed materials. Ignoring these standards reduces the overall experience in many applications and websites.
More importantly, these standards provide free and comprehensible design. Many people already know what a citation looks like, for example. Following a standard format communicates citation-ness for free. Using one particular format for citations communicates even more, it also reflects the flavour of the standard and its history.
Typography and document layout standards are something I have been studying more recently. This is another aspect to the design of content that is well-solved, where applications and websites are often ignorant of rules of thumb and practices that have been tested over many centuries.
Combining grammar, writing standards, and typography provide ranges of colour that reflect different, specific meanings. These palettes of meaning can be used intentionally in design to reflect different qualities in the content presentation. For example, using tabloid case for headlines and colloquial grammar with a relaxed (sans-serif) style body font can be used to suggest content that is more down-to-earth or provincial. Content presented with Oxford rules, classic serif body fonts, and classic styling screams academic. Realizing the meaning communicated by the various stylings and that there are standard styling mechanics, is something that I’ve known about for a while but only recently been using together as a larger design brush.
Skeuomorphism versus minimalism are application design principles we have seen in applications and websites over the past few years. The trend has mostly been a movement away from the literal metaphors to simpler constructs (and then back a bit again).
Apple, for example, used very literal constructs for applications in early versions of iOS and OS X. There were aspects of this that were good, but it (in the opinion of many) lacked tact and style. More recently, Apple dialled the literal back to zero resulting in a very minimalist set of applications and constructs. Google followed a similar path, but they stopped short of zero, coming up with a design language with concrete (but still abstract) metaphors. Understanding these extremes, their history, and effect is important.
There are aspects of skeuomorphism that are important. The idea that concrete objects deliver a familiar and intuitive base understanding is a useful design tool. It is interesting (though somewhat obvious) that the completely-literal, over-the-top metaphors of earlier OS’s and applications felt off. And it is unsurprising that if you strip away too much of the concrete that the result can be confusing. The concrete and literal reflections of our world are important in software, and the balance is an significant factor in comprehension and comfort.
There are a number of anti-patterns that also help shape how I now think about design. These are simpler concepts that I find lead to poorer overall experiences and content.
Attack of the boilerplate (or death by template), for example, results when obvious technical implementations box content into shapes that work in some cases but not others. The best example of this is the blog post in both full and summarized forms. Blogging tools tend to use single representations for things as it is technically simple. There are many types of content and ways to present content, and templates will often force all things into a single format.
The best example of how a template format can degrade content is to compare Reddit and Digg versus Twitter (or the old TabDump/HackThePlanet). Reddit and Digg display all links and posts identically, including superfluous controls, links, and attributes. Twitter boils links down to a statement and link(s), in a more freeform manner. Both techniques are viable, but the latter is more focused and more pliable to different types of content.
Boilerplate-itus is also a problem in applications. User interface toolkits provide a set of tools that can often be used out of convenience and not based on what a problem deserves. Many video applications, for example, stick to tiny play controls in the bottom left of a video. It’s a tiny target for a mouse pointer (or finger), and is mostly the result of using what is already there. Is it good design? Probably not.
On the other end of the spectrum, the not-invented-here anti-pattern can be worse. Just take a look at Google applications on iOS. They feel out of place, breaking standards left and right. This isn’t always bad, but following the Google-Way ™ on another platform is an extreme that can also produce awkward experiences. My favourite example here is the Google YouTube application for iOS, which while functional, has a navigation system that is counter-productive on iOS.
Much of what we think in design is based on examples and the patterns these examples boil down to. The concept of examples and patterns is one of the strongest communication tools in design, as an example can communicate so much with so little effort.
And finally, information modelling and architecture is what I consider to be the starting point for good design. Finding the concrete ideas and constructs in a problem, mapping those to experience and flow, and using this underlying reality to select brushes from standard techniques like UI literalism, content, typography, is what design is to me. Know the landscape of the problem. Use the tools available, with knowledge of the tools together with your map of the landscape. There are many valid paths through a problem and the best ones are found with an honest understanding of it all.
I used to talk about this domain modelling by describing finding the underlying things. A thing is a thing, and finding these constructs and staying honest to them is critical to understanding a system. Unfortunately, using the word thing so frequently in talks and writing is confusing, so I’ve been searching for better ways to describe it. Really, I was looking for better ways to talk about affordances and domain models, terms the are more difficult to use in conversation than thing and object. Ultimately, I think landscape, paths, and tools are better metaphors to describe why things like affordances and models are needed.
So I see design techniques as part of a set of tools that allow us to manufacture experiences and function. The techniques are often based on a history of knowledge and experience with resulting standards that we should understand, that if we ignore we are choosing a path of unfortunate ignorance. There are often multiple standard techniques, each with purpose and meaning. Realizing the value of these different approaches is important for finding better paths through the problems we design our ways through.
The techniques, however, are only the tools available to us. They are not design.
Things like information architecture and domain modelling are cartography techniques. The allow us to define and map the landscape of the problem, using various methods of dissection, diagraming, and documentation. These are also not the entirety of design, but rather the mapping part of the problem.
I see design as the mapping and tools we use to find a great path to solve a problem, where great is defined by things like, time, overall experience, appropriateness, safety, and so on. Design is pathfinding, which relies on mapping and the tools of our trade.
So where do I go next?
I will write more about the various aspects of design. I will continue to doodle, draw, and diagram my use of the standard techniques. And, I will build more, and write more about what I build, as design is only as useful as the projects that use it.