Category: Keynotes

Bill Scott describes Yahoo as a tribal platform, particuarly with the acquisition of various API-based services such as Flickr and del.icio.us as part of the Yahoo Developer Network.

Attention needs to be driven by engaging and relevant interactions to create the loyalty neccessary for successful services. The successful interaction patterns Scott outlines include:

• Immediacy
• Directness – drag’n’drop, inline editing.
• Invitational – hover and tooltip invitations to further interaction,
• Removing boundaries – endless scrolling, in-context expands, hover details.
• Light footprints – Remembered collections, rating items.
• Cinematic – slides and fades, self-healing transitions.
• Rich content – Sharable, bloggable, clonable and findable items.

Alongside the Yahoo Developer Network and UI Library,  the Design Pattern Library is a set of thirteen patterns that attempts to surface a vocabulary and expose solutions for this paradigm of UI components.

Joel Spolsky attempts to deconstruct the winning characteristics of of what he terms Blue Chip products (iPods, Julia Roberts etc.). Spolsky hypothesises a ‘Formula’ for good products and services – making people happy, creating an emotional connection and obsessing over aesthetics

• Reddit.com – the cartoon mascots and karma points creates an emotional bond with the user.
• Motorola’s PEBL – this phone’s surface is highly ‘fondle-able’, like a real pebble 🙂

Spolsky cheekily highlights three trends in web design over 2005 – ugly ugly and ugly! Berating designers for seeking to emulate Google with the pastel Arial aesthetics – essentially copying the wrong part of Google’s design. Google’s design only makes sense as an aesthetic statement when contrasted with older cluttered portal layouts. He’s not impressed 🙂

Ask and you shall receive (something)…Plum’s Hans Peter Brondmo outlined his views on collecting almost anything, sharing it and then connecting it to others in order to discover. Brondmo sees this as a solution to search results that can overwhelm the user with choice.

Plum appears to be a web-based clipboard, where clips can extend to data and people – ostensibly this doesn’t appear to do anything that del.icio.us, Scuttle or our very own Plasma and Klippr do not already do.

Plum does include some interesting features such as

• integration with iTunes. Skype
• Views – collections can be rendered as blogs and other layout formats.
• Storing actual fragments of pages, rather than just URLs or thumbnails.

Brondmo went on to announce the launch of a developer API (written in the last three days!)  and invited the audience to embrace the APIs and extend Plum into new application areas.

Brian Dear of the Events And Venues Database (EVDB) is on a mission to maximise event discovery. Announcing an event is in essence a flyer or an attempt to focus and gain attention.

Initally EVDB was considering either building a portal or platform. Eventually, EVDB decided to build both:

• Eventful.com – let users share information about known events and scanning the future for events that matter to you. Features include search, calendars, tags, RSS & iCal feeds, groups, friends/family/contacts, auto-submission, an iTunes import, an AIM EventfulBot and ‘performers’ data
• The platform – centered around a RESTful API, EVDB is eminently mashable, the most notable example being PodBop.

Interestingly, EVDB is looking to demand aggregation – tools for users to express desire for events they wish would happen. Eventful demand is more than a simple wishlist, but includes the promise of fulfilment. The company doesn’t seek to sell intention data to record labels or promoters (for example) but create a marketplace between fans and performers.

Incidentally, Dear claims that Eventful.com is the largest utilisation of microformats in a live service…this is borne out somewhat by the fact that Ray Ozzie used Eventful.com as a source site in his demonstration of LiveClipboard yesterday morning. Finally, Dear demonstrated some of the viral event campaign tools – creating stickers for use on blogs and other third-party sites.

Adobe/Macromedia’s Kevin Lynch began with a recap of the AJAX rich internet application paradigm and some of its limitations (sockets, cookie storage, vector graphics etc.) before leading into the announcement of a Flash+AJAX web application model – only recently adopted by the AFLAX and Dojo.storage frameworks.

Interestingly, Flash is the most widely distributed client in the world, more so than even browsers, and is constantly refreshed and updated to ensure each client is running the latest code. Flash, Flex and the Flex+AJAX bridge are now the core of Adobe’s RIA strategy.

Lynch demonstrated an application that utilised a Ruby-based AJAX form to populate a Flash-based carousel of Flickr images. Notably, embedded Flash movies now include a View Source command to support remixing and reuse of Flash applets.

Lynch notes that though server-side development has shifted to an SOA and ESB-based models, but client development remains fragmented; a Service Oriented Client could be a solution to this problem and actually make client code more portable and consistent in use of local computing resources.

Clay Shirky introduced a pattern language for moderation strategies, using Slashdot as a case study:

• Freedom’s just another word – with increased freedom to publish, the level of annoyingness increases and mechenisms are required to mitigate increased noise.
• Slashdot’s defenses – members defend readers from writers by ranking comments on individual posts…around 20% of comments are therefore hidden (those below a certain threshold).
• Tragedy of the Commons – Move comments to a separate page, treat readers & writers differently, let users rate posts and include defensive defaults.
• Who Will Guard The Guardians – Treat users and members differently, measure good behaviours, enlist comitted members – judges can’t post.

Shirky went on to describe Bronze Beta, where there aren’t ‘features’ as such, comments are central to the content base of the site and logging in is an optional interaction, removing a barrier for user contributions. Shirky’s ITP students at NYU have made available a wiki and mailing list to foster collaboration and discourse on patterns for moderation.

Jon Udell‘s session on Attention Focussing Strategies was framed around four key concepts:

• Heads, decks and leads – drawn from print publishing metaphors, giving people the abilty to scan information at multiple levels of detail.
• Active context – resources that are self-updating and queryable, rather than a static list of information (wishlists, feeds, linklists, blogrolls, podcasts).
• Canonical names – URLs, ISBN and other namespaces are a powerful source for focussing attention.
• Multimedia storytelling – examples like locative mashups, the iPod-by-Microsoft YouTube clip and others illustrate the potential for attention grabbing content, being created at the grassroots.

I read about Jeff Han’s work on multi-touch screens just a few days ago, but seeing the technology demonstrated live indicates that a transformative paradigm in UI experiences may be unfolding.

The multi-touch screen is essentially a touch screen that responds to simultaneous touches at multiple points on its surface, This enables users to form more intuitive, expressive and sophisticated gestures as well as providing an inherently multi-user interface, as different parts of a screen can be used by each user. The touch surface itself is based on the total internal reflection properties of the glass above the screen itself.

Han indicates that the social, intuitive nature of the multi-touch screen, takes computing away from the historic WIMP direction of interfaces and opens computing culture to children and older users. Certainly, the playful, unintimidating nature of the technology appears to lower the usual barriers of complexity that those unfamiliar with computing environments face.

Han’s demonstrations included:

• Navigation of a 3D globe using gestures for panning and zooming.
• TV channels displayed as individual scalable ’tiles’ and windows.
• A scalable, translucent, virtual keyboard.
• A sock puppet, animaed through gesture, created by a drawing simple strokes on the screen.
• Puzzle games with users using different parts of the screen to solve their puzzles in a time-trial.

Han’s technology seems eminently useful for highly interactive, manipulable visualisations of complex data, but it’s not difficult to imagine versions of Photoshop, fabrication control interfaces, 3D design and sculpting applications being crafted from the same technology.

Ray Ozzie, founder of Groove Networks and creator of Lotus Notes, now one of Microsoft’s many CTOs demonstrated LiveClipboard.

Coincidentally, Ian and I were just last night speculating on a network enabled clipboard that would allow users to copy and paste from other screens to their own. It seems Ozzie’s work on Windows Live has yielded a similar, yet more sophisticated application.

The clipboard is the original mashup mechanism, enabling data from one application to be re-used in an another – Ozzie began his talk with the premise ‘Where is the clipboard of the web’ and ‘Why isn’t this just the regular clipboard users are already familiar with?’. Ozzie speculates that such a mechanism would become a lightweight mechanism for connecting the data silos of the web.

LiveClipboard uses the native CF_Text format (the original microformat!) of the Windows clipboard and simply inserts a piece of structured XML data, rather than plain text. This isn’t disimilar to the way content can be copied from an IE web page and pasted into Microsoft Office applications, preserving the original pages’ formatting. In this case the content of the CF_Text construct is HTML.

Ozzie’s examples of LiveClipboard included:

• Copying event information from and Eventful.com web page and pasting into an Outlook Calendar.
• Copying RSS feed data from a web page and pasting into an aggregator – simplifying an otherwise onerous process for new users.
• Copying a live location feed from a Windows Mobile / MSN Spaces field to a Facebook.com profile.
• Copying a Flickr image from a web page and pasting into a desktop folder.
• Copying and pasting a Flickr photostream feed into a folder populates the folder with live Flickr content. Is this a pre-cursor to WinFS functionality?

Ozzie closed out his session with an invitation to help proliferate LiveClipboard support by helping to shape a common set of microformats, conceding that the success of LiveClipboard depended on collaboration between Microsoft and the wider web.

Introduced of Cory Doctorow, science-fiction author Bruce Sterling, spoke on the themes of ubiquitous computation. Now based in Belgrade, Sterling continues to blog and write fiction as he observes the rehabilitation of a failed state. The Internet of Things, predicts Sterling, will take around thirty years (as did barcodes) to emerge fully…also predicting that at age eighty-six, he’ll be needing an Internet of Things…

Currently, the field is still defining its vocabulary and terminology as experimenters begin to tinker with the technology. By ‘freezing’ the terminology of artificial intelligence and thinking machines, Sterling contends that AI’s potential was funneled into narrow directions. Google appears to be intelligent, but is not presented as a thinking machine like Microsoft Bob and Ask Jeeves…a linking, ranking and sorting machine, not a classical ‘AI’ – Turing vs Google! Artificial Turings and Feynmans may not be able to efficiently compute the cheapest local plumber…perhaps talking about sorting machines rather than AI would have produced a Google in the 1980s…

The socially generated knowledge of the Internet is not about intelligance, but changing our relationship with the physical world through laptops, phones, PDAs and cameras. Objects can be labelled with electronic tags that can be sorted, ranked, located and connected…objects that are auto-Googled?

Sterling goes on to describe spimes, fabjects and blogjects as the artefacts of an Internet of Things – our posessions are no longer inventoried in our heads, but by machines around us. Google my shoes! Trackbacks to my car! Interestingly, at ETel, Peter Cochrane spoke ot logistics and things as a large global economic growth area; Sterling seems to underline the potential here and reflect the timescales noted by Cochrane.

After introducing ThingLink as a generic term to encompass identifiers for objects, Sterling contrasts spimes with Blogjects, with the latter providing a near-term realisation of the potential of Things. Evocative Knowledge Objects, Ubiquitous Findable Objects,

BTW, I wrote about Sterling’s fabjects a couple years ago as part of a rif on fabbing and P2P….the napsterisation of things? Things 2.0…pick your meme 🙂