dConstruct Microformats Workshop data portability

Tantek Çelik and Jeremy Keith

Archives questions

• How old are your email archives?
• How many times have you moved them?
• What format do you keep them in?
• What app do you use to browse them?
• ...to search them?
• What about photos?
• ... or any other kind of data you create?

More archives questions

• What are the oldest data archives you have?
• What disk formats?
• ...or what bus/cable/connection format?
  • E.g. SCSI?
• What file formats?

Moving your data

• What data do you move when upgrading computers?
• What transfers and what doesn't?
• What do you leave behind?

Recovering your data

• What data do you move when your computer breaks?
• Or when one of your web services breaks?
  • Desktop.com and many others...
• What do you lose forever?

More recovering your data

• What do you recover when your hard disk crashes?
• Ever recovered from fragments of a file?
• What's recoverable and what isn't?

Why do we care?

• Archiving, moving, recovering data
• Microformats are NOT trying to solve all these
• Why is this important?
• Might help with designing open data formats

Obligatory Web 2.0 themes

• "You control your own data"
  • Users should own their own data
• Mix & match data & services
  • Requires open data formats & protocols
  • Users are free to move their data
  • away from "broken" interfaces/programs/sites
  • ...into new and better ones
• Why would any company support that?

Companies & open formats

Why companies are supporting open formats:

• Some actually "do the right thing"
• Trust
• More and better tools/source/tests
• Easy importing
• Network effects
• Outgrow the competition

Open data formats

• What do open formats enable?
• Rapid evolution of user interfaces
  • ...independent from the data
  • Both web-based and local
• Key requirements
  • Highly interoperable
  • "web friendly"
  • Truly open

data portability: definition

• is about you, the user,
• being able to move and use
• your data - whether you created it or purchased it
• across space - different websites, devices, media
• across time - archiving at one point in time, retrieving at another
• helped drive microformats
• see: microformats.org/wiki/data-portability

social network portability: problem

• Why does every social network make you:
  • re-enter all your personal profile info?
  • re-add all your friends?
• In addition, why do you have to:
  • re-turn off notifications?
  • re-specify privacy preferences?
  • re-block negative people?
• AKA "social network fatigue problem".

social network portability: goals

• New social networks should let you subscribe to:
  • your profile information
  • your social network
• from any existing profile of yours.
• Would be nice if preferences for:
  • notifications
  • privacy
• also transferred between profiles.
• AKA a "social internetwork", a network of social networks.

social network portability: how-to

• Publish microformats in user profiles
  • implement hCard on user profile pages
  • XFN rel="me" on hyperlinks to users' blogs/home pages
  • hCard+XFN friends lists
  • XFN rel="me" to link to separate friends list pages
  • XFN rel="me next", rel="me prev" for friends pagination links
• Subscribe to microformats for user profiles
• when signing up a new user, let the user:
  • subscribe to their existing hcard-supporting-profiles
  • subscribe to their existing hCard+XFN supporting friends lists
• see: microformats.org/wiki/social-network-portability