Challenges

Guiding Questions
- What are the trends and directions research is taking?
- What are the unsolved research challenges?
- What is the one thing you feel doesn't work well in your favorite OSN?
Develop a distributed graph processing & storage stack
- Store the (social network) graph such that graph locality is maintained when partitioning
- Support the execution of, both, batch aggregate graph algorithms and online low latency local/personalized queries
- Support many-hop graph traversal queries in real time when needed, rather than requiring that they are computed as batch jobs periodically
- Develop concise graph query language(s) that make it possible to quickly prototype new algorithms (e.g. recommendation algorithms)
Apply event processing/matching to social network event streams
Identify the (open source) technologies that are used in popular social networks
- Which (types of) technologies do all social networks have in common?
  - RQ: What taxonomies of technologies for realizing social networks exist ?
  - RQ: What open-source technologies are used by social network market players (commons paterns) ? Why and how ?
  - RQ: Does open-source software speed up the setup of social network infrasctuturre ?
- Do companies contribute back to open source communities?
  - RQ: How is open-source community benefiting from the social networking hype ?
  - RQ: Are social networking players contributing up-stream with open-source projects ? are they forking open-source projects ?
Search results and recommendation results have an impact on user behavior
- How do such systems impact the way a social networks grows/evolves
Methods of identifying inappropriate content
- Is it possible in decentralized social networks? Is this the cost of privacy?
- How do you crowd source when content is private?
- Different "laws" in different sub-graphs
- How do you decide what is appropriate behavior
Recommendation in decentralized environments
Using social network data to predict/forecast real world events
Anonymization techniques for data sets that you release
Comparing topologies and usage patterns of different (types of) social networks