Introduction

In May 2008 Airside turned 10. As we approached our decade of design it was suggested to us that we should put out a book to celebrate. The initial idea was to produce a small booklet. So how did that turn into the 296-page, 2kg behemoth that is adorning the most fashionable coffee tables right now?

This is that story. It’s supposed to be a 500-word blog, but given what happened with the book, let’s see where we end up.

Interviews and commentary in the book are by Fiona Sibley.

Booklet To Book

Our agent in Japan was very keen on the idea of an Airside book and predicted that our large Japanese fan base would be overjoyed to have access to the narrative in their native tongue, and so the decision was made to tell the story in both English and Japanese. Immediately we realised a small booklet would not cope with a dual text format. For that we needed whopping great big pages, which served the additional function of showing off the work.

Having decided on a larger page size, Chris designed a grid based on an 80-page book that met the team’s general approval. As Fred said “it was a slam-dunk from the off”. With all the style and content changes to come, the flexibility of this grid would prove to be invaluable.

What’s The Story

Meanwhile we started considering what story we needed to tell and how it should be told. We reached the conclusion that the unvarnished story of Airside’s evolution, warts and all, was an interesting one and complemented the visual work perfectly. Inspired by a Mojo magazine feature on the history of punk, we decided the narrative should be presented in interview form with each participant chronicling their version of events and the interviews divided and arranged chronologically to tell the story from the group’s perspective.

Various other specific subject matters warranted their own chapter, such as the Airside Aesthetic and the Lemon Jelly project, and so a structure was put in place whereby every odd-numbered chapter told the story of Airside’s growth, while the four intervening chapters covered other topics, like the Airside Process or Self-Initiated Work.

On occasion we also needed a structural divide between the story and the work. While the text would naturally be illustrated with relevant examples of work, we also wanted certain pieces of work to have their own pages. So we introduced the double-page hero spreads, designed to show off those projects that were of great significance to Airside.

It soon became very clear that what we were planning would never squeeze into 80 compact pages. As Roy Schneider might have put it, we were gonna need a bigger book.

True Grid

Our smart, considered, well turned out grid suddenly found itself facing a rampaging hoard of contrary content. What’s a grid to do? Simply stiffen its upper lip and show how resolve can lead to greater flexibility. A number of tweaks later and we now had a layout that could cope with a varying number of pages, deal with large scale images and adapt sufficiently so we could alter the pace and tone of a now quite lengthy book simply by changing the look of a page.

With the grid in place, we started to think about the general style. One of the most important factors was to give the impression that the layout wasn’t necessarily designed by Airside. We wanted it to have a grown-up, quite straight feel that people wouldn’t immediately associate with us, leaving our personality to be expressed more directly through the work itself.

Deciding on the typeface was crucial. We wanted something that wasn’t too obvious and too Airside but would still suit the tone of the book. We eventually went for Bodoni Poster for the headings, which Fred describes as “classy, but in your face – a bit like Airside”. The bold serifs and formal feel felt like a radical departure for us. The contemporary workhorse Akkurat was chosen for the body text and together the two typefaces struck a unique and pleasing balance.

Finally we were ready to layout the prototype chapters and happily it all pretty much worked. The grid was adaptable and the pages were big enough to show off the work. Now down to the nitty-gritty.

Hairline Fracture

As the narrative is told in sequential interview format we had no clear way of introducing new people before their monologues began. We toyed with the idea of having a sidebar that could carry profiles of these clients and collaborators before deciding that the context of the text itself should be enough. The exception was for Airsiders who are introduced by having their name printed in a unique colour, which matches the one used on their business card.

These little touches were important as we wanted the text to feature details that you wouldn’t get in any other book. For example, each paragraph is indented and Chris came up with the device of filling the indented space with a hairline. This controversial move was widely debated. At one point we got the fear and took them all out, before deciding originality was what we were about and had to endure the pain-staking process of putting all them back in again for our sins.

Jap-lish

The next problem was how to present the English and Japanese text on the same page in a clear fashion that awarded equal billing to both languages.

The solution was inspired by an old Situationist fanzine called Semiotexte, which would often run two narratives on one page.

Our flexible grid meant the English and Japanese text could be at opposite ends of the page or sit beside each other depending on the requirements of the page. It helped that the two languages are so different that it’s hard to confuse the two. We also liked how this fluid presentation of the book’s bilingual text references how both Western and Japanese cultures often use each other’s language as a stylistic design element.

Of course this bi-lingual jiggery-pokery could have been well nigh on impossible, were it not for the seamless work of our intern Yoshi who prepared all the Japanese text and implemented the design template throughout with ruthless efficiency. He also makes a damn fine cup of coffee.

Photography

After attempting to cobble together a selection of images from our photographic archive it became very clear that in order to give the book the visual clarity and excellence it demanded, our entire back catalogue would have to be re-photographed specifically for this project. Anne and Jamie undertook this mighty labour with grace and enthusiasm and created a specific photographic aesthetic that gives the work the room to breath perfectly.

A mountain of clearance issues then had to be navigated as in many instances of past collaborations with clients we were not the sole copyright holders of the work in question. It was gratifying to discover how many clients were only too happy to be featured in our humble tome, and in many cases they were also happy to contribute to the text, taking the narrative out of the studio and into the wider world.

Cover Versions

The one thing we knew about the cover was that it would follow the tradition of Lemon Jelly’s record sleeves, so naturally one of Fred’s rhythmic, abstract patterns fitted the bill. It was the last consensus on the cover we had for a while.

Fred’s chaotic, lurid first design was hated by nearly everyone. Except Fred of course who thought it was damn good. (And I still do! – FD) Tweaking the pattern didn’t work, changing the colours didn’t work, nothing worked until Fred fired everyone adapted the design for the book’s end papers.

We decided the end papers should be a very traditional, one-colour design (just like the Tintin books and many others) and the obvious contender was a repetition of the cover design, so Fred reworked it accordingly. Suddenly in the harsh light of monochrome, everyone began to see potential in the pattern. Reducing the scale of the circles made it more attractive and colours were reintroduced but more muted than before.

Finally there was a consensus and it was time to mess around with it.

The cover looks like a simple pattern but in fact there is no repetition at all. The trick is to establish a rigid set of rules (in this case you have eight different types of semi-circle, lines of equal distance apart and a grid pattern) and then see where you can introduce variety. On the front cover the semi-circles opposing each other are different colours, on the back they are mostly the same.

This echoes the lemonjelly.ky sleeve design where the blobs on each panel of the foldout cover are heading in two different directions, but these directions are different on each panel.

This subtle contrast is also a way of differentiating between the front and back of a book with no type on the cover. Like the Lemon Jelly records and Fred’s Triptych and NuBalearica mix albums, the book has no text on the outside. The idea is to create something mysterious that draws the reader in, so they think, “what could this possibly be?” and are drawn to explore it.

However this theory was not one that our distributors, publishers Die Gestalten Verlag, initially agreed with, feeling that the lack of type on the cover might compromise sales. Convincing them otherwise was a difficult task but eventually we plied them with Lemon Jelly CD sales figures and design awards and they relented.

Besides, we knew we’d have to make concessions to the retailers who would need some kind of promotional blurb on the cover, let alone a bar code. Like all Lemon Jelly releases we designed a sticker with all the relevant information that would go on the book’s protective shrink-wrap, which could be discarded once the customer had bought the book. Of course now we quite like the sticker and have some regrets about not incorporating it onto the book itself. Maybe the second print run will be different!

Finally Die Gestalten Verlag on seeing the final design insisted that we brighten up the colours a tad, and in hindsight they were quite right to do so as the original was far too dull. The power of collaboration and compromise!

Airside By Airside

Nine months, ten cases of RSI, 19 nervous breakdowns and one great collective effort later we sent that book off to print never caring if we ever saw it again.

Until we did see it again in all its 296-page, 2kg, full-colour glory.

Then like a mother with a newborn babe in her arms, all the labour pains melted away into exhausted sighs of pride and joy.

Our book. Enjoy.

Following the positive response we had to our earlier process post, we thought it would be interesting to continue the series with another ‘transparent’ project. However, to avoid this post succumbing to the notorious ’sophomore syndrome’, we have decided to cover a smaller project from 2008 but in greater detail: the exercise of designing a trademark for our Japanese agent Dream Ranch.

It is important to note that regardless of the project covered, it will always be difficult to present absolutely everything that informed our final design. It is challenging not only due to technical restrictions, but because for every quantifiable idea there will be some ephemeral factor that is either lost in the process or omitted for professional reasons. The delicate waltz between designer and client is not always appropriate to document.

Dream Ranch has handled Airside’s Japanese representation for over four years, and approached Airside to re-design their trademark to help increase their visual presence.

At the time of briefing it was clear to both parties that Dream Ranch’s logo wasn’t working hard enough. On consideration we agreed their existing mark failed at the first hurdle of logo design – it just wasn’t all that recognisable.

The graphic was fairly presentable and it had obviously done well (our Japanese intern Yoshi assured us the logo was quite unusual for the type of business it represented), but it didn’t posses many of the attributes needed to be identifiable and above all memorable in the market it occupied. The main concern was that the Dream Ranch trademark functioned very well as a graphic, but not as a logo.

The definitions and qualities that make a good logo have been endlessly debated elsewhere. It is certainly worth a more exhaustive post in the future, but for now we must at least touch on it in order to explain just why we thought the Dream Ranch figure failed as a logo and (more importantly) why it could work harder.

For a trademark to be successful it needs to surpass mere representation and gain a degree of ‘excellence’, a standout attribute that imbues either logo or logotype with the ability to stand out through visual noise.

Although ‘excellence’ can be seen as a bit of a catch all word for ’something cool’, the term is best referenced in Per Mollerup’s insightful identity book Marks of Excellence, as a way to describe a trademark that employs semiotic muscle to elevate itself above the boundaries of straight representation.

Mollerup succinctly summarised this through his ten ‘Kinds of Identification’: uniqueness, value, holding power, description, association, tone, graphic excellence, reputation, discretion and repetition. A successful trademark usually fulfils at least seven of these attributes (as some of the attributes work against each other), and is essentially why we talk of a logo ‘working harder’.

Excellence’ can be found crafted into most of the classic logos (take a bow FedEx, British Rail, V&A) and can sometimes develop from the most unexpected sources – Raymond Loewy’s logo for Exxon was a gift thanks to the inherent ‘excellence’ found in the name’s double ‘xx’.

In comparison, whilst not possessing a typographic flourish in the same league as ‘Exxon’, Dream Ranch’s wonderfully descriptive name provided us with a starting point full of potential ‘excellence’.

It was clear from the outset that the logo could easily occupy the territory of ‘dreams’ or ‘ranches’, so it is interesting to note how one of our first ideas ignored the name completely and attempted to represent how Dream Ranch worked as an agency.

Despite the obvious challenge of illustrating an abstract concept, we set out to create a graphic metaphor to describe Dream Ranch’s “draw for talent”. Bob Gill once referred to a method of “designing with noise” for illustrating the abstract, and while a geometric mark would have provided some solution we felt it would have been a tad impersonal for a company that prides itself on face-to-face business.

Our solution above solved the problem by introducing ideographic ‘bric-a-brack’. Both representative and, to some extent, noisy, the logo’s strong graphic runs parallel to an ever darkening half-tone pattern that symbolised Dream Ranch’s work. The logo felt representative of something being the product of many parts and its flying elements all fizzed with an energy suggesting excitement. On presentation this logo nearly went all the way, but fell early when the client started to see comparisons to a toy box.

Even after some furious tweaking to reduce the ‘toy box’ comparison, the damage was done and the logo had to be shelved. Thankfully on approaching Dream Ranch with our initial sketches, we were careful to cultivate a spread of ideas.

The logic being that of the Hydra – to lose one big idea provides space for two smaller ones to grow.

Amongst our early responses to the brief was an exploration of a bold letter ‘D’.

Although it had great impact, a single letter just seemed to lack the holding power a trademark should have.

Shown above are a number of shape exercises compiled from our early sketches that aimed to communicate the notion of ‘dream’ in as few strokes as possible.

These sketches all attempted to emphasise the ‘dream’ quality of the client’s name, and although slightly unfocused (the horse and eagle were quickly sidelined), one sketch proved to be the seed of the finished design.

Glancing over the rainbow’s arches it suddenly became clear that the arcing forms could easily be abstracted into initials of the client’s name.

In true Hollywood style the sketch on the right was scribbled on the way out the door in a snap reaction to the rainbow graphic.

Following that first scribble, it was a matter of chipping away at our vague notions of what was possible with this idea to create something a little more focused. At this stage we knew that there was an idea to use the initials of the client’s name – we just weren’t entirely sure how it should be done.

There was the obvious rainbow parallel, but by sketch number 4 the idea of creating a landscape had started to develop. Setting suns and nightfall seemed an appropriate precedent for ‘dreaming’ so we began to explore the different ways the D and an R could be contrived to form a landscape.

By sketches 13 – 15 we had started to think about the possibility of creating a trademark with a logotype formed from elements of the logomark.

To get our heads around the problem we presented the client with the above sketches to help visualise the solution. Oddly, at this point, we were intent on depicting the sun setting in-front of the hill.

It was important for the logomark to read both as a sunset and as Dream Ranch’s initials so many sketches were produced to try and pin down exactly how this could be drawn.

Some way into the sketches it became clear that there needed to be a degree of context for the logomark to function in this dual role, and consequently an ideographic ‘ranch’ was introduced in the later sketches.

At this stage of the project we were beginning to feel frustrated, as although we knew how we wanted the logomark to look, we found getting the right proportions for the logomark much trickier than we initially thought. It seemed that with every hand-drawn sketch we were able to capture a much nicer set of proportions to those we were creating on the computer, so after copious sketching and heavy experimentation with ratios we managed to hit some something that approached the beauty of those little sketches.

The inclusion of the ideographic ranch was also proving to be a headache. In our sketches it seemed to sit perfectly well on the hill, yet in practice this not only was a nightmare to read at a smaller scale, it seemed to disturb the logomark’s balance. To solve its positioning we looked to a typographic trick (white type is much easier to read on black), and punched its shape out of the hill.

With a completed logomark our attention turned to the trademark’s typography. Although we had an idea of marrying the logomark and logotype together, our numerous sketches of the logo with a hastily scribbled name did have a certain charm.

Thankfully we quickly saw sense and passed on these, and set out to find a typeface with initials roughly matching the shapes in the logomark.

Since the letter ‘D’ and its corresponding shape would always be rather similar regardless of the typeface, the letter ‘R’ proved crucial in the search for the right typeface. After a quick search through the various foundries we eventually found five typefaces that all possessed a broken ‘R’ that could be easily replaced with an ‘R’ of our own design. Although it could be argued that it didn’t really matter which typeface we went with, we felt it was important to find a typeface that already possessed many of the attributes we were going to end up with as a considered choice would make the dark art of inserting custom characters a whole lot easier.

With its strong uniform characters Avant Garde was eventually chosen from our shortlist, however Litera did get a look in thanks to its broken ‘e’, but our belief that this would compliment the broken ‘R’ proved to be incorrect as early experiments with two broken characters looked rather shambolic.

On considering how the logomark and logotype would work together it was clear that a straight pairing of the type with the sunset shapes was not going to work, the weighting was wrong and the ‘R’ could barely be read at any level. A contour-hugging option was also considered, and although this was favoured by some of the design team (as it possessed more of an overt relationship with the logomark), its terrible readability forced our hand for a punch-out version of the sunset shapes.

These characters were then typographically tweaked to match the line weights and tapering of Avant Garde giving a consistent pair of letters to join the set.

With the relationship between the logotype and the logomark solved, Dream Ranch’s trademark was complete. In our final presentation to the client the design team were confident the logo would be a great success, however nobody saw a problem that nearly sunk the idea and prompted a complete redesign.

All visual communication is subjective, and since design’s purpose is to act as a conduit for communication, anything designed (especially a logo) is wide open to interpretation. With trademark design ambiguity is not an option.

To explain Dream Ranch’s problem consider the image on the left: presented here are two classic graphic standards of ambiguity – the ‘glass half-empty / half-full’ and the ’sunrise / sunset’. Our Dream Ranch logo belonged to the latter.

Although they loved the mark nobody at Dream Ranch could tell whether the sun was rising or setting. This may not have been such a problem in the West since sleep and the sun’s transition from either night or day possessed a close enough meaning, but in the East the symbol of a setting sun is seen as a very bad omen. A company with a setting sun – is a company going out of business.

However, there was a solution – by applying a semiotic short-hand to an image some sort of context can be achieved. The images on the right, although containing the same base features, have had their contexts altered by the addition of the straw and the rooster.

Applying this thinking to the logo, we altered the logomark to feature a rooster instead of ranch. This change gave the logomark a narrative of a rooster announcing the dawn, and so set the logo’s outward representation of a rising sun. Thankfully Dream Ranch agreed.

On tidying up the trademark we came to an agreement that the rooster was too complex and clashed with the simplicity of the mark so as a final step the rooster was redesigned, and constructed out of a basic shape set to increase its readability at smaller sizes.

Also to give what could be a rather cold ideogram some character, an oversize eye was added along with slightly exaggerated features to help the rooster graphic maintain its shape at different scales.

With the sunrise/sunset question answered the trademark was completed and presented to Dream Ranch. Since this was not an identity project, no rollout was required – however Airside maintains a strong relationship with Dream Ranch and we hope to develop the ideas that informed our trademark design into a wider identity in the near future.

As a comparative piece to the earlier Airplot post we thought it would be interesting to round up some of our preliminary notes, sketches and worksheets that informed the finished identity.

On the occasions Airside has presented its process talk ‘I Don’t Like It’, were puzzled by the audience’s surprise at just how many sketches and worksheets contributed to a finished design.

We hope that by presenting the ’scrappier’ parts of a project that most agencies would seek to hide, a lot of the mystery behind the design process can be swept away and reveal the work that goes into such an undertaking.

Pictured above are a collection of sketches (gleaned from both computer and sketch book) that document the design team’s initial response to the first client meeting. Those with a keen eye may spot the beginnings of the finished identity, but interestingly there was an overall feeling of tessellation emerging from these very first ideas.

As the project continued the idea of using type as a constructive device became the favoured route, however the design team were concerned that the concept of ideographic type would be hard to explain.

After culling the directions considered unsuccessful, we approached Greenpeace with the above idea. From our point of view it was important to communicate how fields could be abstracted into an underlying grid.

This point was crucial as it would allow us to use the field metaphor, but through a much more manageable form. To explain this point we presented the 4 sequential images featured in the above presentation to hammer the point home.

Once the idea had been explained our attentions turned how the identity should feel. The design team did not want it to look too sterile, so it was decided early into the project to avoid using a computer throughout the typeface’s creation.

We quite liked the sketched typeface we were using in the first sketches, so taking the letters already used we were able to expand these into thick, medium and thin sets, which were then extended across the entire alphabet.

Pictured above are a couple of scanned sketchbook pages illustrating the evolution of the Airplot typeface.

These rough weights were then scanned, cleaned up and printed back out to be mounted and turned into cardboard stamps. Much arduous printing followed – we found ink to be too thin so we opted for acrylic paint instead, this proved to be a messy choice!

Unfortunately no black paint was to hand, but blue was, and in huge quantities too. In the following days not only did I slowly turn blue, but most of the make-and-do-room did too.

When the letter sheets had completely dried the alphabet was for a final time scanned back into the computer to be cleaned and set.

The typeface had to look ‘natural’ but still be readable so in places where the type looked a little too natural the letter was either re-printed or tweaked digitally.

With Airplot Thick, Medium and Thin ready to go we set about compiling the identity. For the results see the previous post…