Steven Johnson opened dConstruct 2008 with the interesting story of Dr. John Snow. Snow created a map showing the clusters of cholera cases in the London epidemic of 1854. By doing so he proved that cholera wasn't spread by smell like people thought but the water from a contaminated pump on Broad Street. Snow showed how data visualization can suddenly reveal useful information that wasn't visible before. Johnson called Snow's map the ‘first mashup’ which sounded like a nice tribute to what Snow's discovery meant to the world.
Johnson explained how is trying to map local news from different sources with his outside.in project. While the idea sounds great and the screenshots looked good I'm not really convinced yet. Sure, this may work fine in Manhattan or Palo Alto but not everyone writes in English. Even when they do you'll still need that ‘critical mass’ in order to collect something useful. They also did some research to find the ‘blogiest’ neighborhoods in the United States, a great idea which I would like to try in Belgium as well.
Aleks Krotoski tried to bring the game industry and webdesign world closer together. Aleks has an incredible enthusiasm but for me the topic wasn't that interesting and I felt it missed a story line.
Joshua Porter went on with social biases and heuristics: how a language can influence decisions or advanced psychology for webdesigners. For example: people tend to be more afraid of losing something than winning, known as the ‘loss aversion’ or how people often do and believe things because many other people do, the ‘bandwagon effect’. Porter referred to a paper which I'm planning to read at some point: “Judgement under Uncertainty: Heuristics and Biases”. An interesting topic but it could use some of the enthusiasm Aleks had in her presentation.
Daniel Burka, the creative director at Digg and on of the founders of Pownce, spoke about the ‘cohesive user experience’ or how Digg tried to allow users to participate on the site without requiring them to follow a lengthy signup process. Nice to have a little insight in the development process behind a huge site like Digg.
Tantek Çelik followed with his microformats presentation. I expected a great deal from his presentation, maybe a little too much. It was a little too advanced for those who have never heard of microformats before and not advanced enough for the others. Maybe I shouldn't have bought that microformats book on last year's edition. ;)
Next up: Matt Jones and Matt Biddulph. My opinion? Splendid! Those two Dopplr guys gave the most hectic presentation of the day and I loved it. They talked about how Dopplr could be as little intrusive as possible, a connector between different social networks and passing on the data to the next (web)app. They explained how they preferred to grow slowly, trying to build up trust first as they seem to understand where you are at a given point is sensitive data for most people. My Dopplr ‘friends’ (a word they specifically refuse to use) tripled in a single day which shows how everyone gave Dopplr a second look after their presentation, nicely done Matt & Matt.
Jeremy Keith closed with a science fiction story which I didn't really understand in the beginning. He tried to prove that the success of social webapps are unpredictable, even though we are likely to think we can. It was an unusual presentation that felt confusing at the start but became better and better towards the end. His point on the strength of weak ties was something completely new to me. Another book to read.
This post is open source. Did you spot a mistake? Ideas for improvements? Contribute to this post via Github. Thank you!