Quote from: DumbFruit on April 04, 2017, 04:28:33 PM

I wish they'd stop referring to DAG as "decentralized", especially when years later there is no cryptocurrency in sight that relies on DAG that is also significantly decentralized, for exactly the reasons I explained... More than a year and a half ago.

Quote from: DumbFruit on September 24, 2015, 02:55:00 AM

I'd like to point out that in DagCoin, it is still only miners that vote. If you're a user that applies work on a transaction and broadcasts it, then you're a miner.
Work is a waste from the perspective of the miners. There is no incentive, or even a method to incentivize miners to do work on transactions in DagCoin. Competition would atrophy hashpower on the network in order to drive up transaction speeds, which is the product that the miner provides, users enjoy, and the only method of remuneration. Mining is just a cost that the miner wants to minimize.
This is just like if we removed the block size limit from Bitcoin and removed the subsidy.

That problem, the lack of funding for difficulty increases in order to slow down transaction speeds, isn't referenced in the white paper unless I missed it. So ultimately to get back to answering your original question "Why 1 sec, not 1 min?" the answer is I don't see any decentralized throttling mechanism that could actually work in DagCoin.

http://forum.iota.org/t/discussion-removing-peer-discovery/939/2

IOTA requires that you manually assign peers because running on it's own it implodes from the bandwidth, because of course it does. It removed Bitcoin's throttling mechanism and now they want to figure out how to prevent these out-of-control blocks in this "blockless" blockchain. Sometimes the real world really is stranger than fiction.

Have you looked at https://byteball.org what is your opinion on that?

It chose to be decentralized but not trust-less, and let the witnesses take the fees.

By the way, IOTA as claimed is for "Internet of Things", and they use Proof-of-Work.

Let me remind you, as someone who actually works on micro controllers at a big enterprise on IoT producst:

Most IoT chips are supposed to be run on batteries, on 3.3V down to 6mAh. Yes, that is right, the latest BLE chips actually use a full 6mAh when radio is on full power and CPU on hard work mode.

Can you imagine how much "Proof-of-Work" such a device can accomplish? Yes, NONE, nada, zilch. It struggles to maintain an TLS connection. DTLS is what people use as https is too heavy. And why? Because you cant ask your customers go around replacing your batteries every year.

And these guys come in here with their IOTAcoin and claim, the stupid devices will somehow magically perform a Proof-of-Working / wasting electricity, in order to transact?! What the fuck.

So in short, iotacoin is not for IoT. Proof-of-Work does not mix with devices which try to do no work!

The only good coin for IoT is Byteball - the cost of sending a transaction/receiving/subscribing to an address - is similar to cost of doing TLS.