Talk for Article "Bitcoin boomed in 2017, where will it go next?"

Talk about this Article

  1. [ This comment is from a user you have muted ] (show)

    I am not a Bitcoin expert, but am pretty sure that the power consumption issue has been misrepresented–not only by the Guardian article cited, but by many others. I don’t know enough to propose a precise rewrite, but hope someone will come along and improve the article in this matter.

    The essential point is that the bitcoin software , as designed, intentionally modulates the difficulty (there is actual a formal parameter of the process named difficulty) of the process the miners execute in order the set the rate of new bitcoin generation to the pre-determined curve. If the miners buy, install, and operate a doubling of capacity worldwide, power consumption doubles, but the rate of new bitcoin award stays the same. Were the price to plummet, miners with operating costs above their new lower reward rate would turn off their machines, but the system would operate unchanged for transaction clearing. Clearly redundant computation not needed to support the actual transactions is the predominant power consumption term.

    Surely there is some minimum computational effort required to support the actual blockchain update, with sufficient redundancy to assure the integrity of the system, but I’m confident it is at a tiny (probably infinitesimal) fraction of the current level.

    So the “can’t scale because too expensive to support” claim is misdirected. It is not the computational requirements which create this situation, but, oddly, the extreme escalation in the market price of bitcoin.

    A rational miner, committing new capital, needs to assess that the value of bitcoins they will be awarded during the operating life will constitute a good return on the invested capital and operating costs (principally power). When the market price for bitcoin in this calculation is higher, it attracts more such investments.

    So it is the scheme by which miners are compensated, coupled with the extraordinary price rise, which drives the (yes, I agree) extremely excessive power consumption. It is not the technical characteristics of blockchain transaction support, nor even the specific version in the current implementation.

    1. [ This comment is from a user you have muted ] (show)

      Total supply cap is around 21 million. Miners after that will not be awarded newly minted coins for their efforts. The only reason to continue mining after that is to validate transactions…what is in it for them at that stage?

      As far as computational effort, it is a function of the block size and the burden of proof to validate the transactions which are capped at a certain number per block. So, at some point the mining becomes less rewarded and many will disappear from circulation. As it is, the majority of mining takes place across about 4 huge operations, perhaps undermining the decentralised goal of the currency itself.

      1. [ This comment is from a user you have muted ] (show)

        Actually the difficulty is adjusted once every 2016 blocks with the specific aim of maintaining a 10-minute block time. The adjustment goes both up and down. A little bit of searching can get you painful amounts of detail on the specifics and the history.

        A published table at the bitcoinwisdom site, which I suspect to be accurate, shows the upward adjustments as high as 21.39% and downward adjustments as strong as -6.09% in a table listing the values (and changes) over the last year and a half.

        In case I’m not being clear–the overwhelming majority of the computational effort involved in bitcoin mining is makework. Deliberate padding is automatically added to the base requirement to balance the computational resource currently deployed to this task to the desire rate of completion (and thus the currently desired rate of new coin production–which is on a scheduled rate decline toward the final cap).

    2. [ This comment is from a user you have muted ] (show)

      I think the assumption there is that miners are actually paying for the electricity they’re using to mine. Which, at least in China, they’re not. Lots of operations have been shut down after being found out (this is one link but obviously not the-be-all-end-all-reference http://www.scmp.com/news/china/society/article/2108486/four-accused-stealing-power-chinese-oil-company-generate-bitcoins)

  2. [ This comment is from a user you have muted ] (show)

    The article is good, but a bit superficial in the analysis by the experts. Bitcoin is the surface of the story – the brand name that gets the space recognised. To understand blockchain and other future currencies the article needs to consider the new breed of companies tackling real world problems with the use of decentralisation.

    1. [ This comment is from a user you have muted ] (show)

      Fair points Chris. It’s somewhat tricky to go deep with prediction articles. And yes, there is a larger ecosystem beyond Bitcoin, but the story is about how Bitcoin itself has risen in popularity and where it’s heading. The story is not about blockchain or cryptocurrencies.

      However, there’s a section about blockchain which explores the humanitarian work of the World Food Programme and how blockchain helps to tackle the refugee crisis in Jordan.

      Nonetheless, WikiTribune will be doing more stories beyond Bitcoin in the future. I’m happy to hear your suggestions on which companies are doing great work in this space. For instance, I’ve been hearing a lot about this one called IOTA, which is a cryptocurrency that doesn’t even run on the blockchain.

      1. [ This comment is from a user you have muted ] (show)

        IOTA is fascinating, Ripple (and their currency XRP) is perhaps more tangible for an audience and the comparison between the two would show the depth of the market and range of projects that are happening. From the machine to machine transactions of IOTA, to the tech that is assisting banks to facilitate cheaper payments with Ripple.

Subscribe to our newsletter

Be the first to collaborate on our developing articles

WikiTribune Open menu Close Search Like Back Next Open menu Close menu Play video RSS Feed Share on Facebook Share on Twitter Share on Reddit Follow us on Instagram Follow us on Youtube Connect with us on Linkedin Connect with us on Discord Email us