Next Generation of Web Applications
[Live Blogged Notes from UX Week 2006]
[A panel on cool stuff led by Jeff Veen]
(The panel participants aren't listed anywhere so I can't keep track of who is saying what but I'll do my best to take notes. Right now they are just introducing themselves. Jared Spool is talking about what UIE does and some other people are going to demo real live stuff.)
Stamen Design Work
- mappr associates a massive placename geotag database with placename tagging from flickr.
- cabspotting tracks yellow caps by GPS on a map in realtime data.
- digg swarm that represents stories with yellow blobs to watch the digg storm in effect.
So how do they get this work (and they did the awesome MoveOn conversation visualization for election '04 conference calls)? They decide what they are interested and draw attention to themselves by doing free research projects.
MindCanvas by Uzuntu
- Demos (Jeff declares "oooh")
No marketing business model, completely word of mouth.
So why flash? Needed sound and some things that Ajax couldn't do yet
Back to Stamen: Team is three people, occasionally four. Often work only on one thing at a time because atomizing work with many projects is too difficult. Working on one thing at once leads to a unified office atmosphere.
Q: So are your desks pushed together?
Definitely. Converted one of the walls into a whiteboard wall.
From post-its and whiteboard drawings they go into data land. For Digg, for example, they had to design an API before building the API apps on top.
Back to Uzuntu.
Very mobile communication. No meetings, just Basecamp and e-mail to have formal communication. It's been two years and it works great for them. At most they have five people on a project. One person plays the conceptualization role but most drawing is done collaboratively. They also use Powerpoint for prototypes because it is the best way to collaborate. It let's them go back and forth between anyone on the staff.
To another guy (I wish I caught his name)
They also used PP for an animated comp for Harvard Business School. This was the only way to do it but realized it was stupid to do in PP. It was a way to get an idea in front of a client.
The back button is a real concern. Do users of Kayak drag a slider and expect the back button to work?
To Spool.
The back button feels vistigial because the browser was never meant to be an interactive window. Now we just have to wait for users who like the back button to die off.
To Uzuntu.
To date nobody has expressed a problem with using the browser for a flash heavy app.
Q: How do you involve users in development?
Veen: For MeasureMap they did a ton of research up front. They realized that people only use a few data points in huge stats packages. They did a little ethnography and then started building working prototypes right there in Rails.
Guy I don't know: They make a prototype just rich enough to do testing.
Veen: The team is very comfortable with throwing away their work. They know that it is iterative and are willing to get rid of ninety percent of their work.
Stamen: Not much user research but definitely iterate to get it right.
Spool: Really loves paper prototypes and then makes them clickable. All that work of doing this is work that doesn't return value. If you would just use paper and have people point with their finger instead of clicking real things. It gets the idea across without risking hearing "I love it, ship it today!"
You don't need fancy laboratories with one-way mirrors. You just need to see and watch somebody.
Q: (odd question with enormous scope about everything in the universe)
Veen: So you mean, where do we start?
Guy (I wish I knew his name): Know your audience. Use tools and research and try to really understand people and what they need. Getting out and talking to people is always where they start.
Q: How do we drive technology.... (What?) Web is neat but pointless? (Did she really say Web 3.0? Me confused)
Veen: Struggles with that himself. Google is very engineering centric. If something interesting comes up. A prototype comes up with explicit techniques and then an implicit UX and then later asks for a designer. Look at Google Base, it's powerful but not user centered.
Stamen: They get that criticism all the time. But not everything needs to be intresting or useful if they push in a good direction. If you are thinking about usability you shouldn't look to the future, to web 3.0, you should look to the past. Wait for things to iterate before focussing on usability.
Uzuntu: They are a small company but they have found engineers who really seem to empathize with users. Really respects engineers who can see ahead and make that happen. Why don't innovations come from us? We must imagine what the next thing is going to be and work with engineers to make what people want.
Merholz: There is time for two more questions and this is the first. One thing that was alluded to in the keynote is a shift away from page design and container design, a space filled with stuff. How do you design when you are not designing for laying things out on the page, for designing things that people are going to change.
Stamen: Don't like to talk much about a project for an architect. Created a space for lot's of photos to go to. Built it on flickr. Expected a dozen photos per project but they loaded it with content and the thing creaked under its weight.
Uzuntu: Focussed on intention. When you are thinking about moving from space to space the emphasis is about showing what's changed and what you can do with it.
Guy: Start with what you know to be true, with the 80% of what users will put into these containers. This is the type of content you expect for the structure so start there.
Q: Here's a back button idea. If the user can't undo what they did with what they just used to make a change the user will go to the back button.
Veen: One more thing... This can continue at a party Jared Spool is organizing. It will be at Heritage India at six pm.
0 TrackBacks
Listed below are links to blogs that reference this entry: Next Generation of Web Applications.
TrackBack URL for this entry: http://www.samfelder.com/mt/mt-tb.cgi/226
2 Comments
Leave a comment
Who is this guy?
Hi, I'm the Guy. :-)
"The guy I don't know" is Steve Mulder from Molecular. Uzunto is Rashmi Sinha from Uzanto.