Polytechnic

The personal brain-dump of Garrett Coakley

Archive of: web development

  • A launch of sorts

    It’s been over a year since I joined WDCS, and to coincide with the 64th International Whaling Commission, you can now see a small piece of what I’ve been working on.

    Say hello to dev.wdcs.org.

    Over the next fortnight, our team will be busy posting news and blog posts from the IWC, offering us a great opportunity to ensure that the structure, work-flow, and multilingual nature of the new site all work as we hope. And in what I think is a brave move by everyone here, we’re doing it in the public eye. If GDS can do it, well so can we (seriously, the GDS team has been, and continues to be, a massive inspiration).

    It’s a bit shabby round the edges: the theme is bland and temporary but hopefully very readable, there are some bugs in the internationalisation framework that I’m still fixing (“Ray, when someone asks you if you want to build a multilingual site, you say ‘NO’!”), and you might find yourself in a bit of a navigation dead-end if you follow the wrong link.

    As I said, it’s a bit shabby, but it’s our shabby.

    If you’re interested in the background of what led us to this point and where we’re going in the future I gave a talk to the Oxford UX group a couple of months ago that fills in some of that.

    Thanks to everyone at WDCS who’s been helping on this project. As a web geek you couldn’t ask for a more passionate and committed bunch to build for.

    And now, some sleep.

  • Rebuilding from the ground up

    Last Wednesday (25th April) I gave a talk at the UX Oxford Speaker Series about the work I’ve been doing at the WDCS over the past year, and why to my friends it seemed like I had vanished off the face of the earth.

    It’s a wide-ranging talk, looking at the problems with the current site, our initial research and findings, and the the content-out/responsive approach we’ve taken towards the redevelopment.

    The talk itself lasts around a half-hour, with another half-hour Q&A session afterwards.

    The books that I reference during the talk are:


    And the list of useful links and further reading can be found on my Pinboard account under the tag uxoxford2012.

    Thanks to UX Oxford for inviting me to speak, it was great to finally show everyone what I’ve been working on and I really enjoyed the questions after.

    Fingers crossed we’re due to go into public beta sometime in June.

  • Egalitarianism and Progressive Enhancement

    Progressive enhancement follows the Golden Rule because it is concerned with the “other”. That’s why accessibility is such a key part of building websites following the progressive enhancement philosophy. It’s about putting yourself in someone else’s shoes—someone whose abilities and situation probably differ from yours. We are a diverse lot after all.

    One hell of a read from Aaron Gustafson on the Golden rule, egalitarianism and the philosophy of progressive enhancement in web design.

  • Done is better than perfect

    Done is better than perfect, or “the best” is the enemy of “the good”. Perfectionism is a form of procrastination. It assumes that time is an infinite resource, that other tasks can wait while you add “just one more touch” and that “perfect” is attainable.

    One of the guiding principles behind Shopify’s apps team. Great advice for any dev team.

  • Campaign Monitor is hiring and now you want to work there

    This is how you design for interest. Wonderful site.

    Via Denise

  • Making content harder to access

    Jeremy Keith on the proliferation of hash-bang URLs and the recent Gawker redesign:

    I’m always surprised when I come across as site that deliberately chooses to make its content harder to access.

    He links to this great post by Mike Davies on just how badly this new anti-pattern – which was only ever supposed to be a band-aid for sites that ignored best practise and couldn’t get indexed by Googlebot – is breaking the structure of the web, and led directly to the site outage affecting all Gawker properties on Monday.

  • Why is Dropbox more popular than other tools?

    Speaking of simplicity.

    Isaac Hall, co-founder of Dropbox competitor Syncplicity, answered the Quora question “Why is Dropbox more popular than other tools with similar functionality?

    It’s a wonderfully in-depth answer, encompassing PR companies, the press, and how to structure a beta, but despite all of those influences it came down to this:

    In the end, it really came down to one incredibly genius idea: Dropbox limited its feature set on purpose. It had one folder and that folder always synced without any issues — it was magic. Syncplicity could sync every folder on your computer until you hit our quota. (Unfortunately, that feature was used to synchronize C:\Windows\ for dozens of users — doh!) Our company had too many features and this created confusion amongst our customer base. This in turn led to enough customer support issues that we couldn’t innovate on the product, we were too busy fixing things.

    […]


    If you’re starting a new company, the best thing you can do is keep your feature set small and focused. Do one thing as best as you possibly can. Your users will beg and beg for more functionality. They will tell you their problems and ask you to fix it. My philosophy is that they’re right if their feature request is right only if it works for 80% of your customers. Until you have a lot of resources, stay focused on your core competency.

  • Keeping it simple

    Rands In Repose interviews the creator of Instapaper, Marco Arment.

    When asked about the early decisions he made in order to manage as a one-man shop:

    The biggest design decision I’ve made is more of a continuous philosophy: do as few extremely time-consuming features as possible. As a result, Instapaper is a collection of a bunch of very easy things and only a handful of semi-hard things.

    Keep it simple, iterate.

  • Matt Mullenweg on shipping 1.0

    There is some great advice in this piece from Matt Mullenweg about shipping version 1.0.

    Usage is like oxygen for ideas. You can never fully anticipate how an audience is going to react to something you’ve created until it’s out there. That means every moment you’re working on something without it being in the public it’s actually dying, deprived of the oxygen of the real world.

    You can come up with all the user stories, scenarios and use cases you want, but the moment your product hits the real world, all that goes out of the window.

    And never wait until something is completely polished and feature complete.

    …if you’re not embarrassed when you ship your first version you waited too long.

    Amen to that. Iterate, iterate, iterate.

  • dConstruct 2010

    This is a longer post, longer than I anticipated, so the tl;dr version is:


    On with my ramble.

    It’s taken me a while to get around to writing up these notes. In the past I’ve just thrown up the bullet points and then left it that, but this time I wanted to let things percolate for a while.

    A nightmare train journey down from Oxford meant that I didn’t get into Brighton until gone 10pm and had to skip the pre-party, so for the first time in four years I didn’t have a hangover going into the opening talk of the day, The Designful Company by Marty Neumeier.

    Marty is the author of The Brand Gap (a book that it turns out is rated very highly by people I admire) and talked about the process of design, innovation, and what it means for your brand. When talking about competition he highlighted that most competition doesn’t come from other companies per se, but from the clutter that surrounds us every day. Whether that clutter be advertising, products or brand choice, you have to differentiate yourself through design.

    Next up was Brendan Dawes with Boil, Simmer, Reduce. This is the cooking metaphor for the process he goes through when creating (although as he admitted himself he’s not the best in the kitchen). The boil is your chance to collect things. Ideas and objects that will inspire. After the boil comes the simmer, let things stew for a while, look out for connections. Then the reduction starts, where you hone in on a solution.

    This is the point where you start to throw out the superfluous and your solution takes shape. Although don’t be too quick to throw things away. An example of this being the elastic scrolling in Apple’s iOS. It serves no real purpose apart from delighting the user (cf: the Dopplr guys talk at dContruct 2008), but like Brendan, I have spent far too much time playing around with it and it never fails to elicit a smile.

    He left us with an encouraging truth. We, here, now, with the technology and tools available to us, are in the unique position of being able to make anything we want.

    Luckily I’m way behind on TED videos as David McCandless announced that a version of the talk he was about to give, Information is Beautiful, had been posted on the TED site a few days before. Not having any real graphical chops myself I was fascinated by the thinking that goes into visualisations, and why a good visualisation can be so powerful.

    They connect visual relationships, that our eyes find beautiful, to conceptual relationships, which our brains find interesting. Data is a creative medium, and in a connected world unconnected numbers make no sense. Visualisation can help us see patterns that connect the data.

    Before the lunch break came Samantha Warren on The Power & Beauty of Typography. This was a call to arms to embrace 2010 as the year of typography on the web and get away from the standard set of web fonts through the use of technologies such as FontDeck and TypeKit. I have to admit that Samantha’s talk didn’t really help me with my biggest typography problem, which is that I have no real understanding of why certain fonts work or don’t work together. I get that a good use of typography can add emotion and personality to a site, but I’m always wary of going too far and making a complete hash of it.

    Still, it’s encouraged me to do some more reading about the subject, so if anyone has some useful links then please drop them in the comments.

    After lunch it was John Gruber’s first talk in the U.K. with his Auteur Theory of Design. A sober thesis on creative endeavour and the tendency for those involved to trend towards the level of taste of the decision maker.

    In other words: “Benevolent dictatorship is more effective than democratic consent”. From my experience the same thinking applies to community management

    Unsurprisingly, given his well known love of Stanley Kubrick he used this quote to illustrate his point:

    One man writes a novel.
    One man writes a symphony
    It is essential for one man to make a film.

    All in all a very thought provoking, meditative talk that was perfect to ease us into what was to follow.

    Hannah Donovan’s What Improvisation Can Teach Us About Design started off with a beautiful improvised musical opening played by Hannah, Jeremy Keith and Matthew Ogle.

    Drawing the threads together between musical improvisation and creativity she went on explain that for improvisation to work effectively there needs to be a foundation:

    1. Tools
    2. Structure & framework
    3. Roles
    4. Communication & expression
    5. Exchange of ideas


    Once you have this foundation in place you can explore and improvise in the space they create.

    She also drew an analogy between musical improvisation and the spontaneous call and response nature of the real-time web, but I’m going to need to go back a listen to the talk again to fully get my head round this idea.

    I’m also going to have to seriously hit the guitar practise again.

    I’m not sure where to begin with James Bridle’s The Value Of Ruins. I really think in a few years time it’s going to be classed as one of those “oh man, you had to be there” talks.

    Starting off with an introduction that left us all wondering what terrible fate had befallen the vanished place that James came from (hint: Yahoo!) he then delved into a fascinating exploration of historiography, cultural preservation and the web.

    Did you know that the Wayback Machine is contained within one of Sun’s portable datacentres? You could steal it with a truck if you were a nefarious sort. Which would be the 21st Century equivalent of the destruction of the Library of Alexandria.

    What about the game of Wikiracing? Start with two random Wikipedia articles and try to get between them in the shortest number of links (no editing and no diving into the article history). It’s using cultural knowledge as game-space.

    (there’s an iOS app which will completely suck any spare time you have)

    James explained that this deluge of information we seem to be suffering from has always been there, but now we have the tools to explore and understand it. To illustrate this he had the entire edit history of the Iraq War Wikipedia article printed and bound. It runs to twelve volumes and covers December 2004 to November 2009, 12,000 changes and nearly 7,000 pages.

    It’s a cultural and historical narrative, and if we don’t start designing for these narratives we’re going to start losing pieces of them.

    A phenomenal talk.

    Expanding on his Designing for a Web of Data talk from 2007, Tom CoatesEverything The Network Touches drew parallels between how the road building of Darius the Great in the 5th Century BC changed the nature of communication in the Achaemenid Empire, and how web APIs are the new roadways of the connected world, bringing together data from social software, geolocation tools, visualisation and the real time web.

    One of the more “controversial” aspects of his talk was the slide entitled “Death to the Semantic Web”. Tom’s point being that the traditional idea of the Semantic Web takes a top down approach, trying to impose a sense of order upon the seething maelstrom of the web. That may work in some select disciplines but the web is more dynamic than that and we’re seeing success with an organic approach using APIs, microformats and simple feeds. A great example being Lanyrd, with its ingestion of data from various sources such as a social graph from Twitter and locations from Yahoo!’s GeoPlanet.

    Another issue is that network connectivity is becoming a commodity (a similar proliferation happened with LCD screens when the cost dropped so low, everything had a screen) so objects are becoming connected, and people are using these connections in new and interesting ways. Personal informatics from systems like Nike+ and Withings are changing our relationships with objects.

    Connected things are transforming the world, we’re effectively building roads for future generations to use and improve upon. As Tom said with his final slide, the world, our Blue Marble, is both the brush and canvas for future endeavours.

    Oh, and 150 transitions in one slide. Kudos!

    And so to the last talk of the day, Merlin Mann with Kerning, Orgasms & Those Goddamned Japanese Toothpicks (there should be an award for a title like that).

    In a wide ranging, slide-less and free wheeling talk Merlin explained that we’re all nerds, but it’s okay. Creativity comes from that nerdy gene that we all have. The important thing is to not succumb to tunnel vision. Look at the horizon every once in a while and don’t become blind to new things to nerd over. You have to keep learning.

    As Xander described it on the way to post-conference drinks, it was possibly the best commencement speech ever.

    And so that was dConstruct 2010.

    Podcasts of all the talks are available on the dConstruct site, reactions and chatter can be found via the #dconstruct and #sausagebap (don’t ask) tags on Twitter, and Jeremy Keith has been curating a collection of photos on Flickr via the machine tag for his take on the conference.

    See you next September.

Page 1 of 2