IOTA and the tangle

In https://steemit.com/technology/@kingscrown/iota-the-cryptocurrency-that-jumped-into-6th-position @kingscrown writes about IOTA which suddenly appeared in the top 10 of coinmarketcap.com recently, basically out of nowhere. In fact as I write it sits at 8th position with a market cap of over $1Bn

I'm not here to talk about the relative merits or otherwise of IOTA itself as an investment, but rather the interesting new concept it involves - IOTA you see is not a blockchain, but a "tangle", otherwise known as a directed acyclic graph.

In the tangle system, each new transaction does a little work, checking two previous transactions for validity and consistency - in that way, it "approves" the two previous transactions on which it's based. In the image above, the shaded nodes at the right hand end of the graph are the "tips", the newest, unapproved transactions, the next new transaction would choose two of those tips to check and approve.

In this way, the node that sends a transaction does a little work to check 2 previous transactions, so there's proof of work of a sort, but no mining in the traditional way. The incentive for nodes to remain honest and approve only consistent transactions is that if they don't, they in turn will have their transaction rejected by a future transaction that refers back to them through a line of transactions.

In an analogous way to a blockchain mining blocks on top of each other, a tangle transaction is more reliably confirmed the deeper in the tangle it gets buried - the more later transactions refer back to it, the more it has been checked and can be relied upon.

Another cool feature is that the tangle is asynchronous:

That means part of the network can branch off, transact on it's own, and rejoin the main tangle later. In this way, the tangle concept reminds me of the Internet as a whole - some routes may be broken some of the time, indeed it's probably that will be the case, but the network can work around failures and is tolerant of faults.

A key enabling technology for the IoT (internet of things) concept, with low-powered computers in all sorts of common objects such as appliances, vehicles, wearables and perhaps control systems for other networks and the like is the ability for node to pay each other for very cheap services such as a little bit of bandwidth, a small amount of processing or a service low in value, but that is of sufficient utility that a lot of nodes might each pay a little bit to use it for a short while. These nano-payments are not really possible with current technology, as too much work is required for a secure network - BTC transaction fees recently have been extremely high due to the network congestion, and IoT applications would only lead to much higher transaction volumes due to the tiny sizes of transactions involved.

The only cost paid by a node issuing a transaction in the tangle system is the small amount of work involved in validating the previous transactions, there's no transaction fees to pay, so tangles will be suitable for sending arbitrarily small payments. This is pretty cool.

I haven't formed an opinion yet about the sudden rise from nowhere of the IOTA market cap, and I don't know enough to know whether it's a good investment, I am frankly a bit suspicious of the sudden appearance in coinmarketcap.com when the bitcointalk thread has been around for a long time, but regardless of the economic merits of this particular implementation the concept as a whole is interesting and I'll be watching to see where it goes - if you feel like digging deep into the technical details you can find the IOTA whitepaper at https://iota.org/IOTA_Whitepaper.pdf