I can't tell you exactly but what I can tell you one solution is somewhere less than one cent per transaction.
Lots of the small ones are more expensive than this - depends on how well you invest.

Sometimes, less is more.
With NXT being able to be used for more than payments, we need to make sure we are not pricing the usage too high. We could very well end up with higher forging per block with a lower minimum fee!

Basically, 99% taxation appears to get a much higher tax revenue than 10% taxation. however, the higher rate kills any economy while the lower rate could easily generate more than 10 times the revenues.

James

+1

Maybe being cutoff from Internet is not so bad after all.

Came up with blockchain FIFO and NXT plugins (both hardcoded and external) and I am getting close to understanding how AE can work with automatic gateways, in fact, that was that specific problem that was generalized into NXT plugins.

I can't see jean-luc agreeing to add Java implementation of bitcoind (does it exist?) into NXT core. If we can hardcode the bitcoind into the NXTcore, then we get verifiable bitcoind ability added to what NXT can do. I know, horrible security issues could arise. However, now that Dr. Evil/Jesse James is here, I am hopeful that we can create an architecture that Evil Bob can't take advantage of.

There are all these trustless escrow algos, zeroknowledge, etc. that say they can do some pretty powerful things. I just wish I was smart enough to figure out how to combine them to create a decentralized gateway for other cryptos. For NXT, I can see that we can send NXT to a special address and the network can credit us with the appropriate NXTcash (or other goodie), then on redemption, the special address converts to NXT. The problem with other crypto is that while everyone can publicly verify that address XYZ deposited BTC to gateway account, how to do a cryptographically safe withdrawal from a single acct stumps me...

One klunky way would be to have N accounts of 0.1 BTC each and limit withdrawals to that resolution. Then using the multisig secret sharing approach I am pretty sure would allow cashing out of the right number of 0.1 BTC accts to fulfill the withdrawal request. The problem is that 10 BTC withdrawal would get hit with .01 BTC fees, but maybe its not so bad?

I am pushing for several different tech initiatives:

NXTcash (zerocoin/zerocash added to NXT)
NXT layers (VM, language + tool set, DAC components)
NXT plugins (solve trusting external code using RT hash of plugin code space?)
NXT automated gateway DAC
NXT crosschain trading of other cryptos

I have called for help in defining the layers, but so far CfB is the only one that posted about VM layer. The above list is my attempt at making the abstract "we need NXT layers defined" into more concrete definitions. I want to anchor super useful functionality (NXTcash, crosschain, gateway DAC) and make sure whatever low level VM we end up with and all layers in between actually helps implement the reference higher level projects.

A simple proof of concept NXTplugin that I hope jean-luc will build into NXTcore would be a NXTsmtp. I think it is pretty simple to generate email from AM contents with some sort of flag to trigger actually sending out the email. If email is encrypted so issuing accts public key can decrypt the contents, then private emails can be sent using decentralized network. I think that is pretty cool even if NXT isn't the first method to be able to do that. Confirmation that email has been sent can be added to blockchain with hash of email content/recipient. Of course, anybody can decrypt the content using public key, but at least the email is not sent in plaintext over Internet.

I am 100% sure I have messed up on details in the above, please use this as starting point to get actual solution. I am sure I have not dealt with any error handling, but using TF we can know what node will attempt to send email, so we know when to check for email sending errors and all nodes can therefore have deterministic way of processing all emails.

If anybody has any really cool high level functionality, just post it. Dont worry if you have no idea how it can be done, I usually dont at first either. All we need is one person to come up with a way to implement part of the cool idea. Then others follow and before you know it, it will end up on the NXT projects list with funding, devs and projected release dates.

James

Sorry, but I am concentrating on things that haven't been done yet, especially the things that are supposed to be impossible, but are in fact just really difficult.

If you run an exchange, there is ~10BTC bounty for adding XCP:

https://forums.counterparty.co/index.php/topic,61.0.html
Description: The Counterparty Team is now hosting a bounty for the creation of a centralized exchange, supporting XCP and BTC, or adding XCP and BTC support to an existing centralized exchange. The bounty funds will be distributed based on a consensus of PhantomPhreak, xnova, and myself. The Counterparty Team will start the bounty funds with a donation of 4.5 BTC and 1500 XCP to the donation address. The bounty will be paid on February 25.

That's pretty good money for adding a single new crypto. I hope all the guys with centralized exchanges for NXT will take the day or two needed to add XCP support.

James

James, have you read up on CoinJoin, CoinWitness or Bitprivacy?

I'd post links but here at work on my phone it's tough.

Check out those three and their respective threads, there might be something relevant there (iirc all are short 2 page threads).

Still looking for someone to do this with.

PM me if interested!

Guys,

I'd like to know if NXTio.org is trustable and if I can keep my NXTcoins there. Do you recommend it?

Thanks,

Maybe I'm too paranoid these days, but this nxt-e IPO seems suspicious.
Could U check them out a bit more for us, Allwelder, someone said the guys behind it were Chinese, AFAIK.