A quick ‘end of the day’ post just to gather some thoughts together…

It’s interesting that the response to FOWA is the same for me this year as it was last. I’m glad I’m here, but it’s a reaffirming of what (I think) we all know already. I haven’t come away full of new ideas and insights (which is possibly what you’d hope from a conference titled ‘Future of Web Apps’), but I have been reassured that my thinking and knowledge is up to date and aligned with the leaders in the field.

The recurring messages (and these were much the same as last year):

- iterate often
- focus on single points of pain
- know your audience
- be honest and transparent when things go wrong
- don’t seek perfection

The only final thing to say before I go do some more geek mingling is that most of these people are REALLY TERRIBLE at giving presentations. I know geeks don’t traditionally have a huge range of social skills, but still. ..these are basic things: don’t rely on a network connection, know your audience (ironic, that), make sure you talk to time, don’t mumble…It’s presentation skills level 1…

Enough moaning. More tomorrow.

Matthew Haughey from Metafilter was next up. All in all not a terribly inspiring speech from a man who is obviously happier in front of some code than presenting. Nothing hugely new here – three messages which have been given by others earlier in the day:

- “be a third place”: people have work, home and the place they go to hang out. If you can be that place, as Facebook is to many, then you’re on to a winner
- “eat your own dogfood”: live and breathe your app; respond to a pinch point in your life or work and build something YOU would use
- “allow unintended uses – build out based on the edges”: watch people using your app or community in different ways and capitalise on this.

Next was Heidi Pollock who gave some interesting insights into mobile app development. In no particular order:

- africa as growth market: ppl using mobile to access the web because they don’t have a PC, not because choose to
- screen size standard 176px, and stick to a 10k download limit. Some phones down to 128px. Motorola V3 is 30chrs x 8 lines…
- more than 3000 phones types, often with browsers unique to each type

Interesting that there are no H or LI tags in xhtml mobile 1, and that B tags are used instead of STRONG (B is shorter, so less bytes!). Ultimately this all means that the Semantic Web is left behind on mobile devices…

I love Etsy. That’s why I’m in a session now with founder Robert Kalin, talking about ‘the future of commerce’.

“Etsy is a marketplace for individuals, but really also a soapbox for people to show off their products.”

Etsy has sold around 1.4 million items. Kalin sees this as a small start for a much bigger future.

“The job of the design is to get the hell out of the way and let the pictures talk”

Core to success is the human to human element of the site, things that have meaning, and not some huge corporate beast.

Shop by colour as innovative experience – used by first timers much more than regular users.

Social shopping: a platform for anything not mass-produced.

Commerce is often below the radar as far as web2 is concerned.

Welcome back. It’s the biorhythmical (?) trough that follows lunch when eyes start closing and web developers slump over their Macs…

I’m in a session with Daniel Burka, creative director from Digg. He’s going to be talking about user feedback and how best to respond to it.

Differences between Pownce and Digg: Digg already has a user base who have ‘invested themselves’ in the site. They have the paradigms of use already engrained. ‘younger’ (by time, not by age) users are sometimes more resilient to change.

How does user experience affect what you do?

First ask whether the change is worth it. Know who your community is by participating yourself. Run focus groups on your new features. Finally, and importantly, define measures of success.

Types of feedback….positive, bug reports, negative feedback.

Important to look at types of negative feedback. Spot patterns – if you get negative feedback as soon as you change something then that’s probably ok. If you’re still getting the same feedback weeks later then you probably need to act.

Don’t react to everything that is asked for, obviously…instead look for patterns of response and react to these.

Look out for expert feedback and make the most of the (free) advice…

Spend a lot of time gathering implicit feedback – watch people using your site. The vast majority on the site will be silent – try to gather data from them too.

So what do youi do first? Nothing. Don’t react, wade in. Instead, sit on the feedback for a few days and identify themes. Then engage with the community – ring them up, email them. Finally, iterate any development.

Went to a session about Google Gears but left as soon as I heard the words “stack overflow”. Instead I met Dan Z from Box UK and we chewed the ecud for a bit.

Next up was Robin Christopherson talking about ‘Attractive yet usable’ sites. It’s always interesting hearing from someone like Robin who is blind himself. Although it’s interesting seeing a blind user on good and bad websites, there wasn’t a whole lot new here, especially if you’ve spent a bunch of time working with accessibility issues already.

Briefly, he covered:

- Benefit to all users of having clear and accessible sites. Demonstrated the use of audio descriptions.

- Disney and GM sites which don’t have proportional fonts or consistent navigation. Vodafone new website – lots of effort for accessibility but still problems with raising font sizes where text overlaps or becomes distorted.

- Drag and drop etc is likely to become more prominent as apps become more advanced. Need to ensure that these sites are device independent so that they can be used by people with hand-eye coordination problems.

The simplest way to test is to try unplugging the mouse and then see if you can navigate.

Derek Powazek, author of my favourite book on communities online and Heather Champ from Flickr are next…

Tips for building and maintaining communities.

Go beyond the legal requirements. Example was for ‘make lemonade’ downtime notice on Flickr. Embraced the downtime as a positive thing and ran a competition to win pro accounts.

Confess! when you fuck up, admit you’ve fucked up. Flickr blog post example – ‘sometimes we suck’.

Don’t keep score. Use scores and leaderboards when you are playing some kind of game. Don’t when you’re not – being ranked in a community scenario can really put ppl off.

Make Real Stuff…producing real world artefacts is a major driver for virtual communities. Print, tshirts, mugs, etc.

Rip that Band Aid – painful changes are better done quickly. Explain them, take feedback, but mostly just do it asap.

Community, Manage Thyself – let users manage their own content. It’s less hard work for you and more satisfying for your community.

Communicate Expectations – ppl don’t read terms of service. Provide them for a legal perspective but also put them into real words. Example: “dont be creepy. You know the guy. Don’t be that guy.”

Don’t Create Supervillains – when you boot ppl from communities you will create an enemies. Instead find other ways of dealing with problem members. Example – one site (remaining nameless) makes the website slower and slower for problem members…

Know Your Audience. Self explanatory, really…

Embrace The Chaos. If you have communities then unexpected things will happen. Let them be imaginative and creative. Be as flexible as you possibly can. Sometimes when people misuse your site it could be that they have a need or an idea which could be embraced rather then reacted to. Example is Friendster > Dogster / Catster…

This is all going to be a little bit stream of consciousness but I’ll clean it up later…

Om malik and Ryan Carson are talking about the future of web apps. Michael Arrington is stuck somewhere. Oop, no – he’s just arrived on stage…

TechCrunch 40 is getting a lot of interest and the general question being asked is ‘what works’.

Mike A talks about the models for charging. Can you offer apps for free and somehow still make money? The other big question is about music, especially given current backlash against drm.

Amie Street is a good model where price is determined by popularity. More expensive is essentially ‘better’. It’s a disruptive model.

What is bad? Attacking a space which is already saturated. Business models should be as disruptive or as different as possible.

Have a single offering which solves some pain. Don’t do too much.

Fwb application in Facebook – Om talking about this being one of the only original good facebook apps.

Google in the social graph. Have they been beaten by facebook.

Mike A – facebook is not the end of the social network story.

Om – don’t put all you eggs into the facebook basket. Portable social apps are the road ahead. Facebook pushing in right direction.

Om – Enterprise widgets, or Enterprise in general is the place to focus on.

Mike A – mobile email, but mobile in general. Also vr in gaming or where you use body to control computer.

—-

10 web apps in 10 minutes – adobe.

Sliderocket – ppt online
Scrapblog.com
Piknik image editing
Mtv Video remixer
Buzzword word processor
Finetune web radio station, built using AIR
Ebay as desktop app
Adobe media player built in AIR
Pownce air application
Google analytics on desktop