Quote from: mthcl on February 11, 2014, 12:47:55 PM

Quote from: ChuckOne on February 11, 2014, 07:56:36 AM

Quote from: joefox on February 11, 2014, 05:29:55 AM

- has to have a client
   - this costs nothing

- risks all his stake if the client is malicious
   - this can be prevented with some forethought... which costs nothing

- risks his hardware if the client is not secured against malicious attacks
   - risks his hardware? This is quite far-fetched and highly unlikely. It reminds me of the StuxNet virus, which made Iranian uranium-enrichment centrifuges spin until they burned themselves out. But I know of very few "malicious attacks" that can fry my home PC.

- has to pay the electricity bill
   - this is a sunk cost. The forger already has an electricity bill, and as has been stated countless times, the cost of electricity for running a Nxt node is very low.

- has to pay for the hardware/renewal of such
   - this is generally also a sunk cost, since people can forge on existing hardware... except for folks who have bought a Raspberry Pi: they're out $50. Anyone who has spent $5000 on a spiffy Nxt forging rig should be a little embarrassed.

- has to pay for a place that hardware can exist in
   - This is also a sunk cost. You already have a place to exist in, which you generally share with your hardware. And your hardware isn't that big.

- has to pay for the bandwidth he contributes to the network
   - This is also a sunk cost, since you already pay for bandwidth. And you can still forge if you don't advertise your node address, which minimizes bandwidth usage.

- has to implement security measures (attacks, heat etc.)
   - "heat security"? I don't even know what that means. As for other security: most people already have an edge firewall protecting their local network. People who don't use any form of security for their home setup are far more exposed already, for their existing hardware (PCs, etc.). So we're at "sunk cost" again.

As your reply shows, you have no idea how precious time and other resources are and how computing works nowadays. Nothing of my previous points is far-fetched or completely impossible.

Furthermore, you just provide many relative arguments: of course forging is cheaper than mining. So what? Cheaper means not: no time, no hardware, no place etc. It still costs something. Compensation is therefore legit.

Agree with ChuckOne. Look also at the example of bittorrent: the torrent trackers had to use rating systems and such, just because otherwise there would be too many leechers for too few seeders. That's not because it costs anything (at least in free countries like Brasil, where there is no risk that you could get prosecuted for that) just to leave the client open, it's because people just don't care about it. There are so many things to care about in this life, so it's better to offer at least some reward if we want that people invest some attention to the network maintenance.

I see what you're both getting at. I think we differ on what constitutes value.

The foundation of my post is that forging is valueless only when there are enough services built on top of Nxt to render it valueless. Given the current Nxt topology, I agree that forging is necessary just to maintain the network. However, once nodes are running code and providing services for other coins, applications, contracts, games, parallel blockchains, etc., those nodes will profit from those endeavours AND also look after base Nxt network consensus, as a pure consequence. I define "value" as what is offered by those services.

I don't think "block rewards" are going away. Someone's account has to forge a block, and that account will still win the fees in that block. Those fees will just be LOW. So perhaps this whole debate is too much in the weeds.

I feel like I'm flip-flopping. In the early days of Nxt I said that the value of forging was in building network consensus, not profiting in terms of Nxt. Or... maybe I'm not flip-flopping at all.