Here are some of the most relevant, interesting and useful articles I’ve come across during March 2010:
Peltier Tech Blog | Jon Peltier provides a makeover to a NBC olympic coverage chart | Link
Juice Analytics | Reviewing the results of a survey which asked ‘Are the Viz-Pundits Really Helping?’ – I’m not sure the survey ultimately answered this question but there were some interesting insights | Link
Eager Eyes | Discussing a month of chart madness | Link
Smashing Magazine | A guide to the most important print magazines for web designers, digital artists, and photographers | Link
Ted Talks (via Information Aesthetics) | Video of Gary Flake’s demo at TedTalk2010 of the pivot technology – a new way to browse and arrange massive amounts of images and data online | Link
Ted Talks | Video of Tim Berners-Lee’s ‘The year open data went worldwide’ at TedTalk University 2010 | Link
Visual Business Intelligence| Stephen Few’s article about ‘Big BI’ being stuck | Link
Eager Eyes | Robert Kosara sets out his case for 2010 being the year of InvoVis theory | Link
Digital Roam | Post with a video of Dan Roam talking about visual thinking | Link
Smashing Magazine | Interesting article about form vs. function – a burning issue in the visualisation field that continues to increase in prominence | Link
Presentation Zen | Garr Reynold’s article about the importance of ‘play’ | Link
Journalism.co.uk (from Feb 10) | Data visualisations that tell the news | Link
Design Research Conference 2010 | Interview with Don Norman | Link
Jakob Nielsen’s Alertbox | Presents findings relating to the amount of time users spend on information presented above and below ‘the fold’ of a web page | Link
Tony Bain | What is Big Data? | Link
Flowing Data | How to make a heatmap using R | Link
Another interesting challenge presented by Nathan at FlowingData to improve the design and clarity of message of the graph presented below which displays the results of a study investigating obesity rates at different ages across people who were born in different cohorts of years.
The primary difficulty in using this display is the problem caused by having to decode eight line graph series which represents the various year of birth cohorts. As Nathan comments, you can eventually adjust your reading ability to draw some insights from this graph, though it does require a multi-slide presentation to truly impart the key messages.
My re-working below simply takes the raw data, moves the year of birth categorical values on to the x-axis and then applies Excel’s cell conditional formatting to present a colour pattern of obesity rates from 40% (darkest red) down to 0% (pure white). I have used a single hue with changing tints towards white as this helps our visual system to better judge a decreasing value. I have left the original values in the cells for two reasons – firstly, I can’t for the life of me remember how to switch them off and, secondly, I actually think they add an extra layer of potential insight being able to reference key values when explaining the insight from this study.
There has already been a great deal of coverage across visualisation-related blogs and news sites about the reaction of General McChrystal, the leader of American and NATO forces in Afghanistan, to a slide he was was presented portraying the complexity of American military strategy. I can’t avoid the temptation to add my own contribution to this debate.
Here is an image of the slide in question, click on it and zoom to see the bigger version:
The focus of the coverage has been on Gen. McChrystal’s quote “When we understand that slide, we’ll have won the war” using it as a further example of the curse of PowerPoint. The New York Times article, in particular, has used this incident to stick the boot in somewhat, reeling out an array of examples of how senior military commanders deride Powerpoint.
This is typical of the ridiculous, misplaced negative sentiment that exists towards PowerPoint, blaming the tool rather than the craft (presentation design) or the craftsman (the PowerPoint designer). Is it PowerPoint’s fault that people use terrible cliparts? Or thousands of bullets? Or dumb background templates? Or useless transitions and animations? Is it my hammer and drill’s fault that I’m rubbish at DIY? No.
It is at least refreshing to see a large proportion of post comments across the web trying to make this very point. However, there are still far too many people out there who totally miss this point and lazily jump on the bandwagon – probably being driven by Edward Tufte (though at least he was willing to backup his opinions).
My experience of the death-by-PowerPoint merchants in the workplace is they tend to be the ones who generally dabble in comedy Friday ties, leave post-it notes in office kitchens complaining that someone used their milk (because they had drawn a line where the level was when they last used it) and they always seem to schedule meetings for 9am.
Of course PowerPoint has it’s flaws and, as with Excel charts, the defaults, templates and wizards are generally awful. But if you look past this you will find a superbly useful piece of software. There are alternatives, such as Keynote and more recently Prezi (which I’m going to discuss in a separate post), but the key issue is the significant lack of capability in being able to present information and communicate effectively.
Data + Message + Display is a simplified formula that explains the key components of effective visualisation. In this example, the data was clearly available with regards to the activities, nodes and relationships of this situation. With my OR hat on, the display chosen, a cognitive map/system dynamics diagram, is a common approach to presenting complex situations like this. And it is the words ‘complex situation’ that are important here because the purpose of this presentation was to present the complexity of American military strategy. This was the message and therefore the display largely succeeds. There’s no doubt the actual execution of the display could have been improved significantly – such as reducing the visibility of the lines, being consistent with the nodes as text labels or boxed labels rather than mixing both, and finding a better way to position the bold category titles (if you look closely they actually sit on and obscure key nodes on the diagram) – but fundamentally the data, the message and the display are coherent.
If you wanted to explore the detail this wouldn’t work, you would need a reduction in the complexity by breaking down the system map into more digestible thematic clusters. Perhaps the senior military officers bemoaning this slide should think back to what they actually commissioned.
End of rant.
Now who’s been using my milk…
It is disappointing to hear about the escalation of the stand-off between Adobe and Apple with regards to Flash, the multimedia platform that enables designers to develop rich, interactive web content. The latest episode being the news that Adobe are no longer pursuing the development of a tool that will allow Flash developers to port software into iPhone or iPad apps.
Whilst Flash is supported by many smart phones platforms, the Apple iPhone and iPad application market is clearly dominant and so any technological exclusion from this will be very frustrating for developers. Some observers have been anticipating the decline of Flash as other technologies emerge and become prevalent and compatible with evolving web standards. This development will only fuel this side of the debate.
Flash continues to be a ubiquitous and unquestionably vital tool for creating effective visualisation solutions, particularly in relation to unleashing the power of sequence and animation to a communication. It will be damaging and hugely restrictive if this ongoing conflict between these two powerful technology brands grows into a VHS/Betamax-type division.
As I have mentioned several times before, my post strategy for this blog aims to minimise lazy repackaging and regurgitating of posts from other sites. It gets very boring when you end up reading about the same article several times across different sites and so I don’t want to add to this.
However, sometimes there are posts from other sites that are really interesting (at least to me) and may be of interest to visitors here. Therefore, each month I’m going to compile a single post which contains links to such items that I have been published or I have encountered during the previous month. It will also allow me to be a little bit more flexible with the subject matter boundaries I normally tend to stay within.
To start with, given it was the month I launched this site, I’m going to list my key links for February:
Smashing Magazine | Applying mathematics to web design | Link
Eager Eyes | Discussion about the future of browser-based visualisation | Link
Ted Talks | Video of Jamie Oliver’s well-received anti-obesity talk | Link
Junk Charts | The art and science of chart selection | Link
Flowing Data | Review of the book ‘The Wall Street Journal Guide to Information Graphics’ | Link
Visual Business Intelligence | Detailed critique of the book ‘The Wall Street Journal Guide to Information Graphics’ | Link
Eager Eyes | Review of Tableau Public | Link
Ted Talks | Video of Temple Grandin sharing her ability to ‘think in pictures’ | Link
Ted Talks| Video of Pawan Sinha on how brains learn to see | Link
McKinsey Quarterly (from Jan 09) | Fascinating interview with Google Chief Economist about the future of web technology and innovation, and the next decade’s “sexy job“… | Link
Don Norman’s JND (from Dec 09) | Don Norman’s critical article about design research in relation to innovative breakthroughs | Link
History Shots | Information graphics telling stories about historical events and periods | Link
Flowing Data (from Jan 10) | How to make a heat map | Link
Smashing Magazine | A useful roundup of print design tutorials | Link
Bit of a tangent from normal posts but I was interested in this promotional campaign by IKEA for their kitchen design service. They have rendered three kitchen layouts and designs to represent the three main political party leaders.
Aside from the comedy names and regardless of whether they necessarily accurately represent the leaders traits, I was interested in the challenge of identifying and translating personal and political characteristics into particular designs and renders. The ability to create visuals that effectively encode qualitative data, as seen in these examples, is a fundamental requirement for information designers. Immersing yourself in solutions like this is an important part of an ongoing development towards mastering the art.
With the UK election process very much reaching its peak in terms of anticipation, discussion, debate and spin I thought I’d follow up my earlier post, which introduced some of the quantitative analysis and interactive scenario tools, to present some examples of qualitative analysis. This area of visualisation is rapidly growing in prominence as media outlets exploit the incredibly rich source of opinion and discussion services such as Twitter can provide.
In partnership with Lexaltyics, this is the BBC’s graphic for tracking the sentiment of Twitter users towards the three party leaders during the course of the live TV debates. The magnitude and direction of the sentiment scores are generated by algorithms which assess the nature of the language being used about each party/leader. Of course this is not an exact science and the results should be taking with a large pinch of salt, particularly as they are vulnerable to ironic abstractions such as the ‘its all Nick Clegg’s fault‘ campaign…
The graph below is taken from last week’s second debate event on Sky News.
Assessing the graph reveals a number of flaws. Apart from the annoying title, logo and the clumsy positioning of the legend key, the main problem is the lack of prominence of the ‘zero’ line which divides the sentiment into a positive or negative direction. There is little or no visual prominence afforded to this and it is made more ambiguous by the useless banded shading. For some reason the vertical axis goes up to 9 when there is no value much over 3 and the minimum value range is -6 which creates a visual inbalance. I also think there is a missed opportunity to aid insight by neglecting to provide useful context. Annotating the timeline to highlight the milestone moments of each question being asked would have been far more useful than just seeing the time across the x-axis. This would facilitate interesting analysis of how each leader was perceived to have performed on each question.
An alternative approach to assessing sentiment is offered by Tweetminster. As well as collecting and collating sentiment scores based on language, Tweetminster’s flagship analysis focuses on the volume of mentions and references made about constituencies and candidates, creating a buzz/word-of-mouth assessment to form predictions for the eventual election results for each party.
We’ve had analysis of who is mentioned in most positive terms, who is talked about the most and now we have who we’d most like to slap via the Slapometer which allows site visitors to vote with the back of their hand.
With all the complicated and advanced algorithms and data gathering devices represented by all the other devices presented here and in my previous posts, could it be that this crude data collection approach will prove the most reliable? We’ll have to wait and see…
Throughout the financial crisis broadcasters, politicians, journalists, authors, academics and other communicators have tried various techniques to try help make their message come of this complex subject as accessible and as digestable as possible.
One of the key attributes of effective communication, as suggested by Dan and Chip Heath, is to make the message concrete. That is to reduce its potential abstraction by structuring the information in alternative ways that helps people relate to it or consume it in a more successful way.
An article in the online BBC magazine today takes this approach by breaking down the headline facts around the UK’s finances into units of ‘Tescos’, equivalent to the retail giant’s £40bn annual sales.
In today’s Guardian are details about some of the interactive tools that will be available for viewers who watch Thursday evening’s first prime-ministerial debate at www.itv.com/electiondebate. It describes a range of “sentiment” tools that will be deployed to measure live responses to the discussion between leaders of the three main parties: Gordon Brown, David Cameron and Nick Clegg. In visualisation terms the most interesting of these devices are:
This will track posts made by 5,000 Twitter users and label each tweet made about the leaders based on a sentiment scale of 1 to 5. These values will then be aggregated each minute to form a dynamic ‘score’ of each leader during the debate which will be presented with a directional arrow alongside a picture of each leader.
Alongside a live stream of the debate an audience reaction tool will track the views of a panel who will use handheld devices to register their impressions. These will be displayed using a worm graph.
Having immersed myself in the world of data visualisation over the past 3 to 4 years, it’s emergence and continued growth is very clear for me to witness. My personal discovery of new blogs, papers, books, technology and, above all, energy around the subject is testimony to this.
On my ‘about’ page I reference a quote from The Guardian’s Charles Arthur back in September 2007:
We’re always being told that our computers are more powerful than ever before. Now, perhaps, we can start to use them to make better sense of the world about us. Visualisation doesn’t have to mean fancy eye candy; it can just be simple, elegant and informative. Look out for it.
I use this quote because I think it nicely encapsulates the purpose of visualisation. However, 30 months on I sense the field hasn’t progressed sufficiently to successfully bridge the still significant disconnect between the advocates and the potential beneficiaries of data visualisation.
In part this is due to the ongoing absence of clearer boundaries and definition of what data visualisation is and (perhaps more importantly) what it isn’t. Of course, anchoring and boxing-in a subject is undesirable but there is nevertheless a need to continually evolve a common terms of reference. This is a matter passionately discussed by people like Stephen Few (most recently here) and especially Robert Kosara who has written about this issue in his most recent item:
Visualization is not a very clearly defined field. There are many variations, ways of doing it, and ideas around it. That is valuable, because it keeps the field moving and brings in fresh ideas. But it also brings with it people who like using visualization’s tools and talk about visualization, but what they are doing is something else. We need to start calling these things what they are: a cargo cult of visualization (read on…)
A second prevalent factor is the mutual back-slapping that occasionally can take place between practitioners in this field. This results in a certain amount of complacency that just because we all get excited about the potential of this subject and its applications, everyone else should too.
In these times of financial challenges across almost all economies, the likelihood is that data visualisation type solutions, unfortunately, will be perceived as ‘luxury spend’ activities by many organisations needing to cut expenditure (particularly consultancy) and to simply focus on their core business. Whilst it fine for all of us inside the business to recognise that it deserves a higher standing and can be deployed to support economic recovery, that is ultimately insufficient. It is vital we all strive towards a better articulation of what data visualisation is to help provide due justification and explanation of the potential tangible improvements for all.
I’m therefore going to be particularly looking at posting items that a) help pursue this idea of a stronger definition and b) help to articulate the tangible need for data visualisation in organisations. If anybody has any suitable content to contribute towards this please feel free to stick then in the comments.