Last week, CERN quietly redirected their URL to the new site. And Mark Boulton Design and the CERN project team saw the chapter close on nearly two years of work.
There is much to talk about with a project of this scale. Probably the first thing to say is that this was not a website redesign project. The project started with the very simple statement: ‘we have a content problem’. I’d like to think it’s a problem we solved, but it also went much deeper than that.
CERN is an organisation that supports collaboration. In fact, CERN is comprised of many collaborations; more community than organisation. We have a bit of experience designing with communities, and the teeth we cut with the Drupal community came in very useful here.
As we all know, the question ‘why wasn’t I consulted?’ can derail almost any endeavour. This was no exception. So, early on in 2011, the Mark Boulton Design team spent months talking and listening. In fact, the majority of our first six months on the project was spent trying to understand CERN’s audience – both internal and external – and what CERN does for them.
As in 2008 and 2009, when we worked with the Drupal community, designing for community relies on one thing: listening for trends. Design by committee doesn’t work because there are relatively few stakeholdersm and sometimes with big egos. Design by community works because of the sheer numbers and relatively flat hierarchy. Instead of ensuring the boss is consulted, it’s more important to consult with just two types of people: those who have influence, and those who have a big mouth. Get both of them on-side, and your job will be easier.
CERN started with a content problem. To really understand how content was to be reorganised, rewritten, and – in some cases – deleted altogether, we needed a plan. Not only that, we needed to understand how this content was being consumed and where. We needed a content strategy.
We embarked on a good few months of content research – audits, interviewing, process-audits and the like – before producing a strategy that sat alongside a new Communications strategy from the CERN comms team. Together, these would redefine how content is created and managed at CERN, and how it gets published to the web.
The new content for CERN is Future Friendly. It’s adaptive – meaning it’s carefully chunked into small fields in the CMS and tagged with really great meta data so it’s able to shift and flex into different containers. No WYSIWYG here. Just really smart, well-structured – and reusable – content.
To my mind, good design decisions rely on good research. And for this project, we had some great research throughout. A broad amount of data was gathered by various means and filtered down to the design and product team. Not by way of reports and documents, though. In my experience, these often get cast aside. No, in our team we had Emma embedded with the design team. Always present with up-to-date insights into the audience needs so that she could literally sit beside us to help us guide some of the design decisions.
I’ve written about the design strategy for CERN before. How the content needed to inspire, educate and inform the various audiences of the website, but we needed a simple model in order to map the content and the subsequent presentation of the content. Our Make Mantra – Create Wonder – was born, and is still used today as way of gauging how content is to be created and presented.
Practically, we did a lot of prototyping and very little else. Prototypes became the way we experimented with concepts, discussed and communicated ideas, worked on responsive elements. These prototypes then quickly moved from static HTML into dynamic lo-fi Drupal prototypes. This was so editors could get writing content quickly – actually using the tools they would eventually use on the site. All the while, we were continuing to iterate. In all, I think there were 5 or 6 alpha releases, 8 betas and many, many pre-alpha prototypes.
The new CERN public website runs on Drupal 7 (the most recent version of Drupal, that we designed the UX and UI for). It’s mostly an out-of-the-box installation with a few bespoke modules that integrate with internal services such as CDS (CERN’s media server) and Indico (the event management database).
I don’t often do it, but this is the bit where I get to gush a little. The people involved in the project is pretty epic. But, there was a core team of designers, developers, product leads, researchers, content strategists and journalists that deserve a mention. Without them, this project would not have turned out the way it did.
Such a smart team. There are of course countless others – probably hundreds of people – I could thank. But, this isn’t the Academy Awards. If you helped, you know who you are, and I’m very grateful.
A special project
This project always was a bit different. CERN is a special place, yes, but it takes a degree of risk and subsequent trust for an organisation of this scale to work with an external company for such a length of time. CERN invested time and energy to work with us at Mark Boulton Design. When we suggested radical changes to process due to our research findings – things that have had disruptive effect on the communications of the organisation – they didn’t flinch. They knew it was the right thing to do. And supported us in our proposals. That takes a very special type of client; one that is willing to truly partner with a design company. And, I’d like to think the results speak for themselves.
Not just a website
When I reflect on the past two years work on this project, it’s easy to think of it as ‘bloody hell, that’s a long time to make a website’. Well, yes, it is if that’s all we did. But, in the words of James Gillies, Head of CERN communications team and project sponsor:
The launch will mark a significant change in how the Organization facilitates communications within the CERN community, as well as how it presents itself to the world. The first thing that you will notice is that the website looks different, but the most significant changes run deeper, and they form the basis for a new communications architecture.
He goes on to say:
There are many ways we could have chosen to build the new CERN website. We could have asked designers to create something beautiful. We could have created a template for content. Or we could have tweaked our existing websites. Instead, we went back to basics and reflected on who we want to reach and what their needs are. Then we asked them to help us. All our stakeholders have had an opportunity to influence our new website, and many of you have taken that opportunity. We have conducted regular audience research through online polling, usability testing, and in-depth one-on-one interviews, receiving feedback and input from thousands of users. I would like to thank everyone who got involved: your feedback has directly shaped many of the design decisions and turned this into a real community effort.
You see, it wasn’t just a web site. It was helping CERN understand who they’re talking to, what they’re saying, and how to say it. Then, giving them the means to do it.
Breakpoints. Break. Points. Points at which things break in your design.
But, at what point did they start becoming ‘the point at which I will create a new layout entirely’, or ‘the point at which I introduce a new canvas’. What started out as a method to optimise your designs for various screen widths has turned, ever so slowly, into multiple canvas design. We’re creating pages again, and we probably need to stop it.
A breakpoint should not necessarily be a point at which we wholesale swap out to another layout. It can be – and I think there are good reasons to make some of those big, macro changes to grid system or navigation patterns. BUT, I think we’re missing a trick for using breakpoints to make lots of subtle design optimisations.
A responsive design can have multiple breakpoints, say for a small-screen phone, then a large-screen phone, then a tablet, then a laptop/desktop. Many teams try to decide on breakpoints using average screen sizes.
However, it’s better to look at what the content and navigation wants to be. By letting the content and navigation drive the breakpoints, teams find they can often get away with fewer screen configurations. For example, a high-resolution Retina iPad might easily share the same configuration as a well-constructed laptop display, while lower resolution tablets might just need a little adjustment to that same configuration.
Our experience at Mark Boulton Design is that we can actually tend towards more breakpoints than fewer if we take the approach of optimisation rather than a device-centric, display-size approach. We’re moving towards including all media-queries in-line in the CSS, rather than abstracting them into their own 480.css file, and we keep the big macro-layout changes in the Gridset CSS. But, interestingly, what we’re seeing is we have less styles associated with more breakpoints as we’re not creating a whole slew of different, fixed layouts and all that goes with that.
‘But, what is content-out design?’
I first discussed the idea of content-out design in the context of a page. A page with edges that is used to define an aesthetically pleasing layout within. On the web – especially the responsive web – we don’t have edges, so it’s best to define your layout from your content. But – and I get asked this a lot – what does that actually mean in practice?
Firstly, it means defining your grids (your big macro break-points, and the columns and what-not) from actual content and not from devices or screen widths. This could be the size of an image, or an ad-unit, or the typographic measure. Just some ‘fixed’ constraint. This will give you your grids.
As you’re building out your responsive design, you should be focussed on watching how the content adapts as the viewport changes. Somethings I look out for:
Type size and leading. Does it need to change?
White space (macro and micro). Do you need to adjust margins and padding?
Vertical space. Do you need to reduce it and make the content more or less dense?
Flow. How is the readers eye movement going to change as you change these elements?
Words. Are there now too many on one line? Or too few?
Source order. Are the right things in the right place?
Focus on the in-between
Content-out design means defining your underpinning design structure from your content, and then focusing on what happens in between ‘layouts’. This approach of optimising your design by adding media queries (I like to call these optimisation points than break points, because nothing is broken without them, just better), means you are always looking at your content as you’re working. You become more aware of the micro-details of how the content behaves in a fluid context because your focus shifts from controlling the design in the form of pages, to one of guiding the design between pages.
I find this period of change we’re in right now fascinating. Not only is centuries old design theory being rewritten, but the process of how design happens is changing too. It’s becoming more collaborative, more fluid, lower fidelity. Less about being squirrelled away in a dark corner crafting a beautiful thing, and more about scribbling away in public and repeatedly throwing our work in a big trash can. And the more we’re doing that, the better we’re getting.
Optimisation points are just one way we can start thinking content-out. And thinking content-out is what we need to be doing to create truly native experiences on the web – wherever the web is.
My four year old daughter attends a rural primary school in the Vale of Glamorgan in Wales. The school performs in the top 3% in Wales. Consistently. It’s over-subscribed with waiting lists to attend. The head teacher greets every pupil – by name – at the school gates in the morning. She even knows my name. It’s a school you feel part of. And the government wants to close it.
We learnt this week that, as parts of plans to change catchment areas for secondary schools, that my daughter’s wonderful school is in a list of a few schools that are pencilled in for closure. The pupils will be sent to other, under-performing, under-subscribed, urban schools. We have a small window to fight these changes, but it seems that once this process gets started, it’s difficult to reverse.
When I picked my daughter up yesterday, she told that the head teacher had told them in assembly that everyone is going to fight for their school. That mummy’s and daddy’s should write letters, they should too, and we should make posters. A wonderfully simple attitude to a dreadful situation. I gave her a hug.
As soon as we got through the house door, my daughter went straight to her desk and asked me if she could write a letter to the men in charge. This is it:
Change. I love my school. Please don’t close it. Change your mind. Love, Alys Boulton.
This is what the men in suits don’t see. They don’t have to sit there and explain it to a four year old that thinks it’s all going to be stopped with a couple of posters. They don’t think about the illogical idiocy of closing a top, over-subscribed primary school. They think about numbers and averages. They don’t think about rural communities, or try to understand why this school is doing so well. No. Thinking about averages leads to average. And who wants that?
So, we’re going to fight. With every last breath, I’m sure this school is not going down without a fight.
On the run up to Christmas, that wonderful annual advent calendar 24 Ways published an article called: How to make your site look half decent in half an hour . In some designer quarters, this was jumped upon as a load of unconsidered twaddle written by a developer who shouldn’t be dishing out design tips. Aral Balkan took this task and has written a great article that follows my thinking as to what ‘design’ is. All good. But, I’d like to comment on what I thought that 24 Ways article was about, and really the indication of a perception of design across web professionals.
For the past few years, I’ve been teaching a workshop based on my book, Designing for the web, to lots of web people, but mostly to developers, front-end devs, UX folk and project managers. All of them have come along to the workshop to learn just one thing:
‘How do I make this thing I’ve done look less shitty?’
Now, if you’ve read my book, or my blog for a while, you’ll know that I see design as a holistic practice; a practice of understanding content, of audience needs and goals, of brand, of psychology, and, of making things look less shitty. As much as designers would like to think we’re the guardians of all things considered practice, taste, and learned process. We’re really not. Some of the best designers i’ve worked with have embraced the passionate learning of students and amateurs. People who are not interested in becoming professional designers, but want just enough knowledge about an aspect of design to make their thing better: be it their resume, their website or their father’s cafe menu.
Getting stuck in
There is nothing wrong with getting stuck in. I’m not a carpenter, but I’ll get stuck in and learn and renovate my bathroom. I’m not a bike mechanic, but I’ll learn enough to maintain my road bike. I’m not a developer, but I’ll hack around with a bit of PHP to get to where I need to be. Many people would not class themselves as designers, but they can arm themselves with techniques to make themselves better at an aspect or technique of design.
There are many ways to Define The Damn Thing, but for me, design is really just the result of a bunch of questions and answers. These questions may be asked by the materials you’re working with, the constraints of the project, or the audience you’re designing for. They may be big questions such as ‘how can we increase quarterly revenue by redesigning this checkout procedure?’ Or, they may just be small questions like: ‘how can I make this content just look a little less shitty?’. These are both equally valid questions.
The answers to these questions are design. Some answers are big. Some are difficult. Some seem like magic. But most, in my experience, are well-trodden cowpaths of practiced techniques. Simple little things we do to make less shitty things.
So, design is veneer. But Aral is right, it’s not just veneer. Even if you wanted it to be, visual design is not a thing, and we should be mindful to not present it as such. But that’s not to say we shouldn’t help people who aren’t designers make less shitty things by giving them a few hints and tips along the way.
The humble hyperlink is an incredibly simple, powerful tool on the web. In fact, many would say that our ability to link one resource to another via hypertext is what makes the web what it is. But, I think this humble link could do a little more than it currently does.
What’s the problem with links?
When the web was invented, the hyperlink linked resources together. That hasn’t changed. What has changed is the type of resource that may be linked. In the early days of the web, resources were documents of text. They then changed to images or tables of data within text, but it was still text. Now, an html document is still the resource, but the primary piece of content maybe something completely different: video, audio, a download to a file, it may be secure. A link may trigger an action along with linking, such as going fullscreen, a pop up, or linking to another site. All of these things are valid links, but we have just one element with a bunch of attributes in the code to describe them. I’m not so sure it’s enough any more.
It’s correctly marked up in the HTML, but perhaps badly framed editorially. You, as the reader, do not know what that destination is until you click it. ‘Just make the link text better’, you may be saying. Ok, how about this:
That’s better. It’s implying a link to an article. The meta data for the user expectation of this content is in the content itself. But, we can’t always rely on our content creators to reliably do this. And what if the content was more complex, like a report:
Is this a link to a web page, a PDF, a word file, a link to download an app from the app store? What is this?
And that’s the issue. I’d like to know what I’m getting before I click it. I think we need more meta data around our hyperlinks that can give us more of an indication what we’re getting before we get it. I became painfully aware of this just a couple of days ago when I clicked a link and started downloading an HD video, over Edge, that started auto playing. Luckily I wasn’t on data roaming at the time.
How can we give more meta data with our links?
There could be a few ways we could do this. Obviously, we could just write the type of link, destination and other stuff in the content. But, if that destination changes somehow, that’s an added overhead for the content creator. Plus, you could end up with quite verbose link texts.
We could use the <type> attribute introduced in HTML5 to specify the Media type (formally MIME type) of the linked document. This is exactly what we should be doing, actually, but the options don’t necessarily map to our content needs, or provide all of the content type requirements. For example, I may want to link to a video, as that’s what I want the reader to watch, but the HTML resource is not a video at all, but a video embedded in an HTML page with a whole bunch of other things. In that case, the <type> attribute would be inappropriate. The same could be said for audio, or any other primary content that is embedded within an HTML document.
I feel we need two things: some visual affordance for the user that they’re going to a particular type of content; and this affordance is rendered through an attribute, or something similar, in the HTML of the link.
We’re already doing this. Sort of.
In designing around this problem, I’ve noticed through my research that we’re already doing this. We already have conventions used on sites like Wikipedia, and others, for indicating, after a link, what will happen when you click it. Here’s are two examples from the BBC and Wikipedia.
The thing is both of these examples use the icons to append list items. They’re not in paragraphs immediately following the link wherever that link may be.
…a small intense, simple, word-sized graphic with typographic resolution.
Sparklines are efficient representations of complex data that typically sit within content to provide additional semantic value to the content. They’re a small hit of additional information. A picture speaking a good few words.
And I think he’s onto something there and it’s something we could use for this problem.
What would it be like to expand on what we have now? How could we provide more meta data – where appropriate – than just an icon as demonstrated on Wikipedia and the BBC.
What we could do, is include some more information, for quick visual digestion, alongside an icon. We could take Tufte’s Sparkline rationale and apply it to this small, inline iconography. We could use a Sparkicon.
I’m defining a Sparkicon as a small, inline icon with additional link meta data to describe either the content and/or the behaviour when the user clicks the link.
I’ve looked a couple of examples of how this might work. The first example shows two icons to indicate behaviour when we click the link. The first is a full screen icon, the second is one we already know: an ‘external site’ link.
The second set of examples are icons for particular content or media types: comments and video. The comments includes a count of existing comments attached to the link document. The video spark icon includes a video duration. This could also include things like aspect ratio, or HD, or whatnot.
Degrading the reading experience?
I think the challenge with sparkicons would be to give enough feedback to the user to let them know what’s going to happen, or where they’re going to go, or download and what are the consequences of clicking this link. That’s the important thing, and history tells us we can’t rely on the words to do that.
But we also need to be mindful how much of an interruption this will be to the ebb and flow of reading long form content. Also, how much added visual noise would this add? At this point, I just don’t know, but I plan on putting this to test soon on a couple of projects: both personal and professional.
There are plenty of challenges, particularly how this could just be some kind of applied graphics through a simple addition to a link attribute. Standardisation is also an issue – this type of icon would be only really useful if it’s a shared convention.
But, it’s just an idea right now that i’m plugging away at. More to come.
Updates: Joe Gooden pointed out on Twitter that this of course is of particular importance on touch devices where we do not have the benefit of hover interactions (such as Alt text reveals), and status bar
At Mark Boulton Design we get quite a few project enquiries and over the past twelve months we’ve seen an increase in requests for these repositories as a project delivery. This is just fine for us, as we tend to propose them anyway in some form or other to document our work. But there is an inherent concern in delivering this type of thing to a client. When is it done? And how is it created? And making sure that is discussed with a client, in depth.
My first exposure to style guides was through print and branding. Corporate style guides are rules of execution; position this logo like this, use these colours, this type, crop photos this way. Adherence to the guide is critical and there’s always someone policing them. Having been involved in the creation of a few of these printed style guides, they are always designed after a body of work and then – almost without exception – never updated to reflect slight organic changes as a brand or design system grows. This is why digital style guides are good. They allow growth.
I’ve read recently that some designers are starting with patterns. Taking small modularised chunks of content and then compiling for display in different circumstances. This, of course, if great for taking an adaptive content approach. Thinking of our content in small chunks is useful. But, I can’t help but thinking that approaching design in this way is a folly and will result in generic, cookie cutter work.
Future coding is a lot like playing the roulette. If you can guess where the requirements of tomorrow will land today, you’ve just scored the ultimate programmer’s prize: looking like a wizard. You saw the future and you were right! High five, brain!
That’s the almost irresistible draw of “flexibility”—often just a euphemism for building half or whole features before you know how they’re supposed to work or whether you need them. Just like in Vegas, the worst thing that can happen is to be right about this once in a while.
And that’s exactly how I feel about creating style guides. There’s a lot of talk about us designing systems for the web. Well, sometimes, we’re not designing systems at all. We’re designing a couple of pages for a marketing web site. Creating a whole pattern library for a project like that is not only overkill, but completely inappropriate.
But the same can be said for a large scale project too. If you don’t know what the patterns are, you could be abstracting too early – bowing to the lure of future flexibility.
I guess it’s a question of balance, and one I’m becoming acutely aware of recently. For me, style guides are still there to document a design system. But they’re not a tool kit of parts. Before you design a tool, you have to know what to fix first.
I started this blog in 2004. Nearly a decade later, I’d like to think my motivations for writing this blog are are same: it’s a diary, first and foremost. A place where I can document my thoughts, observations, theories, design critique and more. It’s also a place where I can engage in a wider discussion. My blog post responds to yours – or your tweet – and so we go around.
Over the past couple of years, we’ve seen more and more great industry-focussed magazines that publish online and off, but the flip side to this is I’m noticing more people writing for these magazines and less so on their own blogs. Why is this a problem?
Considered, well-written, edited, and produced content is hard – and time-consuming – to do. Writing an article for one of these successful publications is not an easy thing (well, for me it’s not). Not only do you have to have something notable and interesting to say, but you have to stick with it through the editorial process. You get pushed, prodded and cajoled along the way – mostly for the good, I should say. But sometimes, you may begin with a pithy throw-away idea that wouldn’t amount to a couple of paragraphs on a blog (kind of like this blog post, really), but through the process it gathers weight – and in some cases bloat – to fit the requirements of ‘an article’. And articles are different to blog posts.
In traditional news journalism, this is the difference between a feature and a piece of news. A feature is a story. It has a beginning, a middle and an end. It’s designed to have a less time-sensitive, longer shelf-life. But ‘news’ is transitory – documenting a period in time. It doesn’t have a structure like a feature does, and probably doesn’t have to adhere to the same editorial workflow. What we’re seeing on this increasing number of publications are features. And because the authors are writing these pieces, we’re seeing less shorter form writing because that type of content has no-where to live, other than blogs and everyone’s busy.
I’d like to see more balance again. More short, scrappy blog posts (like this one), written off the cuff and in the moment. Sometimes, I’d like to read the author’s actual voice, instead of a homogenised edited one. There has to be a space between 140 characters and 3000 word feature. We all need to blog more, I reckon. Roll on March.
Every couple of years I write a short post to document the year before. This was my Two Thousand and Twelve.
The year started with me taking a month off client work to finish writing my book. It went well, but one year on and i’ve still not finished it. It’ll be done soon.
I blogged more in 2012! Hurray! I redesigned this blog and took it back to how I wanted it: simple, designed for long-form content. Most importantly, it now runs on Statamic and is designed for how I write. I’m really rather enjoying it so far.
2012 saw Emma, Alys, Nansi and I take a holiday in Portugal for a couple of weeks well-earned rest. Alys started school and in September, we all moved house. Going from a relatively new house that we’d spent ten years getting exactly how we wanted it, to a two hundred year old cottage that we’re currently ripping to bits. It’s been interesting and 2013 will be the year I skill-up on DIY. From plumbing to plastering, I plan on doing it all myself, starting with the bathroom.
Balance and Focus. I’m reading a lot, recently, about people ditching social networks, over-sharing, spending more time making axes and keeping bees etc. But this is happening to a certain age of person who’s been doing this web-thing a while. This reactionary behaviour is something socialogists have been seeing for a couple of decades now in the general public in the UK with regards to technology. What it represents is a desire for balance. I’ve spent the last two years working on balance, and up until the last few months of this year, I’ve been better at it. And that doesn’t mean using Twitter less. It means making sure I give myself, my family and my work equal amounts of time and attention. I’ll be spending 2013 getting even better at doing that.
On to work stuff…
Community. I hope it’s a more considerate year. I’ve had a gut-full of embarrassing, unconsidered discourse on Twitter and I hope we see a lot less of it this year in favour of responses in blog posts or in person. I hope it’s a year of respect.
Web Design. I don’t think responsive web design will be the defacto method of web design. Not yet. As long as advertising revenue models exist, and platform channels (eg. m.sites with their own profit and loss), there will be difficulties getting that to work responsively. As I talked about at the end of this year, designing responsively impacts many parts of organisations. But, I do think responsive web design is no longer a trend. It’s a bunch of techniques that have proved themselves useful and relevant, and will probably continue to be so.
I hope it’s a year of a CMS shake-up. I hope it’s a year in which open source CMS communities break out of their IRC bubbles and start working with writers, editors and designers. I hope they become less feature-focussed, and more human and organisation-focussed. I hope we see as much effort going into the design of our back-end systems and work flows as we put into our web site designs and user flows.
I hope it’s a year where we stop Defining The Damn Thing. Design is design. I hope the term ‘visual design’ stops getting used. It’s not a thing.
I hope Google stops giving us great, free, cool things and then deprecating them. I’ll pay from the off.
I hope Twitter starts behaving like a media platform instead of a product. I’ll pay for flexibility, freedom and content.
I hope Adobe stops acquiring everything. They need a big, scary, great competitor.
I hope it’s a year we have a more balanced industry view. I think we need more publications, more voices, more speakers, more blogging, better journalism, better reportage, broader guidance, better conferences, more meet-ups. We rely too much on a small set of voices that guide the industry one way and another.
I hope we all take a moment, every day, and think about the importance of what we do. We’re defining, designing, launching, and iterating upon a medium. It’s big. And important. But… if you have a bad day, nobody died, right? That’s important to remember.
So. That’s that. I can but hope.
Oh. And, in 2013 I hope I stop starting sentences with the word ‘so’. I’m British, and we don’t do that.
I hope everyone has a successful, fruitful, happy 2013. Thanks for reading.
The evolution of the grid from geometric form and canons of page construction is quite clear. In no other period of history was the grid used as a core aesthetic as was in the 1950s and 60s where it emerged – almost simultaneously – from several design schools in Europe. From then, the grid system’s influence has been pervasive.
Today, the grid is viewed by many designers with equal amounts of distain and fervore. Its detractors – and there are many – claim a grid system is visual straight-jacket, designed to inhibit creativity and that using one produces dull work. Of course, I think that’s rubbish; there are no bad grids, just bad designers. In the hands of a competent designer, a well-constructed, considered grid system is the frame on which the fabric of the design is hung. It should create balance, provide landmarks and visual cues. It should allow the designer to exercise just the right amount of creativity and provide immediate answers to layout problems.
For the past 800 years, the printed page has been constructed in pretty much the same way; from the edges. The desire to create the most aesthetically pleasing book always starts with the size of the physical page. That page is subdivided to give us the optimum place to put our text and images.
Fast-forward 800 or so years to 1997.
The web is just about hitting the mainstream. I was working as a junior designer in a small firm in Manchester, UK. Typically, as the young guy in the studio, it was my job to create the websites for clients whilst the ‘serious’ designers looked after the large fee-paying clients on their branding, print design and advertising and what not. Remember, this is the early days of the web.
Designers who were exclusively designing for medium back then were doing what they knew; applying the rules of print design to the screen. We used tables for layout, shim-gifs and all manner of terrible ways to achieve our goal of ordered, controlled layout. And it drove us all crazy. And you know what? Despite all the great progress made in the last 15 years – web standards, inclusive design, UX, semantic thinking etc. – when you really think about it, as designers we haven’t really grown that much. Or rather, we’re still trying to force what we know into a medium that it doesn’t quite fit. Our practice of creating designs for the web is still mired in the great thinking that was done over the last 800 years. But who can blame us? 800 years of baggage is hard to get rid of! But that’s what we need to start doing; we need to start thinking in a new, and different way.
‘There is no spoon’
For print design, the ‘page’ is the starting point for creating your layout. The proportions define the grid within. The content is bound the book for pleasing effect. The ratio of the page is repeated in the smaller bodies of text for a feeling of connectedness when the reader moves from one page to another. For print design, the process of designing grids, and the layouts that sit on top of them, is a process with one fixed and knowable constraint: the Page. On the web, however, there is no page.
Consider the browser for a moment. The browser is a flexible window into the web. It grows and shrinks to the users screen size. The user can move it, stretch it, scroll it. The edges are not fixed. It is not a page, but a viewport.
Let us pop back to 1997 again. I’ve just been asked to design a website for a new art & architecture gallery in Manchester. The project is an exciting one, with some great potential for some really creative design and layout work. Typographically, it was a bit of a dream project. I’d been involved in the branding, the logotype design and the design work for the publications. I’d designed a grid system that would work across all of the media from flyers to signage – a kind of universal grid with the proportions of the logo as its starting point.
The time came when I had to knuckle down and design the web site. I started the design, as I usually did, on paper. Then Photoshop. Then Dreamweaver (trying to avoid looking at the code it produced – not because I was purist, but because it scared me to death!). The website design I created was based on a fixed width, fixed height modular grid. It had it all: ambiguous navigation, hidden content, images instead of text, all created with tables. It was cutting edge. But I know now, I hadn’t created a website, I’d created a brochure that happened to be on screen. I knew then, as I do know, that it was wrong. What I’d created had no place on screen at all – the wrong design for the medium. Instead of trying to understand the web, and the browser, I’d taken my existing thinking and shoe-horned my controlled design into it.
Now, let me ask you a question. If you replace Photoshop with Fireworks, Dreamweaver with Textmate, and tables for layout with Web Standards, how many of you are still designing this way? How many of you are still thinking of pages and edges? It’s ok. I am still, too.
800 years of baggage is hard to shed. There’s a lot of engrained thinking. Thinking that is, in fact, really great design and compositional theory. But, because of the attachment to the fixed page, we’ve not accepted the web for what it really is: fluid, chaotic, unordered, open. On the web, there is no page.
If there’s no page, no thing with edges, then how can we begin to define a grid? One of the goals – as described by Jan Tschichold – was to create a layout that is bound to the book. How can we bind anything on the web if there is no page from which to start? I propose we stop trying to find the browser’s edge. We stop trying to create a page where there isn’t one, and we welcome what makes the web, weblike: fluidity. We start creating the connectedness Tschichold talked about by looking at what is knowable; our content.
It has been said that as web designers, we’re not designing around content, but rather we’re designing places for content to flow into. Particularly in large organisations, it was commonplace for designers not to know what the content is, or would be, but rather, at best, they’d have some idea of how the content would break down. At worst, they wouldn’t have a clue and basically guess. ‘Oh, this is an article page, so it must have a bunch of headings, some body copy, some lists’. Feel familiar?
Separation of content and presentation is the mantra of the Web Standards movement. So you may think this disconnect started when the web standards movement was in full flow, but it started way before then. I look back at when I worked in web design agencies in the early 2000’s and, as a designer, I was off in my little corner designing the three templates that the client was paying for, and that the Information Architect had defined. I had wireframes of these exemplar templates and was pretty much following them the number. What I was doing was designing in the dark. I had my blinkers on. I had no idea what the content would be in 6 months, 1 year, 2 years time. In fact, I’m pretty sure the client didn’t, either.
What I was doing was designing a box. A straight-edged, inflexible box that couldn’t grow with the content as it didn’t take into account existing graphical assets, either. Thankfully, skip 10 years to the present day and things are getting better. We have content strategy. We have a relatively mature UX industry. As designers, we’re in a much better position to know, not just what the content will be right now, but what it will be in 1,2 ,3 years time. Now we have this knowledge, we can use it to our advantage: by using it to create our grid system.
A NEW CANON
I’ve talked about baggage. Hundreds of years of thinking in the same way: canvas-in. We’ve taken grid design theory from the world of the physical page and tried to make work in a medium where there are pages with no edges. A medium where the user is able to manipulate the viewport. Where context matters – is the reader sitting at the TV, a desk, using an iPad or hurredley walking from one meeting to another snatching some news on the way on their mobile device. Where do we begin to design on these shifting sands and still retain the reason for using a grid system? Before I get on to that, let’s remind ourselves what those reasons are:
Grid systems help connect or disconnect content. They help people read and aid comprehension by chunking together similar types of content, or by regular positioning of content, they can help people navigate the content. Connectedness helps brands tell a consistent story in layout.
Help solve layout problems
We all need answers to layout problems. How wide should a table be? What should the whitespace be around this boxout? Grid systems help us with that with predefined alignment points.
Provides visual pathways for the readers eye to follow
A well-designed grid system will help use whitespace dynamically and in a powerful way. By filling a space with negative space instead of content, we can force the direction of the readers eye more effectively than anything else.
As with the printed word, the word on screen would benefit from some rules of form; a new canon of page construction for the web. A way of constructing harmonious layouts that is true to the nature of the web, and doesn’t try to take constraints from one medium into another. That starts with the content and works out, instead of an imaginary fixed page and working inwards. I’m going to repeat that, because it’s important: we start with the content and work out. Instead of starting with an imaginary fixed page and work in.
The new canon can be best described as an approach. A series of guidelines, rather than a single diagram to be applied to all. This first part of the canon are a series of design principles to adhere to. These design principles were created to provide a simple thought framework, an Idea Space; a set of guiding principles to be creative with.
Define a relationships from your content. If none exist, create some.
A grid for the web should be defined by the content, not the edge of an imaginary page. Look to your content to find fixed sizes. These could be elements of content that is out of your control: syndicated content, advertising units, video or, more commonly, legacy content (usually images). If none of those exist, you must define some. Make some up. You have to start somewhere, and by doing it at a content level means you are drawing content and presentation closer together.
Use ratios or relational measurements above fixed measurements.
Ratios and relational measurements such as pixels of percentages can change size. Fixed measurements, like pixels, can’t.
Bind the content to the device.
Use CSS media queries, and techniques such as responsive web design, to create layouts that respond to the viewport. Be sympathetic to the not only the viewport, but to the context of use. For example, a grid system designed for a small screen should be different to that intended to be viewed on a laptop.
By using these principles to design to, we’re drawing closer relationships between our layout, content and device. Tschichold would be proud.
– This blog post is an excerpt from my upcoming book on designing grid systems for the web, published by Five Simple Steps.
As usual, it started on Twitter and a flurry of blog posts are cropping up this week to fill in the nuances that 140 characters will not allow. So, here’s mine…
[Aside: I did actually make a promise to myself that I wouldn’t get involved, but, I find that cranking out a quick blog post, gets my head in the space for writing generally.]
I started speaking at web conferences in 2006. After attending SXSW the year before, I proposed a panel discussion (with the lofty title: Traditional Design and New Technology) with some design friends of mine: Khoi Vinh, Jason Santa Maria and Toni Greaves and moderated by Liz Danzico. I was terrified. But, in the end, it was fun – there was some lively debate.
I wanted to do a panel at SXSW since seeing Dave Shea, Doug Bowman and Dan Cederholm sit on a CSS panel at SXSW in 2004. Not because I saw the adulation, but because I saw – for the first time – what it was like to contribute to this community. To be part of it. To give back: be it code, techniques, thoughts, debate or discussion. And I wanted a part of it. So, that’s what I did. I started blogging – I felt I had some things to say, about typography, grids, colour theory. All of the traditional graphic design stuff that mattered to me. Not because of some egotistical trip, but because I genuinely wanted to make things better. Trite, I know.
Fast forward a couple of years and I’m speaking at the inaugural New Adventures conference in Nottingham for my friend Simon Collison. On that day, every speaker up on stage was trying to give the best talk they could. Not because of the audience, not because of who they were, but because of Simon. It was personal.
The talk I gave at new Adventures took about two years to write. Yes, it took me that long to write a twenty five minute talk. You throw that into the equation, a high-risk personal favour for a good friend, plus my family and best friend in the audience, and you’ve got a recipe for bad nerves and vomit. And I did vomit.
But, I got up there. Cast aside the nerves and held my head up and spoke for twenty minutes on things I’ve been thinking about for years. It was received well. Afterwards, all I did was sit in the green room for about two hours and didn’t really speak to anyone.
My natural preference
It may surprise you that most speakers I know are not extroverts. I don’t mean extroverts in the way you may think, either. I mean it in the Myers Briggs type: they are not the type of people who gain energy from other people, they gain that energy from themselves. I’m one of these people, too.
Being on stage is firmly out of my comfort zone. Firmly. I’ve had to learn to harness the nerves and put them to good use. A good friend of mine calls this ‘peak performance’ – the nerves help you bring your ‘A’ game.
My natural preference is to be on my own, working. Either thinking, sketching, writing, building, exercising… whatever. All through my life, I’ve enjoyed solitary sports and pastimes, from angling to cycling. Now, that’s not to say I’m a hermit. I’m not. I’m pretty sociable when I need to be, but it’s not my preference. So being on stage – sticking my head above the parapet – takes incredible effort, and then afterwards, I generally want to go and hide in a corner for a bit. It wipes me out.
So, why do I do it? Why does anyone do it in this community? If you’re a regular speaker, or your first time? Almost everyone I know does it because they want to give back. They have something they’d like to share in the hope it may help someone else in a similar position.
This brings me full circle to my opening sentence. Why, then, knowing all of this, is there a general feeling of discontent in a vocal minority that speakers who do this regularly are:
In it for the ego
Not doing any real work
Not leaving room for new speakers
I’d like to address these points from my own experience.
In it for the ego
Why would someone get up on stage and speak to hundreds of people? Sure, some may get a kick out of that. People applauding you feels nice. But, let’s be clear: that’s not egotistical. That’s being rewarded, and there’s nothing wrong with that.
Not doing any real work
I wrote about defining ‘work’ last week. I see speaking as part of my work as a graphic designer. If you’ve studied graphic designers, art directors, ad copywriters and the like, you’ll know that a lot of them speak to their peers – either at conferences or through publications. Writing and talking about what we do with each other is work. Not only that, it’s fucking important work too. Without it, there would be no web standards, no open source, no progression.
Not leaving room for new speakers
Experienced speakers leave room for everyone. Experienced speakers do not run conferences: conference organisers do. And conference organisers need to put bums on seats. Just like a big music festival, you need the draw, but also you need the confidence that a speaker will deliver to the audience. Every experienced speaker I know works damned hard to make sure they deliver the best they can, every single time. They’re professional. They don’t screw it up, or spring surprises. They deliver. And that’s why you may see their faces at one or two conferences.
A couple of months ago, I saw Heather Champ talk at Web Directions South in Sydney. Amongst many hilarious – and equally terrifying - stories of how she’s managed and curated communities over the years, she came out with the nugget:
Will you tolerate a conference circuit swamped by supposedly the same speakers and vote with your wallet? Or will you tolerate conference organisers being continually beaten up for genuinely trying to do the right thing? Will you tolerate speakers being abused for getting on stage and sharing their experiences?
Will you tolerate harassment, bullying and exclusion?
As I’ve said before, Twitter is like a verbal drive-by. It’s fast, efficient, impersonal and you don’t stick around for the consequences. Let’s stop it.