Thoughful post on the difference between symmetric (Facebook) vs. asymmetric (Twitter) relationships, courtesy of Wes @ Plinky. Advantage to Twitter for enabling attention inequality (receiving more attention than you give). Bottom line for me is that edges in the social graph should have (dynamic) meta-data as opposed to being hard-coded to “friendships” or “following” or whatever. Of course, this will be hidden from users in simple abstractions such as connecting/friending/following.
Subscribe
About Simeon Simeonov
Events
Get invited to events I am organizing or speaking at.
Top Rated
- The best vesting schedule
- Monitoring Redis with MONITOR and WireShark
- Apache Spark native functions
- Startup Founder Agreements
- JSON and JSONlines from the command line
- About
- Startup Anti-Patterns
- Metcalfe's Law: more misunderstood than wrong?
- Making Money From the Paradox of Choice
- Nokia & Navteq: Vertical Integration or Diversification?
-
Latest Content
- Apache Spark native functions
- Unicorn pressures and startup failures
- Advertising marketplace design
- Angel investing strategies
- JSON and JSONlines from the command line
- My most favorite math proof ever
- Monitoring Redis with MONITOR and WireShark
- Google and the ecosystem test
- Anatomy of an online ad
- Startup anti-pattern: platform risk
@simeons on Twitter
- New England doctors, hospitals launch voter registration efforts bit.ly/2PI6Kmh https://t.co/qF5THxK3uJ 2 days ago
- In the years since we open sourced Spark-records this has continually proven to be true. #spark #DataEngineering twitter.com/kebabmaihaddi/… 3 days ago
- @kebabmaihaddi Job flights are an organization and/or application specific correlation identifier to help with oper… twitter.com/i/web/status/1… 3 days ago
- @kebabmaihaddi I pulled the 98% idempotent Spark job auto recovery rate from Swoop’s own petabyte scale systems. #spark 3 days ago
- Somewhere along the way, we forgot we are all in this together... Partisan Pandemic: How Partisanship [...] Affect… twitter.com/i/web/status/1… 1 week ago
- ack09
- Adobe
- Advertising
- AJAX
- amazon web services
- angel investing
- Apple
- Automattic
- AWS
- B2C
- Better Advertising
- Blogging
- bootstrapping
- Broadband
- Bubble 2.0
- business
- cloud computing
- customer development
- Digital Media
- e-commerce
- entrepreneurship
- FastIgnite
- Flex
- founder agreements
- FutureOfWebApps-SF06
- Geotagging
- IE7
- Industry News
- iPhone
- lean startup
- Life
- Long Tail
- Macromedia Flash
- marketing
- Metcalfe's Law
- Microsoft
- Microsoft Office 2007
- MIT 100K
- Mobile
- MySpace
- Nantucket Conference
- network effect
- Office 2007
- OpenOffice
- opensource
- Plinky
- Polaris Venture Partners
- privacy
- Ruby
- Ruby/Rails
- SaaS
- Security
- Shopximity
- Social Advertising
- Social Commerce
- Social computing
- social media
- social networking
- startup anti-patterns
- startups
- Swoop
- technology
- TechStars
- The Long Tail
- Thing Labs
- VC
- Venture Capital
- virtual worlds
- Web 2.0
- Weblogs
- Web Services
- Windows Vista
- WordPress
Twitter is pretty much in the spotlight nowadays, but isn’t the only asymmetric platform, neither the first. From personal experience, Flickr comes to mind as an example, also the humble Ecademy (nowhere near in membership size, but worth studying for its 10 years of consistent innovation and very advanced social functionality, populated by a small but very active constellation of communities)..
The asymmetric model better reflects real social relationships: parellel co-existence of (to use classical relational terminology) both one-to-one and one-to-many links from the same node.
An interesting aspect is the concept of ‘attention relationships’ vs ‘human relationships’. While ‘human’ relationships are the fundamental (and semantically superior) fabric of social networks, attention has become an extremely important attribute of human relationships. A scarce comodity in the IO (information overload) world, attention merits modelling as an entity / object in its own right – especially because its better understanding is expected to make it more manageable and yield economic (or political) benefits.
Before Twitter, ‘plain old’ blogging was the area where dynamic metadata was (nearly) mastered in assessing the attention ‘value chain’ and propagation. Volumes of such data were much smaller, as most models were based on links, talkbacks and quotes.There are also countless dispersed blogging platforms (and non-platforms :). Twitter, with the concept of ‘following’ brings an unprecedented volume of such data, concentrated on a single platform – perhaps, as Joshua Porter suggests, the key to its monetization?
But he is,IMHO, over-obsessed with Facebook’s copycat actions and the (perceived) war between the two. If they have anything to fight over, it is just the center-stage position in the hype circus, not much else.
Vlad, I think your point about attention meriting its own modeling is spot on. We are all short on time and how we spend our minutes says a lot.