Week 5

This week, as well as catching up on my Flash skills (I'm having some problems uploading the .swf files of my practicing to the blog, will try to add later) I have been working on designs for the website. The Flash stuff we have covered has been fairly basic - making stickmen walk across a screen, making a little point and click game - useful stuff really, because I had completely forgot some elements - so it really helped. I don't feel like much of the practice will really benefit my project, but it's good to remember how to do certain things. I am thinking of trying to use stuff like moving grass or something in future review conversions. But anyway, on to the website...

I have decided I want it to be quite basic and minimalist to give more focus on the images that accompany each review. Here is a quick sketch of how I want it to look;


Click the thumbnail to enlarge

That's just for the home page, but obviously the content box would change on each page; the reviews page would display each review in the boxes, the news would have updates etc.

I think in the initial stages, I will just stick to my original idea of nothing fancy, just the images next to the text. At least in the pilot stage. After this, I may look into some other approaches, like doing artwork myself and focussing on more of a data collection process, like I mentioned before with looking at frequency of words used etc. - as well as looking to implement a rollover AP Div style word highlighting mechanism. So when the image is rolled over, the corresponding text in the review will become highlighted.

Week 4

This week I conducted my first experiment with the concept of my project, in order to put my theory into practice. I wanted to try and make the first image for my Pilot, and here is what I came up with;

My first experiment consisted of quickly checking NME.com for a review that I thought would be particularly interesting to cover. Being a fan of M83 myself already, I knew their music was often atmospheric, dreamy and usually consists of a vast soundscape to pick apart. Whilst I haven't yet heard the album this review covers, I felt the language was adequate enough to convert, so from the following review, I compiled this image:

Review taken from NME.com (http://www.nme.com/reviews/m83/12375);

A grandiose double-album paean to childhood dreaming that only occasionally needs the naughty step

7/10On ne voit bien qu’avec le cœur. L’essentiel est invisible pour les yeux(One sees clearly only with the heart. What is essential is invisible to the eye). Not our words, but that of a clever fox in Le Petit Prince, Antoine de Saint-Exupéry’s profound children’s book. The book tells the story of a child-like alien who travels to other worlds. Visiting Earth, he meets a crash-landed aviator in a bleak desert.

While the stranded pilot is a weary grown-up, the prince is yet to be tainted by adulthood. If Anthony Gonzalez (M83) is like the aviator, having imagined this double album while in the Joshua Tree desert in California, ‘Hurry Up, We’re Dreaming’ is itself the Little Prince: guileless and dreamy. Quite a bold statement to make, but this is an album of equal valour.

‘Intro’ really captures the spirit of the album, guest Zola Jesus’ voice both otherworldly and powerful. It’s a jaw-dropping start that fades into ‘Midnight City’, arguably one of the best tracks to be released this year. Its dancefloor-filling melody boasts the finest sax solo you’re likely to have heard in the past 20 years.

It’s this pomp that makes the album work, and on ‘Reunion’, The Breakfast Clubsoundtracked by Toto, the fantasy is properly realised. Gonzalez has said that each track on the album is an interpretation of people’s dreams. But there’s another theme that runs through both sides of this epic: childhood innocence. At times this works well (‘Splendor’), but the Dora The Explorer-like voiceover on ‘Raconte-Moi Une Histoire’ is a mega peeve. There’s a lot to take out of ‘Hurry Up…’ but moments like this, and the synth interlude of ‘Klaus I Love You’, detract from the overall vision.

But, if you can look past that and use your heart, you might rediscover your own absentminded childhood and an album full of hidden pleasures.

Jamie Crossan

Research: 1

Edward Tufte


I have recently been exploring the work of Edward Tufte, as recommended to me by Rod. My initial thoughts are that whilst Tufte's ideas on the relationship between data and visuals is fascinating, and has some scope in the ideas I have on this project myself, his actual practises in creating visuals of data are far too niche, far too focussed on scientific ideals. 


In Beautiful Evidence, Tufte touches upon some of his own methods of data visualisation, such as the Sparkline, which has been made popular in visualising rises and falls in the stock market, or temperature activity, and I feel it doesn't really have much to do with visualising the imagery a reviewer uses. A lot of Tufte's methods are about compressing as much data into as smaller an image as possible, just for the sake of efficiency and whatnot. With reviews, if I conclude something to be a key element, it's not about being dense, sparse and cutthroat with the way I display it. Taking creative data and re-imagining it creatively does not really call for the same kind of methods in order to preserve time or page space.


Again, as I briefly mentioned before, it would be great to show which words show dominance in each review with a method like these; whether or not the word 'chorus' fluctuates more and more with one certain reviewer's work, for example, but to simply convert text to an image seems a lot more of a creative process than an analytical one.


In my initial search, I also came across Tufte's Beautiful Evidence lecture, from Intelligence Squared in London, Wednesday 19th May. Here is a brief highlights video;

The full version is available to view for members only on the Information Squared website. In this lecture, Tufte explains the reasoning for some of his arguments, gives some examples of his work and generally tries to put his ideas into context. Here are some quotes I found particularly interesting:

"Evidence that bears on questions of any complexity typically involves multiple forms of discourse. In modern scientific research, for example, about 25% of published materials are graphs, tables, diagrams and images, and the other 75% are words. The spirit of real science in publication is whatever it takes to explain something."

I realise Tufte's ideas are a lot more scientifically orientated, but the message I want to explore is largely the same. I think in regards to my experiment and music reviews, the format is irrelevant. There are ideas, emotions and feelings that the reviewer tries to evoke with his writing, but I don't think the writing aspect is essential in that. This is why I want to see if it really works on a visual level as well. 


By default, music reviews take one idea and almost transcode it to a completely disconnected and unrelated format. I don't think it is data that they are dealing with at all; data collection on music would just be a much more sterile, emotionless take - what kind of guitars were used, how was it mic'd up, how many strums of each chord pattern and so on. That would be a more Tufte-like approach to exploring creative writing dealing with music.

"Evidence is evidence, whether words, numbers, images, diagrams, still or moving. The information doesn't care what it is. The content doesn't care what it is. It is all information. For readers and viewers, the intellectual tasks remain constant regardless of the particular mode of evidence. To understand and to reason about the materials at hand and to appraise their quality, relevance and integrity."


This quote is perhaps more key in understanding what it is I'm trying to achieve. Obviously there are questions about how I approach this creatively, but what I am looking to do is to make a visual representation of existing texts. There will be certain things I can't really incorporate into a collage, such as specific words that make up sentences; the "ands," the "buts" the "the's". But it is more so the message that I am trying to visualise. "The content doesn't care what it is." For the experiment to work, I need to draw my own conclusions as to which medium does the job it is set out to do in a better ability, but that aside, analytically, all I really want to know is if the user gains the same, or draws the same conclusions regardless of whether they've experienced each review visually or textually.

"We shouldn't deny ourselves the uses of every possible mode, and we shouldn't segregate the information by the modes of production."


Again, Tufte's ideas play into how I want to look at the relevance of text as a format. Either way, sound is difficult to visualise, and all music reviews seem to do is create a mental image of how music sounds. It's quite complex, but Tufte definitely takes a similar approach with his presentation of data as I intend my project to.

Week 3

This week I have been thinking of how to put my project into practice. After initial discussions with Rod, it seems that my approach is worth perhaps expanding and experimenting with. I have been told to look into Edward Tufte's writing - covering Data Visualisation and different ways of producing data. Whilst, again, I wouldn't really say my focus is on data collection as such, an initial Google search proved interesting.

I have been recommended books such as Tufte's Envisioning Information (2001) as a starting point to add more theoretical context to my idea, and this paired with websites such as Information is Beautiful have really helped me to pinpoint what it is I'm trying to do. I'm not really trying to make a picture of the music as such, I'm trying to convert what the reviewer is trying to say the music looks like. Whether or not the images are particularly striking is not really the point of focus; if anything, if the image lacks visually it says more about whether or not music reviews serve their purpose in the written form than anything else really. It's more difficult to use visuals to describe audio than it is to use audio to describe visuals.

It has been suggested that I should look at work such as Tufte's and try out a similar approach. For this to really work, I think I would need a lot more skill with freehand drawing and photoshop than I currently do. It would be a lot more time consuming and perhaps too artwork based if I were to take this approach. I think the whole point of this is to take key ideas and splice them together, rather than create an epic visual landscape. If I were to look into the idea of drawing everything myself, I don't think it would be before the Pilot version is created. I think for this idea of data visualisation to work in the context of my idea, I would need to focus on the fact-based element far more - for example, observing the amount of times a certain word is used, and then creating a visual chart to correspond. For example, if 'guitar' is used frequently in one review, I could make a visual tally with images of guitars I have created myself. This idea may be worth looking into, but it doesn't really fit into the framework of my approach.

Tufte's work seems largely based on facts, figures, graphs and charts; scientific experiments being visualised rather than artwork made for the sake of art, although in Envisioning Data, Tufte seems to see himself as an artist with no active audience, at least in the field of art itself. His ideas and actual theories on why visualisation is important seems a good starting point for my own. I will post a blog about my reaction to his theories and ideas shortly.

Aside from this, I have been revisiting ideas of intertextuality. Within my images, even from initial searches on music review websites, I have found that the texts are often laced with references to other bands, other places, other images. In Semiotics: The Basics (2002) Daniel Chandler writes

Although Saussure stressed the importance of the relationship of signs to each other, one of the weaknesses of structuralist semiotics is the tendency to treat individual texts as discrete, closed-off entities and to focus exclusively on internal structures. Even where texts are studied as a 'corpus' (a unified collection), the overall generic structures tend themselves to be treated as strictly bounded. The structuralist's first analytical task is often described as being to delimit the boundaries of the system (what is to be included and what excluded), which is logistically understandable but ontologically problematic.

So he backs up my ongoing argument that music reviews, in text format, do impose a sense of a strictly bound and limited system. The idea of internal structures is interesting to me, as the text written in a review is self-contained really. Illustration doesn't seem to come into contention, and a world of its own, a visual landscape is created within the writing, and it's the (apparent lack of) consideration or focus on that I am interested in.

Music in its raw format is obviously audio, so text doesn't often translate as concisely as may be required. The same goes for food critics. Obviously if you have already tasted something, you can relate to the ideas, as my O'Toole quote last week highlighted. But to try and describe something, there are often derivative references. This music sounds like this, this food tastes like this. I think it is incredibly difficult to describe a taste to someone who has no idea. With the web as a tool, is it worth expanding on the limits? After quite a deep search, I haven't found any music review websites so far that take the same kind of approach I am experimenting with, but almost all text based review websites feature writing that evokes some kind of mental picture. A picture speaks for itself, it's all of the intertextuality and derision in music writing that takes away from its creativity. It seems like it isn't language itself which presents a problem when trying to convert to an aesthetically pleasing imagination of it. It's the writing style. Is it simply the traditional formatting in publishing as text that is limiting the potential? I'm not so sure.

Chandler, D. (2002) Semiotics: The Basics. London: Routledge

Week 2

This week, we in the web department established that we could do with some refresher lessons in Adobe Flash. I am intending on using some elements to possibly highlight text in the reviews I work on when scrolled over, using AP Divs, and it's tricky remembering how to do this. Should be very useful.

Also, whilst discussing everything with Phil, my attention was drawn to http://www.informationisbeautiful.net/ - a website which kind of specialises in data visualisation, superbly.

The above image is one of the latest posts on Information is Beautiful. I thought it was quite interesting as this particular piece was sort of a competition for the website's audience. They were simply asked to sketch what they thought their soul looked like. Obviously this is interesting to me, because in a sense I am trying to figure out whatever the reviewer of a piece I am studying thinks the music they are listening to looks like, so the creative aspect spoke out to me, I suppose. The thing about this website is the creative aspect, in theory rather than design, is slightly limited. Obviously the way each piece of data is presented takes a lot of creativity, but the fact my subject matter is really just creative writing in the first place, as opposed to Information is Beautiful's data research, means there is more room for manoeuvre in my project. 

I know, in essence, it seems like this is similar to what I'm aiming for with the reviews, but I think it's important to highlight that what I'm taking from the reviews is not really 'data' - it's more imagery, visions, language. There is a danger that what I'm extracting from the reviews could be too specific and personal to the reviewer who wrote the review, but at the same time, I think it's more just a way of displaying a form of media rather than collected data. I'm trying to explore the relationship between language and what the user really gains from it; whether or not my images really correspond with whatever the audience is encouraged to visualise.

In order to really engage with what it is I am trying to dissect, I have been reading into the relationship between image and text. I came across a book by Michael O'Toole (1994:4), called The Languague of Displayed Art, which argues that

semiotics – the study of sign systems- can assist us in a search for a language through which our perceptions of a work of art can be shared. I believe that we should start with the impact the particular work has on us in the gallery, or even in a book of reproductions, but this semiotic approach will also allow us to relate the nature of this impact to the scene portrayed

So in this, my perception of even 'language' itself has been challenged. O'Toole emphasises that recognition and awareness of a context plays a large part on how we perceive whatever it is we're observing; in his case a work of art. I think in order to find the answers I am looking for, i.e. whether image can evoke the same response as text in music reviews, it's arguably not even language I am more focussed on, it's the audience's understanding of semiotics, and their recognition of culture. The image can't really say much if you don't recognise some of the more iconic imagery I will be trying to include. There is a lot to think about.

O’Toole, M. (1994) The Language of Displayed Art. London: Leicester University Press.

Week 1

After a number of discussions with friends, family and coursemates, I decided over the summer that I wanted to take my own personal aspirations for what I want to get out of this course, and challenge the boundaries. I came to UWE with the intention of aiding my aspirations to become a music journalist, so this year I have been trying to gear my work in that direction. 


Tom told me about an interesting art installation he had seen in which wine critics tried to make a visual representation of the blurbs on the back of bottles of wine, or something similar. The details escape me. But from that, I thought it would be interesting to look at music reviews and really dissect them, highlight what they actually are, what they hope to achieve, and then mess with the conventions to expose them. 


Does the language in music reviews really help us to imagine what music sounds like? Or does it simply paint a picture or a storyboard for how an accompanying music video may look? I'm not sure that flowery poetry about how a guitar 'soars' really helps the audience to understand what to expect. Anyway, I wrote a vague proposal to try and summarise what it is I'm looking to do;


For this year's project, I intend to examine magazine style reviews, and explore whether or not the language they use is successful in really describing what to expect of whatever product or event it is they are featuring. Due to my personal interests and intentions as a journalist and academic analyst, I am more interested in music reviews in particular. Obviously what music reviews are trying to achieve is to not really paint a visual picture, but to paint an audio prediction. Through this, though, they often use colourful imagery and whether they intend to or not, inadvertently achieve a much more visually rich description. Does the audience actually know what to expect musically?

I aim to take a number of existing reviews, or possibly use some reviews I have written myself for Guide2Bristol.com and try to convert them into visual translations. I aim to take any of the more particularly rich imagery and make collages depicting what it is the review is really trying to say. To make the whole experience more coherent, I will include the original text in which the image is based on, and perhaps use AP Divs to highlight particular phrases when certain parts of the images are rolled over. I will present my work on a reviews website I have created myself, loosely based on the sort of design that features on websites such as NME.com.


To put this into context, I was thinking of creating a reviews website, not too aesthetically dissimilar to websites such as the NME's, and just having links to each including the original text and then the image displayed with it. The user will engage by trying to dissect which part of the image corresponds with each key concept, idea or imagery touched upon in the reviews. I need to start working out a basic website structure, but I think initially it won't be too fancy. The images will do the talking; I just need the bare bones to work.


Also this week, myself and Amy teamed up with Tom (the usual sort of grouping) in the tutorial workshops for the mini practical projects we've been assigned with. Myself and Amy are doing a couple of Web-based pieces - the User Unfriendly website being an obvious choice so far, and Tom is working on a couple of bits of video. Screenshots etc. coming up.