I am not talking about average fees per block. I am talking about how much that would be in fiat in the future, we don't know that. We don't know how much fees per block will be in the future. All this 'analysis' based on past events is not relevant. It could well justify renting a VPS, labor is not a lot, 30 mins per month(?), NXT is already quite stable. Especially funny if someone projects fees based on their testnet, as in the case of BitShares.
So no analysis can be done using past figures. Everything needs to be done using the unknown future? Is that what you are saying?
Are you a dev at NXT by any chance ?
I once tried running a NXT node. This was before the AE release. The thing ended up taking up 5 gigs of memory after a couple of weeks. I'm not sure what was going on, but it was nuts.
30 minutes = $20 for an average developer or admin. So that makes it $25 a month now. Just take the average fees in a month and see what sort of stake is required to make $25 a month. Do you know what the average fees charged per month is ? Oh wait, we can't use facts. My bad.
Is it true or not true, that if I wanted to do an expensive operation, I could rewrite the wallet to put my transaction on a block that I am forging ? A selfish-miner of sorts and thereby taking away fees from others ?
People try to do analysis using past figures all the time, you're free to do it. Only future will show if an analysis had anything to do with reality. Often it doesn't.
No, I am not a dev at NXT.
NXT could not be considered stable before AE release, it's much more stable now, and will get more stable in 1.3.xx when some code will be reworked. It's still beta quality, just like most software it should get better with time.
30 mins is to upgrade NXT 2-3 times a month as new releases (bugfixing or new features) come out. Of course, in the future, new releases will not come out that often, it's still in beta stage, so you should understand it's all beta-testing going on now, that's why have to spend time on it, just like beta testers work on BitShares. The difference is BitShares delays releasing to live net, is that good or not, opinions can differ. In any case, the main point of running a node is to forge on it and support the network. You gain not from forging mostly, but from being invested in NXT and assets on the exchange and price appreciation, so if someone's looking to profit from forging, they should re-think their strategy.
Here is a link to see distribution of forgers:
https://nxtblocks.info/#section/blockexplorer_chartsWhen you check the blockchain, the 'Others' that make 41% constitute more than 300 accounts, so there are 300+ accounts in NXT network that successfully produce blocks. We'll have to see how this changes in the future.
A block with the highest cumulative difficulty wins, not sure what you mean by 'rewrite the wallet'. For more info and specific questions you should better visit nxtforum.org and ask experts.
Here is a whitepaper on NXT:
https://dl.dropboxusercontent.com/u/2948110/Nxt/NxtWhitepaper_v122_rev4.pdfsomeone's still working on it, so it's far from complete.