The 5 rules of paper prototyping as delivered at break neck speed by @harryharrold and @ruperredington from Neon Tribe!

1. It has to be paper

2. It has to be a prototype

3. You must test it with real users

• Pair testing can be useful to get suggestions from users about how to make improvements.
• But if you have more than 2 users, it becomes a focus group. Avoid!

4. You must make one change to your prototype using scissors during your testing

• Validates the user involvement.
• Users feel they can suggest changes because it looks unfinished.
  
• Can turn into a co-design session.

5. It's always your fault

• Empower users by saying it's all your fault - it's us not you
• Lab tests put users under scrutiny so they feel the software are their failures
• Paper is malleable

A discussion followed. These are some, but not all, the points raised in the session.

Paper prototyping isn't hard. It's quick and easy, cheap and accessible.

Very tangible e.g. you can tear, cut and shuffle paper.

Could be tricky to prototype some interactions in paper e.g. gestural interface of iPad.

Don't wireframe too early (if at all) as they commit you to a design.

Should lo-fi wireframing tools be used in place of paper prototypes? There is a risk that users might interpret anything on screen as being a more 'finished' product than if you use paper and be less likely to give useful feedback.

Sometimes doesn't work. E.g. you can make some interactions seem more onerous than they actually are.

A couple of resources on paper prototyping:

http://www.paperprototyping.com/

http://rosenfeldmedia.com/books/prototyping/

@pubstrat explained the context for the session.

Traditional policy and legislative development may not deliver the optimum customer experience. How do we marry up Agile principles with policy development?

@curiousc talked about breaking risk down and making it manageable. How do you re-train the politicians and media to not want the big story impact? It was noted that at the same time they don't necessarily want a big story if it's a failure! @curiousc outlined some points as follows:

• Agile is a commitment to communication.
• Everyone is responsible for making it work.
• In traditional policy making the plan becomes the objective, how do you separate them out?
• How do you adjust policy as you go along based on evidence? How do you use the users to make sure you are on track.

Some of the potential barriers and issues of going Agile in government were discussed. This doesn't by any means cover the whole discussion, just some key points I picked up.

Policy people are not good at changing their minds and tend to bury evidence where policy implementation hasn't worked well.

The current process helps policy makers to avoid facing possible failure and huge costs further down the line. Better to fail fast, early, cheaply. Failure can be positive. Can sell failure on the basis of cost savings.

You could surface issues to Ministers by saying you're going to use Agile approach.

Separate the plan from the objective.

Could a policy making process be tested out in a more Agile way? Hard to test small atomic bits of a large policy in the same way you test code.

Is Agile crossing over with co-production?

Need to test policy across a wide range of demographic groups.

Policies need to be delivered in a way that suits a networked, online society. Transparency is key. A digital footprint means that you have to deliver policy outcomes in line with original policy statements.

@willperrin introduced the session by showing a variety of local websites that Talk About Local have been involved in supporting communities to set up. For a map of hyperlocal sites see Openly Local.

Most of the people who Talk About Local work with are of an older age group e.g. over 60's.

@hughflouch talked about Networked Neighbourhood's research into local neighbourhood websites.

Are communities thinking about resilience of hyperlocal sites? @willperrin pointed out that WordPress.com is a very resilient service. But resilience of people is also very important - volunteers can keep things going.

The Localism Bill is all about de-centralisation of services. There are specific changes to the planning process in the Localism Bill which are quite radical and will require local democratic involvement.

What is the attitude of local government and the local media to hyperlocal sites? The landscape is changing. A few years ago local government officers wouldn't have been interested, but now some are on board and others coming up to speed.

@hughflouch highlighted findings from research about the benefits for local people and local government of engaging with local websites. There are different models of engagement and these will be detailed in the forthcoming SOCITM report.

Involvement of public services in local websites is dependent on their perception of the local area.

Example of using Cover It Live (live blogging) for police engagement.

@willperrin suggested looking at Damian Radcliffe's slides on hyperlocal and also the Talk About Local guides and tips.

Discussion turned to local media and fact that many local newspapers are funded by local government notices about planning. Is there a possibility to offer space on hyperlocal sites in a similar way? @willperrin pointed out that very few hyperlocal sites make money or need to make money to exist.

Scope for including PDFs on hyperlocal websites. @hughflouch mentioned that members and council officers have an obligation to engage a representative population so will still revert to offline engagement, producing PDFs etc. I mentioned a finding from my own MSc research - some council officers are not willing to engage with local websites because they feel they are not representative. This point was also picked up on Twitter by @sharonodea.

Important for local government to work in partnership with other public services (e.g. police, fire, health), community and voluntary sector, parish councils etc.

@getgood said that people who are not literate in IT can email articles to her to put on her Digbeth is good website.

@podnosh made a great point from his work with neighbourhood managers, that those with the right personality have successful hyperlocal blogs. You can't force people who don't have the right personality to be conversational online.

@podnosh talked about social media surgeries being a place where volunteers have a space to ask questions about social media, look and learn. Look at http://www.socialmediasurgery.com/.

@willperrin people never believe they can set up a website. Especially older people. @podnosh said that hyperlocal work is laying the ground for future skills in data matching open data on the web, by the type of women over 60 who run community groups.

The first session of the day - all about how to fix consultation.

@lesteph introduced it with three questions:

• Why haven't more engaging approaches become mainstream?
• What kind of approaches work well at answering the questions policymakers actually have?
• Is consultation as a political and bureaucratic process too screwed up for digital to try and solve?

Below are some notes from the discussion. I definitely missed a few comments.

Need to be realistic about how consultation processes works. Especially when there isn't enough time.

There isn't the budget or will to maintain resources in house to build cheap WordPress tools and procuring them in is expensive.

Important to layer consultation. People have different motivations and levels of engagement. Some people only want to do the equivalent of 'like' something on Facebook.

Crowdsourcing exercises have been difficult. Raises expectations and can attract unsavoury views. Required considerable moderation resource. Led to some small changes but perhaps needs to be more focused.

Open ended consultation data (i.e. freeform responses) can be hard to interpret and with limited resources it's often easier to take the easy, structured route.

Need to pilot consultations with target audience.

Important to involve statisticians early on.

Interesting discussion about how to develop participation and democracy around consultation issues.

Opportunities to focus on how to present information to people to help them understand what you are consulting on. Improved transparency.

Balance of digital engagement and formal consultation.

Important to think about how people are engaged after your consultation.

Need to build in 'low-tech' consultation so not to exclude minority groups or those who are digitally excluded.

MySociety and me (aka confessions of a slightly lame volunteer)

Joe Lanman was explaining that he works as a volunteer at MySociety, a charity that makes web applications to help people engage with government

How it all began. When Joe first moved to London he used http://www.theyworkforyou.com. He found his local MP was Diane Abbott and could see her voting record. Then when the expenses scandal broke Joe used http://www.writetothem.com to write to his MP and found they were developed by the same organisation.

He got involved in the IRC chat where MySociety discusses ideas and makes decisions. Joe wanted to get involved with them and found out while they have lots of applications for central government they don't have much for local councils.

MySociety scrape data from local government, so Joe got involved. Then Joe got involved in the design of icons and thought there was an opportunity to get more involved in the user experience of the applications.

Joe attended a MySociety public meeting to discuss some ideas about user experience. Next Joe attended a planning weekend.

MySociety's core staff is made up of developers, so there are no designers except the volunteers. Joe put forward ideas about research, usability testing and found some challenges in figuring out how to engage with a predominantly developer led organisation.

Joe enlisted some help from Leisa Reichelt (http://www.disambiguity.com/) who is working on UX design for Drupal, another open source project and Johanna Kollmann to try and propose how UX Designers could get involved to improve e-democracy applications.

We had a fascinating discussion about how UX Designers could get more involved in hack days such as http://rewiredstate.org/, http://charityhack.org/ and http://www.sicamp.org/ so that applications are informed by user research and design activities. Some of the issues we discussed:

• user research and design takes too long so can't easily be incorporated into hack days and documentation takes too long
  
• developers have immediate access to make changes to an application by adding to the code base - how do UX designers get similar access?
  
• In terms of privacy/ethics would it be a problem to put participant data up online in a public space?
  

Some suggestions that were proposed:

• UX Designers do a pre-hack day
• Use guerilla research and user testing techniques - e.g. taking a laptop out into a café and recruiting participants there and then

Animating sketches

The session was led by Chris Neale (@chrisneale)

The session has started by discussing issues with wireframes and sketching.

Problems with wireframes

• Static vs. dynamic
• Speed - easy to get precious
• Sometimes too abstract e.g. black and white
• Not collaborative
• Graphic designers skin them / colour it in
• Version control difficult
• Can't see the interactions that lead from one screen to another - hard to communicate to clients and developers
• Detail can be overlooked

Chris demonstrated a new tool/technique his design team developed when he got fed up of doing 32 iterations of wireframes and had trouble communicating to the person who develops prototypes.

Enter the Animationizer!

The Animationizer is basically a way to do stop frame animation using wireframes. A high definition Logitech web cam is mounted on a shelf bracket on a tripod.

The web cam connects to a laptop running the Animationizer software that Chris is giving away on his blog: http://e102.co.uk/?p=13

Chris also uses post-its, sharpies, card, scissors etc. to do the sketching

When you start, position the prototype and press the space bar. Each image is stored as a file so if the software crashes (as it tend to in a live demo!) you can stitch together the separate shots. The software also has 'onion skinning' so you can the last shots you took.

The purpose of the animations is to get sign off from management on the interactions. Then they are given to the developer who creates prototype.

Chris proceeded to do a live demo of faceted navigation for shoes. The facets were:

Materials
Size
Style
Colour

Chris has some useful tips e.g. you can use acetate to draw circles round selected items to demonstrate 'on clicks'. Or use cardboard icons/symbols. Spray mount can help lock down things you don't want to move much. Someone else suggested pritt non-permanent markers.

Chris showed how you would step through the animation very quickly. He showed us a video of an animation he'd done earlier in the week to show how a faceted search would work.

Once compiled the animation speed can be altered using screen capture in e.g. Silverbackapp.

Chris suggested introducing the technique into an organisation gently as it can come across as a bit weird!

In summary the technique can be used for communication and discovery and helps designers visualise if something is going to work before building it.

From post-its to personas

Slides from this session

Lee McIvor talking about a technique based on Alan Cooper's method which enables you to use your research data to drive and inform development of personas.

It's important to get buy-in to use your personas throughout the project. Therefore they must be based on accurate data.

You collect a wealth of data from interviews, ethnographic research, user testing etc. How do you make sense of it?

Pick out behavioural traits

• What do users do?
• Why?
• How does it make them feel?
• How do they think it works?
• What are they trying to achieve?

Lee said he usually pulls this info. out of data onto post-its - sometimes a different colour for each participant.

Cross-case analysis. Find the patterns and collect broad groupings e.g. using affinity diagrams.

The next step is to define behavioural "scales" in terms of the behaviours exhibited. Sometimes you come up with a multiple choice e.g. 2-3 types of behaviour, but quite often a scale. If your scale is based on say frequency a user interacts with a product in a day you can see on a continuum where your participants sit.

Enables you to see where groups exist and identify personas based on those groups.

Pattern matching

• Map feedback to the scales.
• Relative positions on the scale matter, not exact positions.
• Groupings start to form to show relationships between participants along the scales. Compare how a participant is grouped on one scale with another scale.
• Look for persistent groupings and base your personas on that. Be aware of basing a persona on two groups who exhibited different behaviours when you did your research.

Lee said that this is a technique that can be learnt quite easily. He said it can take anything from a one day (10 participants) to a couple of days (20) to a week, depending on to what extent you involve your colleagues as a collaborative process.

Be aware of anomalies. These can be incorporated and are interesting but may not fall into a usage pattern. Not all behaviours will correlate - people are unique. Doesn't necessarily affect you overall patterns.

Define goals

You need to define goals for the personas as well as behaviours. What do they want to achieve?

Developing personas

Now you have your behavioural traits, quotes from the data and goals you can create your personas. Lee mentioned not to forget using scenarios.

Discussion

Someone asked how you ensure the personas get used and don't just sit on the wall. Lee stressed that you can get clients involved in this process and they are generally impressed with the technique.

However it's important to ensure designers are also involved early on and have visibility of this process.

@alphabUX mentioned that personas can be mistaken for 'average users' not users who represent the extremes.

Resources
The inmates are running the asylum - Alan Cooper
Designing for the digital age - Kim Goodwin
About Face 3.0 - Cooper, Reimann & Cronin
http://www.cooper.com/journal/personas

Slides from this session.

Michelle Adams ran a session on Agile and UX. Michelle said that she is an in-house UX team of just one!

Michelle ran through the Agile manifesto and Scrum process and talked about reducing down documentation and increasing face to face communication.

Michelle said that one of her challenges is knowing what her role is in the team. The UX person can potentially be the product owner or a member of the sprint team (pig) or a consultant/advisor to the team (chicken). Michelle usually works as part of the sprint team, working a sprint ahead to produce artefacts for the next sprint. She works closely with the product owner and attends current sprints. Michelle recommended pushing to be part of the sprint team (pig) so that you have a voice in the process (as chickens can't speak in sprints).

Challenges and opportunities for UX

• One of the real benefits is collaboration with the team. Everyone having a say in the design/development.
• Need to get a balance between working ahead and working within the sprint. Worth getting to know your dev team to find out where the best balance is - 'big design up front' vs. 'just-in-time' design.
• Flexible deliverables - reduces down the amount of deliverables because working in rapid cycles.
• Don't always get to do as much user testing as could do because of access to end users and incorporating feedback isn't possible in the short timescales.

Michelle had some useful resources in her slides.

We talked about the need for flexibility - developers and UX people need to be willing to compromise and meet in the middle.

@adrianh mentioned a Ken Schwaber video that talks about levels of detail and when it's appropriate to go into details.

Someone asked how Michelle protects herself from demands from the rest of a business when working in sprints. Michelle said her manager is very supportive and understands that she can never be 100% on a sprint.

There was a discussion about how to ensure UX value is built into Scrum. Enhancing features is often seen as unnecessary and new features are generally more appealing. Michelle suggested that if there's anything big or controversial - put in a 'test and refine' task in next sprint.

Tom mentioned that you need a design literate product owner. This chimed with me as someone who also works client side.

There was a discussion about how you feed what you learn back into sprints. Someone talked about seeing sprints as experiments. @adrianh also talked about 'the spike' - a small phase of experimental development.

If anyone is interested in doing a talk on Agile and UX at the Cambridge Usability Group please contact @micheleidesmith

Session on Interactive TV

This is a live blog post from the first session I am attending at #uxcamplondon led by Patrick.

Patrick is explaining the interface restriction for interactive TV. He is stressing that interactive TV has a simple interface and it's important to be easy to use and navigate, clearly visible and legible, usable with a remote. You can't use certain colours due to flare, can't use thin lines.

There are four main broadcasters but limited access. Web access is minimal.

Some style guides e.g. BBC. Little good research about interface standards. Tieresias is an industry standard font developed by the RNIB.

All services have different middleware, so you have to develop for each one. Standards exist, but haven't been implemented across the board.

Project Canvas is a UK partnership project with various broadcasters/providers involved "to build an open internet-connected TV platform, subject to BBC Trust approval" (quote from http://www.projectcanvas.info). Will have standardised middleware.

Patrick used the terms 'lean back' and 'lean forward' to describe different user behaviours. Not something I'd ever thought about! But assumption is users are 10ft away.

We are looking at some examples of how Twitter can be integrated with interactive TV - but each tweet has to be moderated. I wonder how many people would use the interface via their remote rather than using their smart phone.

Now looking at some example interfaces e.g. for selecting videos. The buttons that correspond with the remote have to be very simple.

The remote is one of the biggest problems about designing for interactive TV. But there is potential to network with other devices using an IP connection.

The discussion then started to focus on how people are actually using interactive TV, the use of other devices alongside.

Norfolk Home Page session

Led by Shane McCracken of http://gallomanor.com/

Overview of the Norfolk Home Page project

How do you get hyperlocal activity happening when you don’t have someone like Will Perrin to start a hyperlocal site?

Doing a project in Diss based around the local library.

Diss has some local sites but e.g. the Diss express, a local councillor website and Badminton group. But there are no www.kingscrossenvironment.com sites or www.pitsnpots.co.uk

Wanted to bring all this local information together, so decided to develop an RSS ‘thresher’, essentially a web service that enables you to combine RSS feeds from different sites into different categories. Admin tool enables you to delete, promote items from the feeds.

Items which come through the Diss Express are not necessarily relevant to people in Diss (e.g. stories from Eye) so can be deleted.

RSS sorting machine gives you the ability to take the rubbish stuff out of feeds and promote the good stuff. Then you can output in a front-end e.g. http://diss.norfolkhomepage.org.uk/

Libraries sort the RSS. Interesting to see what they choose.

The purpose of the site is to bring together the disparate information sources and bring together where it can be found. People have ability to click through to the stories, standard WordPress functionality so comments can be moderated. It’s supported and safe.

Because local council manage it there is more chance of getting elected representatives involved.

Diss a very sleepy place. Not much existing  information to aggregate. Now moving it to Great Yarmouth and training library staff.

Still lots of battles to fight at a local level. E.g. when doing a demo found the internet connection was really slow.

Lessons learnt:

Lack of understanding of benefits.

• Hard to explain the reasons why they should recommend that local sites are developed using RSS, updated regularly etc.

Blog is still a perjorative term

• People don’t understand what a blog is.
• Showed lady that blogs are not boring and can have different designs, pictures etc. e.g. http://beautifulbirthdaycakes.co.uk/ !
• We are still a long way from people getting past the barrier of what they think it’s about.

Need more support for training

• Lacking resource.

Thing that had the most impact in Great Yarmouth - showed how if you search for Amateur Dramatics Club in Google, but nothing came up. If you knew it was called the Mere Players you would find straight away. Because the site is built in Flash it doesn’t rank in search engines.  Compared with the Badminton site that uses RSS and it comes up top in Google. WordPress is very search engine friendly - http://beautifulbirthdaycakes.co.uk/ is getting 2000 visits per month.

Note:

I didn't have a chance to write up the discussion bit because I was getting tired and also contributing to the discussion. But we had a really interesting discussion about the barriers to getting local communities in very rural areas active in the hyperlocal scene.

I wrote up the list of words we decided should be banned (!). http://micheleidesmith.posterous.com/list-of-banned-words-localgovcamp. We felt these words are barriers to getting people interested in, and using, technologies that can increase their involvement in their local community and as a result, improve their local area.