- Rank aggregation
- Argument map aggregation
- Budgeting / collective modeling of risk/costs/utility, etc.
- VCS (Git as an MVar)
- Distributed CI (build, test)
- Contextual chat
- Support line (for people using code, to developers actively maintaining that code; also: for those developers back to their userbase (on demand A/B testing) )
- Issue crowd-funding
- Subjective deprecation (method can be marked as deprecated; is automatically excluded from compilation for clients that don't use it; is automatically removed once no clients use it)
- Password store
- Document store
- Identity store
- Public chat
- Wikipedia
- Images
- Music (streaming)
- Movies (streaming)
- Private message exchange (there are many legitimate uses for such a service, but I am still divided between to what extent it should be easy to have completely anonymous, perfectly secret discussions. By no means do I mean to buy into the terrorism/pedophile arguments for blanket government surveillance, but there's risks on the other side too. Balance is key in this one. The proper design of such a service could very well be one of the first fundamental problems we could attempt to tackle as a society, using social tools developed in Demograph. As a sidenote, I think that reducing cypher strength to the extent that a modern supercomputer can crack payloads might be a way to provide a pragmatic balance; supercomputer resources will be put there were deemed necessary, protecting the majority from blanket surveillance as well as from the terrors. It's unclear how long it would take for such texts to become practically public, due to the developing speed of the average computer, aiming for after death might be good thing?). Do note that any services following here that do exchange of data in secret could be abused for non-legitimate purposes too if not carefully designed.
- Secrets exchange - passwords / certificates to be shared between teams
- Scalable anonymity - We want to allow a person to manage their degree of anonymity when making a statement, without introducing single points of failure through trusted parties. Context: we want to allow an inhabitant of country X to reliably declare that (s)he is an inhabitant of country X, that (s)he is opposed to the current regime and will aid in overthrowing it if X% of the population is against the regime too. Here it is important that we can reliably establish that unique identities have added their support, without being able to identify the exact source of the statement. One way of attacking this problem is:
- Identity decomposer: Splits a given identity into its bits and signs the value and location of each bit. The result of this step is then n signed bit-values, where n is the length of the personal identifier. In order to thwart attempts to discover initiatives to obtain anonymity, we have each client periodically query an arbitrary peer to obtain a decomposed identity.
-