Quote from: MalReynolds on October 02, 2014, 01:20:31 PM

Quote from: GreXX on October 02, 2014, 01:44:41 AM

The new simplified solution still bases forging on nodes being active, but it is a more random solution as it doesn't require sending of the extra overhead bits and isn't an http request so consensus isn't a problem in the timeframe we are aiming for. This solution works and will get us back online, but it still isn't ultimately our goal or the system we wish to implement long term. It is reminiscent of an early build we had before implementing the up-time calculations we had in the launch version....
<...>

Remember the goal was to reward users for running nodes in an equitable manner. There are definitely other solutions and we will be looking into them.

Is the goal of rewarding users in an equitable manner for running nodes the right goal for Crypti to have?

This leads directly into a critical philosophical discussion and a practical solution for Crypti that I was going to wait about bringing up until things were more stable, but maybe it's time to have that discussion now.

Start from first principles. In any cryptocoin system, SOMEBODY must be selected to encrypt the current block and add it to the blockchain.

PoW Bitcoin first motivated these "miners" by giving them a reward of newly created coins. This created three problems.

First, the new coins created every ten minutes is a huge source of monetary inflation for the Bitcoin economy (despite Bitcoin's false propaganda to the contrary).

Second, the miners are motivated to start a huge ASCI arms race that cannot be sustained and is headed very soon to becoming a classic case of ecological exponential growth / collapse (see http://blockchain.info/charts/hash-rate?timespan=all&showDataPoints=false&daysAverageString=7&show_header=true&scale=0&address=) . Like the death of the dinosaurs, this will soon kill Bitcoin stone cold dead and open the landscape for a new 2.0 cryptocurrency to become king.

And finally, the Bitcoin approach is incredibly wasteful. During every Bitcoin 10 minute block period, literally thousands of millions of billions of candidate blocks are generated for possible inclusion on the blockchain. It takes tens of millions of dollars of gear to do this burning huge amounts of electricity. Every single one of these block is mathematically correct as a block chain add-on. All (except eventually for one) are rejected because they don't have enough "leading zeros" in the first part of their bit string to "win" the (illegal in all US States) DICE THROWING GAMBLING GAME being played by the miners to select who gets the new coins. Bitcoin is no different than shooting dice in the back alley except they use computers for the dice and the internet for the back alley. BESIDES BEING ILLEGAL, THIS IS A HUGE UNNECESSARY WASTE OF COMPUTER POWER.

PoS NXT solves the last two problems by cooperatively designates in advance one single person to generate the block every blocktime instead of competitively selecting in realtime the winner of a huge worldwide gambling contest every ten minutes as Bitcoin does. Instead of the warehouse full of $2000 ASIC supercomputers the Bitcoin miner needs to win his lottery to forge a Bitcoin block, all a NXT forger needs is a single $35 Raspberry Pi (or better still, something just a little more powerful but still incredibly cheap like an ODROID-U3 : http://hardkernel.com/main/products/prdt_info.php). This HUGE advantage alone means NXT will ultimately win over Bitcoin.

But NXT still feels the need to "reward" people for running nodes. Since no new coins are being created like Bitcoin does, NXT "recycles" to node operators / "forgers" a small proportion a coins gathered as "fees" during a blocktime. Well, that's OK, but to win the NXT lottery that single Raspberry Pi or ODROID-U3 has to be loaded up with as much NXT as possible (the "stake" in PoS) to maximize the chances of winning the NXT lottery. This leads to a mentality that NXT needs to be hoarded by individuals instead of dispersed among many to form a merchant-based sales economy.

Crypti wants to become the foundation for merchant-based sales. So the devs have come up with "PoT/PoI/PoP" to replace NXT's PoS and so break out of the PoS inevitable coin-hoarding mentality. This is indeed the next big step and why I am interested in seeing Crypti succeed.

But in addition to breaking out of NXT's PoS coin-hoarding weakness, Crypti needs to break out of the real flaw that both Bitcoin and NXT share : the perceived need to reward node operators.

Rewarding node-operators just isn't necessary AT ALL. Computers have become so cheap and so ubiquitous that it really isn't necessary to motivate people to operate them anymore. There will be no stopping the cryptocoin that realizes and implements this fact.

Think about it. In the real-world retail environment, the concept of "blocktime" is "ringing up a sale". Are reward fees used to finance the operation of a cash register? Is a cashier paid a "tip" for pushing the buttons and swiping a card / opening a cash drawer and making change? NO! Both the cash register and the clerk are so cheap they are both considered "overhead" in the operation of the sales environment and paid as an expense in the overall operation.

With cheap sub-$100 computers like ODROID-U3 capable of making up ALL nodes in a Crypti network, the idea of coin-owners being needed to operate the nodes AT ALL, much less needing to be "REWARDED" for doing so, is ridiculous. If Crypti is really being used by merchants as true money, the MERCHANT POOL running Crypti processing nodes will provide all the nodes required to secure the network, and can adjust their prices so running a cheap node is just another incidental business expense.

So to repeat, there really isn't a need to have Crypti coin holders running nodes at all, much less be "rewarded" for doing so. You've got 750 BTC worth between a quarter and half million dollars. Go buy a thousand ODROID-U3s for $100,000 and let the devs set up a huge 1000-node Crypti network on those at relaunch. As merchants sign up, power down an ODROID and mail it to them for FREE to start back up in their store. Or let them download the code to put in an SD card and stick in a ODROID for a configuration-controlled addition to the network. If Crypti is a successful cryptocoin of commerce, paying $100 for a little plastic brick to sit in a corner so they can join the fun is no big deal for ANY commercial vendor of ANY size. They don't need to be rewarded every so often for such a small buy-in requirement, especially since the reward is a mere pittance anyway.

Remember, the key is not rewarding a node operator - it is cooperatively designating one to add the next block.

The real key is to get merchants to want that little plastic ODROID brick / Crypti node. No mystery remote server setup for them, no bulky windows computer running 24/7 for them - just pull it out of an envelope, plug it into a wall socket, and access it thru a cell phone app / wifi to setup and processs sales transactions. That should be the pitch.

Crypti alone isn't very enticing to lure a merchant - it's just one of many cryptocoins and has a had rocky start to boot. So the brick needs to offer something more than just Crypti to generate immediate interest. Now, if its code could handle / automate BITCOIN operations NOW for a vendor as well as CRYPTI operations as those ramp up in the FUTURE, I think there would be some real merchant interest in getting one. Offer merchants something that can help them handle Bitcoin AND Crypti. We gain market penetration based on the Bitcoin capabilities. If and when Bitcoin falters, we have already arranged a seamless merchant transition to Crypti. Merchants come for Bitcoin, they stay for Crypti.

Don't lose sight of what your primary goal for Crypti is: to become a non-hoarded coin for merchant use in sales transactions.

Is working to perfect a PoT algorithm to reward coin-holders the most direct, most efficient, and most likely to succeed effort that will achieve that goal?

Or is concentrating on creating a dual Bitcoin transaction access / Crypti block creating cheap node that merchants will want for themselves NOW a better path?

Maybe this is not an either - or situation. Maybe...do both?

My parting thought: if you are doing a major system reorganization anyway, take a step back and see how you can more tightly integrate a cheap, simple standardized building block like the ODROID-U3 into your revamped effort and so create something that will better advance Crypti with an additional hook like Bitcoin access to MAKE MERCHANTS WANT IT NOW.

Making merchants want it NOW: that's the path forward for Crypti.

Now that is one radical approach, in my honest opinion I don't think it would be feasible to use this kind of concept, also it would make Crypti initially heavily centralized.