Is this a similar concept to compartmentalized consensus? https://mwhittaker.github.io/publications/compartmentalized_...

The compartmentalized consensus separates follower as acceptor and replica, and divorces command-log replication from data replication. It also seems to use two acceptors group as in horizontal scaling.

The compartmentalized consensus does not have relay nodes that do relay/aggregation. The idea in PigPaxos is simply that using randomized relay/aggregators have surprising power for vertically scaling Paxos.

The bipartisan Paxos, seems to apply the compartmentalized consensus idea to EPaxos.

Repo for the TodoMVC demo: https://github.com/swannodette/om-next-demo

Morse code instructional video from the US Army: https://www.youtube.com/watch?v=Li8Hiwbc664

The "Juliette" and "Romeo" suggested mnemonic imagery on 4:57 was hilarious and very revealing about other suggested "attitudes" of the time and context.

A series of CIA memos describes how Israeli Mossad agents posed as American spies to recruit members of the terrorist organization Jundallah to fight their covert war against Iran: http://www.foreignpolicy.com/articles/2012/01/13/false_flag

Concerning sampling and aliasing, you might find this video series from xipf interesting: http://www.xiph.org/video/

Except that the government has a monopoly on force. If a private company harms you, there are often civil and potentially criminal concequences. If the government uses your private data against you and violates your rights, who do you turn to?

> If a private company harms you, there are often civil and potentially criminal concequences.

It is really difficult to sue a company that has a team of lawyers while you are a individual just trying to get by, esp. if said private company did something that really disrupts your life, like wrongly foreclosing on your home or wrongly firing you.

1) It's unequivocally bad if the government spies on its citizens.

2) Some people manage to obtain justice when they are wronged. Often it's just not that easy.

Earthflight has similarly beautiful footage: http://www.youtube.com/watch?v=XIKnol2i850

The same is true of arithmetic coding, which separates the probability model from the encoding process. Feed an arithmetic coder a stream of random bits and it will efficiently sample from your model. See section 6.3: http://www.inference.phy.cam.ac.uk/mackay/itprnn/ps/105.124....

Here's a list of predictions: http://wiki.mises.org/wiki/Austrian_predictions

I would like to point out that the Austrian school teaches that economic predictions are impossible to make which is incredibly convenient when one of their theories breaks down completely (e.g. the Austrian Business Cycle Theory has been shown to be empirically false during all recessions we have data for).

Also, mises.org isn't exactly an objective third party.

Just to clarify, Austrians define monetary deflation as a contraction in the total supply of a given money. In this case, it would mean a decrease in the total number of Bitcoins.

Supply contraction won't happen in Bitcoin until sometime before 2140, when the number of coins lost to hard drive crashes or which are otherwise difficult to spend start to outrun new coins minted.