Putting users first
(Came in a bit later to this one. Sorry.)

User is a dirty word. Focus on people. Me. My interests, my services, my devices, being able to connect my information. But all the users are thinking about is me, and what interests me. Whether they have created the information or not, they think of it as mine, and it's all about me.

When we think about real people we need to understand their real desires, wants, needs. Think about Maslow's hierarchy of needs. When people want or need information, resources, object, they expect them to be there, to be usable, re reliable. We need to think about user's needs.

Focus not just on people but real people, including the 95% of people who don't live their life on the web, probably not anyone who is in this room. We live much of our life on the web. Think outside the alpha and beta users.

We haven't made it easy for information to be portable so people can use it in their real life, away from the computer, away from the browser, and are in the shop.

Technology pain: syncing devices, syncing services, suffered by real people not just alpha geeks.

Biggest tech pain is refindability - it's one of the fun things if you are evil and say "Have you ever had a time in your life when you had to find a piece of information and you knew you'd seen it but you couldn't find it, can you tell me a story about it." Everyone has a story, everyone suffers it. Tools are addressing that, social bookmarking, it's about context.

Taste. If are looking for cake, is it a big American-style cake, or a continental tea cake. Mahalo, for example, filters based on its editor's taste for Martha Stewart cakes.

Been on the web since 94 so established, but everyone who sets up for the web now, has to set up a lot of profiles. Everything is all about me, so people want to use profiles, but they don't want to repeat a lot. Want portability. Need ease of use.

Portability, move information where we need it, in our pockets, on our dashboard.

Privacy. Some people think that privacy is a lost cause and we should get rid of it. But we really do still need privacy, and need to get smart.

Attention, focus, energy are limited. We only have so much of it. We need to start filtering. We are doing the same stupid things over and over and over.

Need to ease tech pain. Tagging, and other features that are easy to implement.

A few different tactics: favourites or flagging - very easy to do. Identity, have to log on so most apps have identity, but also can do more and link people together. Tagging, takes a bit more work but helps you understand your own context and helps with refindability, and is really good. Ratings, takes a bit more effort, but sometimes people spend quite a bit of time figuring out how many stars to give something. Titles, things aren't always well titled, and need to be more informative to be valuable. Abstracts and long text annotations take a lot more work.

Tagging brings up the 'F-word' - Folksonomy. Result of personal free tagging of pages and objects for one's own retrieval. Saves refindability for me, but also for people with similar interests, terminology, taste, etc.

Tagging is usually done in social event, shared and open to others. But not always - private tagging is just as valuable. Done by person consuming the information so provides a point of reference for them.

Object, identity and metadata triad. Add community, people who use the same terms on the same object. But vocabulary becomes terminology when you have a community. Then it becomes a culture when in a community because it's not so tightly defined, it's fuzzy.

So if you have an objected identified by a tag and we want to find other things we can use the tag for it. Can find without having to re-see all the things we saw before because the metadata shows us things that are alike but new.

(Goes through Magnolia and the way that they use tags, bookmarks, etc.)

It's all about sharing and social. Sharing and social is how we got out of caves, it's how we move forward as a society. We need to be very reticent about how we think about this as we move forward. As we look at social systems out there, they don't give us a good way to actually create new friends.

Sphere of sociality, i.e. things private to you at the centre - personal infocloud; friends that you have a private conversation; collective - everyone on the service, e.g. on Del.icio.us you can see everything on the public system; outside the system there's a mob that doesn't get it.

Directional sociality - relationships are not equal, can be one-directional. Unequal access at granular level.

Real relationships can break down into smaller categories, public sharing, private sharing, listening. Social software needs to understand this. but want to be contextual as well, e.g. "only want to pay attention to friend A when they talk about thing B".

Jaiku good for granular listening.

Want to be able to do things with information - Facebook, you can't do anything with what you put into it. Twitter lets you favourite things, so you can remember and re-find things that people say. Need to hold on to good things that have been said that we appreciate and like.

Ease of use. Web pages need to be easy, portable, need to put the information into our life. Stikkit gives you a chance to save snips of information, recognises date strings and puts it in to your calendar. Clearleft give their information as a v-card and h-card, so it's easy to grab it. Stikket understands a v-card and knows that it's an address and puts it in an address book. It does to dos, and notes.

Lots of really good ideas but need to test early and test often and testing with real people. Need them to test with - we are not necessarily our own best audience.