ChuckOne
Sr. Member
Offline
Activity: 364
Merit: 250
☕ NXT-4BTE-8Y4K-CDS2-6TB82
|
|
March 19, 2014, 09:46:11 PM |
|
[2014-03-19 21:43:05.325] DEBUG: Starting 8 background jobs [2014-03-19 21:43:05.364] DEBUG: Initialization took 15934 seconds [2014-03-19 21:43:05.375] Nxt server 0.8.11 started successfully.
15934 seconds? I do not think so.
|
|
|
|
jl777
Legendary
Offline
Activity: 1176
Merit: 1134
|
|
March 19, 2014, 09:47:58 PM |
|
Does anybody know how long the recent fork on the testnet was?
If it was less than 10 blocks, then using 10 blocks for confirming transactions if fine, but if it was more, then 10 blocks is not enough...
Assuming that it was significantly longer than 10 blocks, this is a potential issue, but since it is relatively rare case it would be a shame to have to wait for a full day before any transactions are fully confirmed. However, what if somebody deposits 100 BTC, it gets confirmed, they sell and cashout, but it all happened on a fork. Seems like a mini-disaster, at least to someone.
So, my idea is to use information from the nodeminers to detect how consistent the NXT network is. I am assuming there is a pretty good way to determine if most of the nodes are in sync. I want to use the most reliable data so I can essentially publish a probability that we are on a fork. I would also think that a specific node can query the combined data to see if it is in the majority or minority fork.
When we upgrade to a new version, it seems that even if you are in a minority fork if that fork is caused by the new version rejecting old blocks or old versions rejecting new blocks, then as more and more people upgrade to the new version, what used to be the minority fork is now the majority fork.
Does anybody see the flaw in the following algo:
A) find the latest version of NRS (probably safer to hardcode this to avoid Evil Bob's tampering) B) determine what fork the latest versions are on using method TBD C) warning nodes that are not on the expected majority fork that they are not on the majority fork D) Using above to generate overall NXT network fragmentation and adjust the number of confirmations needed based on this
Also, what is the best "method TBD" to determine that two nodes are on the same fork? Is it as simple as the blockhash of the latest block?
James
P.S. For nodecoin haters, do you still think nodecoin is bad for NXT?
|
|
|
|
pandaisftw
|
|
March 19, 2014, 09:48:09 PM |
|
Not sure about the tagging thing, I would just perhaps do some global categories only, like "currency", "commodities", etc... (and not allow the user to choose custom ones).
Hm, how would your client know that "fakecoin" or "fakeNXT" are currencies?
|
NXT: 13095091276527367030
|
|
|
wesleyh
|
|
March 19, 2014, 10:04:27 PM |
|
Not sure about the tagging thing, I would just perhaps do some global categories only, like "currency", "commodities", etc... (and not allow the user to choose custom ones).
Hm, how would your client know that "fakecoin" or "fakeNXT" are currencies? Asset issuers has to say it. If it's not true, we could use service providers or something. Service provider is 3rd party. Has an api that returns information about the assets, like: "scam", "ok", "not_checked", "incorrect_information", etc.. It's a centralized service, but there could be more than 1 service provider. (If anyone is interested in becoming a service provider that is). Didn't you have tags feature in your proposal? Issue is the same?
|
|
|
|
S3MKi
Legendary
Offline
Activity: 1540
Merit: 1016
|
|
March 19, 2014, 10:05:42 PM |
|
P.S. For nodecoin haters, do you still think nodecoin is bad for NXT?
Is nodecoin good for NXT?
|
|
|
|
jl777
Legendary
Offline
Activity: 1176
Merit: 1134
|
|
March 19, 2014, 10:13:32 PM |
|
P.S. For nodecoin haters, do you still think nodecoin is bad for NXT?
Is nodecoin good for NXT? If it helps detect when there are significant forks in progress and especially if your node is on a minority fork, do you think that is good or bad?
|
|
|
|
ChuckOne
Sr. Member
Offline
Activity: 364
Merit: 250
☕ NXT-4BTE-8Y4K-CDS2-6TB82
|
|
March 19, 2014, 10:17:52 PM |
|
Does anybody know how long the recent fork on the testnet was?
If it was less than 10 blocks, then using 10 blocks for confirming transactions if fine, but if it was more, then 10 blocks is not enough...
This conclusion is not correct. One does not draw such conclusion from 1 (in words ONE) single example. [...] Does anybody see the flaw in the following algo:
A) find the latest version of NRS (probably safer to hardcode this to avoid Evil Bob's tampering) B) determine what fork the latest versions are on using method TBD C) warning nodes that are not on the expected majority fork that they are not on the majority fork D) Using above to generate overall NXT network fragmentation and adjust the number of confirmations needed based on this
Yes. 1) People should be encouraged to write their own node code. NRS is just one of them. 2) A) does not work out. Tampering is still possible.
|
|
|
|
ChuckOne
Sr. Member
Offline
Activity: 364
Merit: 250
☕ NXT-4BTE-8Y4K-CDS2-6TB82
|
|
March 19, 2014, 10:19:30 PM |
|
If any of these node providers sees this: fix it! [2014-03-19 22:11:16.234] DEBUG: Invalid peer address: emoneyru2-nxtcoin.rhcloud.com:80, java.net.UnknownHostException: emoneyru2-nxtcoin.rhcloud.com [2014-03-19 22:11:16.862] DEBUG: Invalid peer address: emoneyru1-nxtcoin.rhcloud.com:80, java.net.UnknownHostException: emoneyru1-nxtcoin.rhcloud.com [2014-03-19 22:11:19.047] DEBUG: Invalid peer address: node7-mynxtcoin.rhcloud.com, java.net.UnknownHostException: node7-mynxtcoin.rhcloud.com [2014-03-19 22:11:19.056] DEBUG: Invalid peer address: node7-mynxtcoin.rhcloud.com:80, java.net.UnknownHostException: node7-mynxtcoin.rhcloud.com [2014-03-19 22:11:19.845] DEBUG: Invalid peer address: globogestion.cloudapp.net, java.net.UnknownHostException: globogestion.cloudapp.net [2014-03-19 22:11:20.460] DEBUG: Invalid peer address: 03-emoneyru.rhcloud.com:80, java.net.UnknownHostException: 03-emoneyru.rhcloud.com [2014-03-19 22:11:21.939] DEBUG: Invalid peer address: node8-mynxtcoin.rhcloud.com:80, java.net.UnknownHostException: node8-mynxtcoin.rhcloud.com [2014-03-19 22:11:22.282] DEBUG: Invalid peer address: emoneyru3-nxtcoin.rhcloud.com:80, java.net.UnknownHostException: emoneyru3-nxtcoin.rhcloud.com [2014-03-19 22:11:23.751] DEBUG: Invalid peer address: 02-nxtcloud.rhcloud.com:80, java.net.UnknownHostException: 02-nxtcloud.rhcloud.com [2014-03-19 22:11:24.077] DEBUG: Invalid peer address: 01-nxtcloud.rhcloud.com:80, java.net.UnknownHostException: 01-nxtcloud.rhcloud.com [2014-03-19 22:11:25.595] DEBUG: Invalid peer address: 03-nxtcloud.rhcloud.com:80, java.net.UnknownHostException: 03-nxtcloud.rhcloud.com [2014-03-19 22:11:25.946] DEBUG: Invalid peer address: 52fc12535973caf866000329@node5-mynxtcoin.rhcloud.com, java.net.UnknownHostException: node5-mynxtcoin.rhcloud.com [2014-03-19 22:11:27.863] DEBUG: Invalid peer address: node6-mynxtcoin.rhcloud.com:80, java.net.UnknownHostException: node6-mynxtcoin.rhcloud.com [2014-03-19 22:11:28.404] DEBUG: Invalid peer address: GornorellaVaginalis, java.net.UnknownHostException: GornorellaVaginalis
|
|
|
|
Jerical13
|
|
March 19, 2014, 10:22:03 PM |
|
Should there be some (general) categories for assets? If so, which?
You could index them according to type of asset: cryptocurrencies, fiat currencies, precious metals, stocks, bonds, loan markers, services, agricultural assets, real estate/land holdings, natural resources, energy resources, tech assets, miscellaneous assets. It would also be cool if you could index the sellers location and length of time that the sellers account has been established. I can imagine in the future that categorizing the AE will be a very smart move. I would suggest using an indexing similar to Craigslist (if you are familiar with it). Craigslist indexes a lot of postings from a lot of geographic locations and includes a large number of general category descriptions in a way that is very easy to navigate.
|
|
|
|
ChuckOne
Sr. Member
Offline
Activity: 364
Merit: 250
☕ NXT-4BTE-8Y4K-CDS2-6TB82
|
|
March 19, 2014, 10:25:01 PM |
|
nodecoins are earned each block based on how much you node is helping the network, especially if you are forging. They are sent automatically after you have earned 10 nodecoins Send to whom?
|
|
|
|
ChuckOne
Sr. Member
Offline
Activity: 364
Merit: 250
☕ NXT-4BTE-8Y4K-CDS2-6TB82
|
|
March 19, 2014, 10:25:53 PM |
|
Should there be some (general) categories for assets? If so, which?
You could index them according to type of asset: cryptocurrencies, fiat currencies, precious metals, stocks, bonds, loan markers, services, agricultural assets, real estate/land holdings, natural resources, energy resources, tech assets, miscellaneous assets. It would also be cool if you could index the sellers location and length of time that the sellers account has been established. I can imagine in the future that categorizing the AE will be a very smart move. I would suggest using an indexing similar to Craigslist (if you are familiar with it). Craigslist indexes a lot of postings from a lot of geographic locations and includes a large number of general category descriptions in a way that is very easy to navigate. It should be tagging system. It is more flexible.
|
|
|
|
Jerical13
|
|
March 19, 2014, 10:29:10 PM |
|
Not sure about the tagging thing, I would just perhaps do some global categories only, like "currency", "commodities", etc... (and not allow the user to choose custom ones).
Hm, how would your client know that "fakecoin" or "fakeNXT" are currencies? Asset issuers has to say it. If it's not true, we could use service providers or something. Service provider is 3rd party. Has an api that returns information about the assets, like: "scam", "ok", "not_checked", "incorrect_information", etc.. It's a centralized service, but there could be more than 1 service provider. (If anyone is interested in becoming a service provider that is). Didn't you have tags feature in your proposal? Issue is the same? That is a good idea, and technically it wouldn't be centralized, it would be a perfect example of "Free Market". It would seem centralized if there weren't a lot of competition in what is essentially a virgin market though, but if it is a profitable enterprise, I would guess competition in the market would be eventual. Think of all the free market opportunities like this that the AE will create!!!!
|
|
|
|
|
L5Society
Newbie
Offline
Activity: 56
Merit: 0
|
|
March 19, 2014, 10:33:56 PM |
|
Hi Brian. Thanks for posting your concerns in detail. I found it easier to digest than trying to follow the back-and-forth nature of discussion here. I agree that the fee thing needs to be sorted out with an eye towards future price appreciation. Unfortunately, I don't think there is a simple solution. Charging fees based on bandwidth usage or size of transaction seemed like an elegant solution when I first heard about it. However, as nxt intends to compete with a variety of financial systems, each with complicated fee structures of their own, I'm concerned that some great nxt services will be "priced out" of the market, while other services will be unnecessarily cheap. When thinking of nxt as bitcoin in the future, I think the analogy breaks down when talking about fees. Specifically, the future of bitcoin assumes much lower price volatility. Therefore, denominating fees in bitcoin instead of fiat isn't a big deal. For us, we expect high volatility for the near term. Denominating fees in terms of fiat (or gold or any other asset with lower volatility) might make sense in the near term. With regards to charging fees for placing an order in the asset exchange, I think that there is room for the nxt AE to improve upon the way markets currently work. If you follow nanexllc on twitter (his work has been featured in the Wall Street Journal, amongst other publications), you will learn that high frequency trading firms have been abusing the stock markets in the United States. They take profit, and justify this profit by claiming to increase liquidity and lower the bid-ask spread. However, in reality, the displayed liquidity (order book) can rarely be accessed by real trades, and orders often disappear before a counterparty can fill them. I know that setting walls is a strategy for day trading. However, for a wall to be effective, it has to be large. This implies that the fee would be tiny in comparison to the size of the wall. Additionally, Nanex's recent studies show that the bid-ask spread has not tightened over the life of the current regulation (Reg NMS). Based on these issues in current stock markets, I am suspicious of HFT terrorizing our infant network, and I think that charging fees for placing an order prevents both spam and these predatory trading practices. We aren't designing the AE for the profit of algorithmic day traders.
|
|
|
|
ChuckOne
Sr. Member
Offline
Activity: 364
Merit: 250
☕ NXT-4BTE-8Y4K-CDS2-6TB82
|
|
March 19, 2014, 10:36:09 PM |
|
|
|
|
|
Jerical13
|
|
March 19, 2014, 10:38:21 PM |
|
Should there be some (general) categories for assets? If so, which?
You could index them according to type of asset: cryptocurrencies, fiat currencies, precious metals, stocks, bonds, loan markers, services, agricultural assets, real estate/land holdings, natural resources, energy resources, tech assets, miscellaneous assets. It would also be cool if you could index the sellers location and length of time that the sellers account has been established. I can imagine in the future that categorizing the AE will be a very smart move. I would suggest using an indexing similar to Craigslist (if you are familiar with it). Craigslist indexes a lot of postings from a lot of geographic locations and includes a large number of general category descriptions in a way that is very easy to navigate. It should be tagging system. It is more flexible. I am not sure I understand what you mean by tagging. To me It would seem be more practical to implement if there were a set of "check the one that applies" categories that would actively limit the types of listings that could be used. Flexibility could be achieved by having a comprehensive set of categories to choose from, that could be added to if needed; and by allowing more detailed descriptions of the asset in the description field during the assets registration.
|
|
|
|
okaynow
|
|
March 19, 2014, 10:42:42 PM |
|
Hey guys, just a heads up, to any of you trading at Cryptostocks:
AlcurEX, a listing for an exchange traded there, has posted a list of coins to be voted by the shareholders and NXT is on the list. If any of you are holding any AlcurEX shares, please vote for NXT.
Thank you -Fo-
|
1PeecNu1J8VNKpgR13nasMZWLcMZrwNJfc
|
|
|
xyzzyx
Sr. Member
Offline
Activity: 490
Merit: 250
I don't really come from outer space.
|
|
March 19, 2014, 10:43:01 PM |
|
How?
Nxt transaction sigs only care about the account id's (not say something like the "genesis block id") so a "clone" (such a test net) in which account #'s that *match* main net (most likely due to using the same password) means that you can *steal* NXT by just broadcasting a test net tx on main net. (this issue will also apply to "parallel chains" if and when implemented) This is why I suggested adding a unique identifier/GUID to a NXT style blockchain. A transaction would have to specify which blockchain it belongs to.
|
"An awful lot of code is being written ... in languages that aren't very good by people who don't know what they're doing." -- Barbara Liskov
|
|
|
xyzzyx
Sr. Member
Offline
Activity: 490
Merit: 250
I don't really come from outer space.
|
|
March 19, 2014, 10:45:33 PM Last edit: March 20, 2014, 07:39:37 AM by xyzzyx |
|
Result: U sold 50k NXT for 0.01 BTC. Crazy! ...thnx brother... Hey, just don't send NAS when he sends u 0.01 BTC. Call me State-The-Obvious-Stan if you want but.... I just want to check You would need 50,000NXT in your Nxt account for this to go through, right? But I suppose they could look in your Nxt account, see you have 1500Nxt and say "hey, I'll buy 1500NAS for 0.01 BTC", right? Move the NXT from your account to a new account before sending anything on the other chain. If you duplicate an account across blockchains, consider them compromised and act accordingly.
|
"An awful lot of code is being written ... in languages that aren't very good by people who don't know what they're doing." -- Barbara Liskov
|
|
|
mcjavar
|
|
March 19, 2014, 10:45:53 PM |
|
Hey guys, just a heads up, to any of you trading at Cryptostocks:
AlcurEX, a listing for an exchange traded there, has posted a list of coins to be voted by the shareholders and NXT is on the list. If any of you are holding any AlcurEX shares, please vote for NXT.
Thank you -Fo-
I did, although I don´t think that they will be able to add it in the beginning.
|
|
|
|
|