This is an interesting article on how the drop in Bitcoin (BTC) versus US dollar rate may mean that a 51% attack on the bitcoin network is getting easier. According to the article 90% of mining capacity has gone offline, as it is no longer profitable at the current BTC price. It argues that if you buy just part of that now worthless (because single purpose) equipment cheap, you can effectively double the mining capacity of the network in a way that gives you more than 51% of the capacity (at least temporarily). Then that entity would be able to influence the ledger.

Of course the big Asian elephant in the room that is left unmentioned is that such a 51% attack is likely to have taken place already. As the article itself states ‘most miners’ were in China, where you can now get all that mining equipment cheap ‘by the pound’. As most mining was already running on a handful of Chinese superclusters, and given what we know about the data driven authoritarian model China geopolitically pursues, the conclusion is rather obvious: The 51% threshold had been reached in China already. So it’s not an emerging 51% attack risk, it’s just that there now may be a window of opportunity for somebody else to do it.

I still wonder in what instances blockchain is actually really useful, meaning that having a distributed database/ledger AND a transparant log of transactions AND a permanent immutable record are needed for such a use. But where that is the case, I am convinced it is, other than maybe for public records, not needed and even risky (see above) to have it run on a global network or platform. As then others, not invested in your actual use case, may have influence on the validity and stability of your use case.

It makes much more sense to me to have use case specific blockchains where the needed computing nodes are distributed across the network of people invested in that specific use case. For instance I can easily imagine a local currency or exchange trading system (LETS), to use blockchain. But only if it is run by the members of that LETS for the members of that LETS. Meaning a small computing node attached to your home router as part of your membership contribution.

Technology needs to be ‘smaller’ than us, run and controlled by the socially cohesive group that uses it for some specific purpose, otherwise it more likely undermines agency than provides agency.


Cryptocurrency Art Gallery, public domain image by Namecoin

I am interested in blockchain as a distributed way of organizing things through software. I have questions that center around in which situations that distributedness, having a public ledger, and having a permanent ledger is actually useful. Also in general for any defined user group, available blockchains are all global by nature. This takes away any agency that group has concerning ensuring the availability and soundness of the technology they use. This is a threat to a group’s resilience basically (e.g. when a group in northern Poland runs their transactions on something that is dominated by opaque Chinese computing clusters). So I am interested in how to deploy blockchain for a specific group (that can then run their own nodes for the needed calculations.) The potential to subvert a blockchain in such a situation is theoretically bigger, but at the same time it is also more strongly embedded in existing social relationships which provides its own robustness.

Here’s a number of links concerning blockchain I came across and read the past few days:

Explaining blockchain

  • A good read, ‘A Letter to Jamie Dimon‘, which takes as perspective that the distributedness is less effective than centralized solutions but also the key aspect for the intended user groups, as this is the only way to avail themselves of specific affordances. Distributedness is a tool to increase resistance to censorship (also read as ‘access’), and blockchain allows creating fully distributed applications.
  • A talk by Richard Bartlett at Re:Publica Dublin, on whether a blockchain is decentralizing power or not.

The ICO hype is unfolding
ICO, initial coin offerings are campaigns to sell tokens for your specific blockchain application. You can buy them usually only with Bitcoin or Ethereum. What amazes me is how much money (millions) are getting invested in short times (the term vaporware comes to mind), and that minimum investments are often in the 5.000 or 10.000 Euro range.

Examples