Digest — a new magazine


I’m excited to hear that early preview copies of Digest are landing on lucky readers’ doorsteps. Digest is a new A3-sized magazine by Elliot Jay Stocks and Keir Whitaker, described as “a seasonal digest of culture that surrounds the world of the creative professional; a lifestyle publication that provides inspiration for the discerning traveller; a reminder that the world away from our computers is ripe for exploration.”

The best part is that Elliot and Keir kindly invited me to contribute an article to the debut issue, so I penned a few words about my favourite places in New York City; a city a love and visit often.

Head over to the magazine’s site, where you can buy the magazine either in print or digital form.

The new Creative Journal


Very quietly last week, I pushed live the new version of Creative Journal, which is my blog in which I squirrel away all non-web design inspiration (for web design work, there’s siteInspire).

It’s not quite finished, and probably will always be in perpetual beta, but it is in decent enough condition for it to be let loose.

The motivation for the redesign was actually technical: I began to hate the way I had set up the old site in ExpressionEngine, and posting became a chore. So I redeveloped it in Rails, and gave it a bit of a facelift while I was at it. I have further plans for it too, which would necessitate a more custom application.

Take a look, and I hope you like it.

Remote working and corporate culture

I’ve just returned from holiday and quickly caught up on the week’s news, one piece being that new Yahoo! CEO Marissa Mayer has banned remote working.

A great deal of commentators have been up-in-arms about the decision, including David Heinemeier Hansson whose company 37signals is predominantly staffed by remote workers, and who has written a new book extolling the virtues of remote working.

I don’t disagree with any of the points made here: there are incredible up-sides to allowing employees remote working, but most commentators are missing the point that for remote working to be successful, a company’s culture and vision has to be extremely clear and more importantly, shared by every employee.

If it isn’t, remote working can just become an excuse to slack off or take some free time off work.

Many moons ago I worked at Accenture, which is name-checked in DHH’s article as a stodgy firm that is embracing home working (incidentally, I don’t believe that any Accenture manager would ever condone home-working; it’ll just be a HR blurb written in their corporate literature). I—and almost exclusively all my contempories—disliked working at Accenture. There were no shared values, and nobody respected whatever vestige they had of a unified corporate vision. The culture was one of work hard, and then work a bit harder (all the while spending your time away from home on the road), so when we “worked from home” (complete with the sarcastic finger quotes), we didn’t.

I’m not ashamed to say I didn’t do anything productive when I worked from home. I checked my email for 10 minutes to check-in and give the impression I was active, but otherwise, it was movie-time in my pyjamas. I simply couldn’t give a damn about Accenture, so given the chance of slacking off for a day, why would I do any work? (Yes I know that is an unprofessional attitude, but I was younger then.)

I would bet that this is exactly the sort of culture at Yahoo! right now.

Mayer taking over the helm is a great move and I’m sure plenty are buoyed by her being there. But if my experience at a company as stodgy as Yahoo! is anything to go by, I’d guess that the majority of the people who are remote working at Yahoo! don’t give a damn about the company either, and are slacking big time. Indeed, the data reportedly seems to suggest they are.

It takes a long time for a strong corporate culture to form unless it’s baked-in from the outset like at 37signals and Github (another company that works extremely well using a remote workforce), and a company as battle-worn as Yahoo! will take a long time for it to heal. Only when it rebuilds and strengthens their culture so that people feel great about being there again will they be able to deploy a remote workforce who are motivated and self-disciplined enough to care.

Talking and writing about ideas gives them form

I think we’re all agreed that not talking about your ideas and slapping an NDA all over them is never good advice.

Chris Dixon eloquently talks about this in “Why you shouldn’t keep your startup idea secret”.

I would go further and encourage you to talk to absolutely anybody about an idea you have listing in your head, without shape or form.

Talk to people who don’t know your industry, or who don’t know about problem you’re trying to solve. If you catch yourself waffling, or searching for the right words to convey the idea, it needs work.

Email people you normally tweet with out of the blue. They’ll welcome the surprise, and pay attention.

Write about your ideas publicly on your blog. If you’re feeling particularly ballsy, light the touch-paper and stick that blog post on Hacker News. (Make sure you stand well back.)

An idea only grows with a thousand bumps and bruises: rebuttals, criticisms, “it won’t scale” and plenty of “that already exists”. Afterwards, the idea might still be unclear, will take on a sharper silhouette.

Then it just takes execution, but that’s a whole other story…

Avoiding the medium-sized stuff

I don’t really post links on this blog these days, but once-in-a-while an article comes along that I feels deserve more attention than a quick link on Twitter.

Connor Tomas O’Brien write an enlightening post about avoiding medium-sized stuff:

The small stuff is okay, too. Tweets and Instagram photos and Vine clips – stuff that’s easy to create and easy to digest. The small stuff can stay. The small-scale stuff is fun.

The medium-sized stuff: projects that don’t really mean that much to me, but that take more than a trivial amount of effort to get finished. Recently, I’ve been taking on a lot of projects with timeframes measured in days or weeks. That’s not long enough to do anything very interesting. These projects are not horrible to work on, but, were I to pan out and see my life on the scale of years or decades, I realise it’s these particular projects that I’ll end up forgetting, these particular projects that will lead me to wonder, “Hey, what did I actually do over this year and that?”

Big projects are scary. It’s much more tempting to take on a bunch of medium-sized projects than one huge project, because in doing so you mitigate the chance of failure. But it’s the stuff that could fail that’s the stuff we remember, not the stuff that’s safe.

I’ve been thinking a lot about this recently and I’m thankful to Connor for giving this thought a shape and meaning.

The launch of Steer: learn design and development


I’m very excited to see that Rik Lomas’ new start-up—Steer—has launched. Their goal is to set up high-quality courses in web design and development. Some of their courses are aimed at people pursing a new career, whether it be creating their own web startup or wanting to become a designer or developer, and other courses are aimed at giving people in the industry extra skills to improve their craft.

Some of their courses are long form multiple day courses, for instance, they have a course called “Learn to make websites” coming up in April that’s Monday to Friday, 10am until 6pm every day for 10 days. Another interesting product is a series of courses called Slow Code: sessions every fortnight for people who want to learn more casually or just don’t have the time to commit to an intensive course. They’re launching with Ruby on Rails and jQuery Slow Code courses.

Their long term goal is to create a platform for learning during and after the courses so that their students can always get help solving problems. There are a few education startups that help do this, but importantly Steer are providing the educational infrastructure than will underpin the platform; tailoring it precisely to their students’ needs.

The staff at Steer—who, incidentally all code themselves—know the pain of learning how to program, so in parallel with the main platform they’re building web apps to make creating websites easier. Their first app, Teepee is for people looking to get static sites online easily. They say, “Think of it as a cross between Geocities and a super simple web host. We’re working with Code Club on this so it has to be simple enough for a 9 year old to use which in itself is a big but fun user experience challenge for us.”

Finally, Steer are donating money from each course to Code Club, to help their programme that helps kids learn to code. For instance, every ticket bought for their 2 week learn to make websites course will fund 12 children to code for a year. “We love Code Club and it’s amazing to see how much enjoyment children get out of creating their own inventions. That sense of fun and achievement is something that we want to replicate with Steer.”

Given the woeful state of design and development education in the UK and the US, I’m excited about Steer, and genuinely believe that it will be startups like this that will carry the mantle: I haven’t seen any innovation by traditional educational institutions and can’t envisage a future with them playing a significant role in educating the next generation of talent in our industry.

“What Constitutes Good and Bad Web Design?” — a response

I recently enjoyed this article by Alice Rawsthorn in the New York Times. As an industry we seldom get a mention in the mainstream media, so I welcomed commentary.

Here’s a pithy summary:

  • A huge number of websites—particularly those of major art galleries like Tate and Centre Pompidou—are incredibly confusing to find even the most simple information.
  • She levies the fault of this at the web designer’s feet: “Shoddy Web site design is a curse of modern life. The more dependent we have become on the Internet for information, the likelier we are to suffer from its design deficiencies.”
  • A well designed site is “fulfilling its intended function efficiently and engagingly”, but “dispiritingly few sites manage to achieve it. A common mistake is to prioritize style over substance.”
  • The principal problem with many Web sites is that, “their designers were neither rigorous nor imaginative enough in planning the way we will navigate them”
  • She uses the example of Quo Vardis as an example of a well designed site that fulfils its function
  • Then, she uses the Milwaukee Police News website of a site designed to convey the complex and time-sensitive nature of its content well.

I agree with almost everything she says, but it’s everything that is unsaid that is missing, and which makes me uncomfortable with the piece. Of course it’s tricky to get into the minuatae of a website’s failings in a mainstream publication, but at the very least I want to explain why the web designer shouldn’t be blamed.

Let’s take a typical art gallery site as an example. I might not particularly like the site in its entirety, but I could love its design. In the context of the article, that could seem like a contradiction. It assumes that the site is a failure because it has been designed badly, and in many people’s minds design equates to style. In fact, there so many pieces of the website machine that can fail, which can have a devastating effect on the overall experience.

Firstly, it’s no secret that the last organisations to enjoy the cutting-edge in content management systems (CMSs) are those in the arts. Either there is no funding available, or they are locked into multi-million dollar government contracts with behemoth IT companies whose systems are held together with string and sticky-tape. Designing for antique systems is a challenge that I wouldn’t want to wish on anyone. Your beautifully crafted designs—and even code—can get mulched into a hideous mess when mangled by such systems.

Then, because the CMSs aren’t a pleasure to use, the people who are responsible for updating it have a hard time adding content. If it’s hard to add content, missing connections start to creep in and the whole experience is ruined. I don’t think I’ve ever enjoyed browsing the V&A site without stumbling into a 404 error page, for instance. Because adding content is a miserable experience, why would they even bother creating—or commissioning—joyful, well produced content, if it means a world of pain to publish?

Suddenly there’s a chain of circumstances that lead to website mediocrity, and yet designers only form a small part of it.

I’ve not met a fantastic web designer who also has the ability to structure and model content as well as a fantastic information architect can. Yet I’ve learned of projects where there has been little input from someone who can see the broader picture and goal of a complex site. In these cases the web designer is working from briefs that usually haven’t been validated by some solid architectural thinking. By the same token, bad structural thinking can only lead to messy design; but that’s hardly the fault of the designer.

Underpinning all this is the technical infrastructure. Sites have to feel punchy and quick for a great experience, and so many either don’t have the resources to achieve this, or their hosting team isn’t up to the job. Tate themselves felt this sorely with the Kraftwerk fiasco, a few weeks ago.

So, let’s take a look at what’s happened with all the cogs in the machine so far: the design agency who created the site is upset because their designs have been mangled beyond recognition. They could also be upset because they’ve worked from client briefs without any solid architectural plans in place. The content team are frustrated and demotivated because their CMS is little more use than a typewriter. The hosting team are frustrated they don’t have the money (or the will) to make things faster.

And yet it’s the web designer who’s suffering the blame.

Finally, I’d like to question Rawsthorn’s use of the Milwaukee Police News as an example of a great site. Absolutely, it’s a technical feat and compelling. But to me, it looks like a promo for a new cop drama starring Damian Lewis. And—while I haven’t tried; I’m only assuming here—I’d like to see how it works on Internet Explorer 6.0, which is exactly the sort of browser that someone using “older, cheaper machines with slower Internet connections” she refers to earlier in the article have at their disposal. The drama of the site will soon be lost.

To genuinely appreciate design for function, I would have liked Rawsthorn to reference the work of the team. They have done a lot of thinking around how the many millions of users in the UK need to access content, on any machine, and in any way possible. This is only successful due to the close-knit team of designers, developers, information architects, user experience professionals, and technologists, that have come together to make a speedy, simple, clear message, using design as it is intended: to make the site effective and functional.

Ultimately, a well designed website is the sum of its parts. The parts tend to be invisible to its visitors other than the visuals that are the end result of a long and complex process. The outcome is that the designer often gets the blame, and that’s a sad message to read in the mainstream media.

A few things I learned from redesigning and redeveloping siteInspire


The new siteInspire is live, and I’m pretty exhausted.

I started the process almost a year ago, which is an insane amount of time full of ups and downs. So I thought I’d pen a few notes as to what I learnt from the process.

I should also caveat this entire post with the fact that I realise a site that showcases other sites isn’t exactly the most humanity-transforming idea nor is the most complex. But with daily visitors creeping into the tens-of-thousands a day, it seems to be something that people like and so there’s a duty to do a decent job, and ultimately not to screw it up spectacularly.

The redesign went through approximately ten radically different looks, each a knee-jerk rebound from each other and—as a result—I felt each iteration was poorer than the last. In the depths of disillusionment and directionless-ness, it took on the look and feel of a vintage, hipster, restaurant menu. Ugh. That “friendliness” and “charm” was replaced by stark coldness. Ugh. Towards the end I became almost blind to what I was doing and trying to achieve, which was when I had to seek the advice of others before I gave up.

In the end, the site doesn’t look dissimilar to the old version: the same(ish) palette, same typography. There’s nothing clever about it, it’s just a small iteration from the original. Yet paradoxically it took a long time to get to a point where it just felt right.

In true piece-of-advice-blog-post fashion, here are a few take-away bullet points:

  • If the design isn’t that broken, don’t throw the baby out with the bathwater. All it probably needs is a little attention and tweaking. Keep the goal simple.

  • If you ask 10 different designers for an opinion, you will get 10 different, polarising opinions. That can only lead to heartbreak and confusion, so be choosy who you seek for advice and be careful when you post on open forums like Dribbble: some feedback I’ve seen for most shots that I have either liked or disliked has been unfathomably bonkers.

  • Seek advice of only those whose work you genuinely respect and whose work you admire, otherwise you might as well ask some guy at a coffee shop who’s wearing cool headphones.

  • Related to this, seek advice only from those who have done similar projects to what you are doing. If you’re creating a web app, reach out to people who have worked on a similar thing and who understand the challenges in both designing and developing one. Designing a product is very different to designing and building something.

  • If someone really, really hates your design, it just means they care. Of course, saying something like “it’s shit” isn’t constructive, and they’re probably a bit of a dick, but don’t take it to heart however hard it must be. This is why people get so angry when loved products like Twitter and Facebook are redesigned. Haters are gonna hate, but haters care a lot.

  • You need focus. Spending a year on a redesign and redevelopment project is totally impractical and is full of waste. Sprint to the end as fast as possible whilst still taking care, otherwise you’ll lose motivation and focus. This is difficult with personal projects because there is no client to impose any deadline, so try to set one. (Mine was actually Christmas Day, but that came and went, what with the all the food and booze.)

  • Try and ditch Photoshop and Illustrator if you’re making a web app. siteInspire is hardly a complex design but I didn’t touch either once apart from creating assets. Having no training in design, they feel like such old fashioned tools to me, and there’s a lot to be said for just diving in and creating everything in HTML and CSS from the outset.

  • This isn’t related to this project in particular, but if you want to learn development the very best way to do it is on your own projects. It’s a place where you can make all sorts of mistakes and experiment. I learnt a lot more about Rails, Compass, and responsive development.

Finally a big thanks and warm hugs to Shelby, Allan, Simon, Al, Lawrence, and Rik for all their advice. They’re awesome and you should follow them.

Now for the next challenge, to re-design this blog, the studio site, Creative Journal, and a new, not-so-top-secret project that I’ll talk about soon…

Verbose content discovery: adding the human touch to aggregation

I’m a voracious consumer of written content. Nothing on the web makes me happier than filling my Instapaper bucket with shiny pebbles. It doesn’t matter that I might not end up reading them all, but I will try.

Finding the pebbles is a hard thing that has been made astonishingly easy—too easy, for reasons I’ll explore—with the recent rise in aggregators that span the automatic to light-touch. The services I use regularly to help navigate content are the following.

There are many more services that pop up with every other Techcrunch post, all attempting to snare morsels of rarefied attention in trying to solve the problem of delivering the most compelling content to you on an Internet that is unfathomably overcrowded with stuff.

Here I’ve deliberately ordered the services in terms of how transparent they are in sourcing the content that they feel is most relevant to you. replaced the excellent Summify for me, which disappeared once it was bought by Twitter, and which powers their new Discovery features. Taking your Twitter friends as its source, it attempts to digest all the most relevant links, and sends a daily email telling you what to read. Each link is suffixed by the faces of your Twitter friends who shared the piece of content.

By the same team behind is the new Digg uses metrics that are far outside of your control to determine what to show. It doesn’t take into account who you follow on Twitter or who you are connected to on Facebook; it aggregates general consensus. This is an important distinction between it and the old Digg. In the old Digg, you invest time in trusting the community to vote up relevant articles, yet new Digg assumes you to trust the entire internet’s social community in determining what is good (which, as Twitter makes very transparent, is something you shouldn’t assume).

It now turns out that is due to be superseded by Digg: personalised aggregation to be replaced by automation.

The part that is missing in these services is the why. The act of sharing a link in itself is stripped of nuance. Someone might have shared a link because it was the most extraordinarily wonderful piece of writing, as much as because it was the most despicable, bigoted piece of writing ever committed to a blog. Yet this nuance is lost, and it’s the nuance that I miss.

This loss was addressed in The Filter Bubble by Eli Pariser: an excellent book if you like pop socio-/psych-/techn-ology books. This passage resonated with me:

David Gelernter, a Yale professor and early supercomputing visionary, believes that computers will only serve us well when they can incorporate dream logic. “One of the hardest, most fascinating problems of this cyber-century is how to add ‘drift’ to the net,” he writes, “so that your view sometimes wanders (as your mind wanders when you’re tired) into places you hadn’t planned to go. Touching the machine brings the original topic back. We need help overcoming rationality sometimes, and allowing our thoughts to wander and metamorphose as they do in sleep.”

In an era of computational aggregation, how do we re-introduce human touch?

An exciting new services that helps bring back verbosity and nuance is It offers you content that your friends are reading right now. It doesn’t matter whether the piece was good or bad, or whether it makes you appear cool or dull. While it demands that your friends use the bookmarklet to mark what they are reading, the output is a very comprehensive set of articles that are likely to interest you, since you’re interested in your friends.

I don’t use it in the way it has been designed. I read all my content offline on Instapaper, and so I can’t share what I am reading easily. And while it has a commenting system baked-in, since I rarely read on screen I’m not compelled to enter into the conversation. I consume the content it outputs yet I’m very aware that the content behind it is being shared by just a handful of people I respect. is a step towards what I consider the future of curation and aggregation. No amount of natural language analysis or computation can analyse deeply personal taste, quality, nor provide context or meaningful links between content (at least not yet) and so, we need to build platforms and services that are high-touch. So—perhaps paradoxically—we will increasingly rely on editors to help us navigate the web.

I keep thinking about the ratio 100:9:1. (I can’t remember who coined it or referenced it, so if you know please let me know.) It refers to there being one creator of content, nine people who share/curate/edit, for the one hundred consumers. The internet has helped us build platforms for each segment of this model of consumption: readers have Twitter and countless other ways to consume, and of course creators have extraordinary tools at their disposal. Yet the same develops can be said for the nine editors: tech has made us jump to explore computerised, automated curation, without considering that a more valuable proposition might exist with the piece in the middle which gives people who are considered sharers of good content a democratic editorial platform.

I don’t know what this platform might look like yet, but I’m excited to explore it further to bridge the gap between noisy Twitter and the relative calm of a traditional editor. I’d assert that verbose human/high-touch content discovery is something we should strive towards to help us find fewer but shinier pebbles.