IA Summit '08 - Miami!

The picture - it is just one of the many deliverables on the "Wall of Deliverables" at the IA Summit. Oh and also - special thanks to Puerto Sagua in Miami Beach for providing me some of the best cuban food I've ever tasted.

First off - just to further elaborate on my post about Jared Spool's keynote now that I am typing on an actual keyboard, the discussion of process > methodology > dogma was interesting. Spool's company (firm? consultancy? - sorry, not sure offhand) did a study to try to examine what distinguished successful teams. Their finding was surprisingly that a project's success has nothing to do with how formally their methodology was followed. In other words, all successful teams have a process, but their success had more to do with what he termed "tricks and techniques" - ways they had of accomplishing individual pieces of the project most efficiently - his example was the plumber who banged on a pipe with a chunk of metal to fix a problem. When he asked the plumber how he know to do what he did he said "I didn't" and when he asked him if that was the right tool for the job the plumber said "No". Then he asked him why he didn't go to get the right tool first he said "I didn't feel like going to my truck". In other words, sometimes the efficiency of getting something done, maybe using less than formal methods, sometimes helps the overall project. It maintains a rhythm and pace to the project.

Some more random notes from Jared Spool's session:

• Best teams don't have a methodology, focused on techniques and tricks


• Struggling teams follow methodology without results


• Tsa = "thousands standing around"


• What inference to make from eye tracking, analytics (he exposed analytics tools as sheerly ways for consultants to have fun. However - he also said this isn't always bad if it is what gets them "to come to the lab"


• U(ser) (e)X(perience)=vision, feedback, culture

Session: IA Tools/Content Analysis

Semi-random notes:

• Axure - kick ass prototyping tool


• Semantic analysis breaks content down to building blocks and allows systems that reinforce business processes


• Connexor: natural English parser


• Calais: Reuters service to provide stuff like what you write based on semantic analysis (didn't quite understand this)

This was Matthew Hodgson's session. He went through a case study of doing semantic analysis on some medical information regarding symptoms and treatments in order to codify it and enable the creation of a backend administrative interface that would standardize the input of the data without sacrificing any of the richness of the content. By basic subject/verb/noun/clause analysis of representative samples, they were able to create some basic maps and wireframes based upon the samples - by using the actual words as the nodes on the maps, they were able to map the relationships of the information and create order out of widely disparate sources.

Unfortunately, I need to actually do some work and can't type this up all day - but I do want to get it out there. So here are the rest of my notes with some mild clean up:

FOLKSONOMY SESSION

Exploratory search:

• hierarchical


• faceted (set of smaller hierarchies) Flamenco


• dynamic clusters (dynamic, post_retrieval, unique/clustering algorithms - good example is automation) - Vivisimo is clustering platform, Grokker and Clusty


• folksonomy - Motivation uncertainties, Activities: searching(cognition), browsing(perceptual) - flat and inclusive structure
  

USER EXPERIENCE TEAM OF ONE
Leah Buhley(adaptive path)

Generative design
1 Brainstorm a lot
2 ad hoc team
3 pick the best idears

Techniques:

• Spectrums, continuums and grids (design)


• Mix and match terms


• Inspiration library


• Screen grab(firefox plugin)
  

Concept sheet:
- description
- draw a pic
- it's like
- it does

Business needs are good, but so are user needs

PLACEMAKING

• Sociably, uses and activities, access and linkages, comfort and images


• Map markup 'slams' 'charrettes' (on site)


• Cheese example - 2 people looking at cheese and also viewing each other's reactions at the same time - triangulation
  

SEARCH PATTERNS
Peter Morville

Songza
Oskope
Like.com
Live search (msn)
Spime search for physical obj (bruce sterling)
Everyzing

PROTOTYPING

paper!
XML/XHTML
Dreamweaver
Adobe flex
Fireworks
Adobe thermal?
Irise
Axure
Michael Schrage - prototyping books

And finally - if anyone actually read this far, Andrew Hinton's "plenary" (uh I guess that means closing speech), summed up perfectly where IA and UX are today. He had some amusing visuals and some interesting AK-47 metaphors, but what it came down to is that the web has evolved from an extremely hierarchical and abstract organization which was only fully understood by the engineers, now it has become a vast and complex, chaotic being in terms of its physical organization. New interface technologies combined with new ways of interacting socially and users contributing to the meta data of a site provide new opportunities for site navigation and organization. This all is serving to blur the links between user experience designers and good information architects - such that any good ux designer is an IA and vice versa.