Author

Topic: NXT :: descendant of Bitcoin - Updated Information - page 410. (Read 2761632 times)

hero member
Activity: 644
Merit: 500
Is this 1626 words dictionary enough now or should we go with a bigger one, also to allow less words?

Why?

If crypto.getRandomValues is working properly then 1626 is just perfect. If crypto.getRandomValues  isn't working then even 20,000 words won't be enough, as real entropy will be limited by the random number generator.

Some people are having difficulty comprehending that 128-bit is a very big number.

Lets assume the worst case.

Lets say in two years, 1 billion people are using wesleh'sy client and  everyone created their pass phrase using these randomly generated words.  

Now lets assume that every bitcoin miner in the world  (millions of dollars of electricity per day)  all work together to find just one of these 1 billion randomly generated pass phrase. The current hash rate on bitcoin network is 3 million GH/s.

How long would it take for the entire bitcoin network combined to find just one of 1 billion pass phrase?

Total number (aprox 128-bit)  1626^12 = 3.4×10^38

1 billion users, 3.4×10^38 / 1000000000 = 3.4 x 10^29

Now the entire bitcoin network combined works to find just one pass phrase   3.4 x 10^29 / 300000000000000000  =  1133333333333 seconds

that's 36,000 years (36 thousand years).

lets assume they get very lucky and find one pass phrase only after 1% of search, that's still  3 thousand years

And after all that work (and electricity) and 3 thousand years later,  they might discover that the account they did find only had 2 Nxt  in it.

The point is that  the problem (if there is) will be with random number generator -- not with number of words, which are just fine.

That's why I keep saying you don't need 50 or 100 character password  (not even 35, actually  -- 25 is fine). . It doesn't add any more security. If there is a security hole (like for example keyloggers) even 1 million character password isn't going to help.


You're right, bitcoinpaul was only suggesting that, from the end user point of view, a 10 words passphrase is a little more convenient than a 12 words one. This is possible with a larger dictionary.

That will require about 7200 words. That's 4.4 times more words. wesleyh will have to find a lot more words that are still simple every day words.  



hero member
Activity: 910
Merit: 1000
Any suggestions about Parallel Chains? If not then I'll stick to BCNext's draft.

A question:

can parallel chains be 1min blocks and 1s, 5s, or 10s out of sync so every 1s, 5s, or 10s one chain is forging?

this might be an interesting answer to the latency problem of 1min
and provide more opportunities for clients to get quick confirmation of transactions...

Interesting thought.
hero member
Activity: 910
Merit: 1000
I've completed the 1st draft of my article about forging from a probabilistic point of view: http://www.docdroid.net/9yub/forging.pdf.html

Comments, suggestions, (and donations Wink) are welcome!

Please, discuss it here: https://forums.nxtcrypto.org/viewtopic.php?f=17&t=836 (or send me a PM). Unfortunately, I won't be able to follow this thread in the next few days.

Me when somebody uses latex

hero member
Activity: 515
Merit: 502
Is this 1626 words dictionary enough now or should we go with a bigger one, also to allow less words?

Why?

If crypto.getRandomValues is working properly then 1626 is just perfect. If crypto.getRandomValues  isn't working then even 20,000 words won't be enough, as real entropy will be limited by the random number generator.

Some people are having difficulty comprehending that 128-bit is a very big number.

Lets assume the worst case.

Lets say in two years, 1 billion people are using wesleh'sy client and  everyone created their pass phrase using these randomly generated words.  

Now lets assume that every bitcoin miner in the world  (millions of dollars of electricity per day)  all work together to find just one of these 1 billion randomly generated pass phrase. The current hash rate on bitcoin network is 3 million GH/s.

How long would it take for the entire bitcoin network combined to find just one of 1 billion pass phrase?

Total number (aprox 128-bit)  1626^12 = 3.4×10^38

1 billion users, 3.4×10^38 / 1000000000 = 3.4 x 10^29

Now the entire bitcoin network combined works to find just one pass phrase   3.4 x 10^29 / 300000000000000000  =  1133333333333 seconds

that's 36,000 years (36 thousand years).

lets assume they get very lucky and find one pass phrase only after 1% of search, that's still  3 thousand years

And after all that work (and electricity) and 3 thousand years later,  they might discover that the account they did find only had 2 Nxt  in it.

The point is that  the problem (if there is) will be with random number generator -- not with number of words, which are just fine.

That's why I keep saying you don't need 50 or 100 character password  (not even 35, actually  -- 25 is fine). . It doesn't add any more security. If there is a security hole (like for example keyloggers) even 1 million character password isn't going to help.


You're right, bitcoinpaul was only suggesting that, from the end user point of view, a 10 words passphrase is a little more convenient than a 12 words one. This is possible with a larger dictionary.
legendary
Activity: 1512
Merit: 1004
I just mean whether can use NSR to send coins which are based on parallel chains.

I think as it is being planned currently you could just use a setting (like testnet) to select "which chain" the NRS will use at startup (i.e. no different code - just a different blockchain).

really great if like as you said.
legendary
Activity: 1890
Merit: 1086
Ian Knowles - CIYAM Lead Developer
I just mean whether can use NSR to send coins which are based on parallel chains.

I think as it is being planned currently you could just use a setting (like testnet) to select "which chain" the NRS will use at startup (i.e. no different code - just a different blockchain).
legendary
Activity: 1512
Merit: 1004
all parallel chains use the same Nxt Wallet?

You do realise that Nxt *doesn't have a wallet*?

Just stupid question:
I just mean whether can use NSR to send coins which are based on parallel chains.
full member
Activity: 189
Merit: 100
Update Clienxt GUI

version 0.0.7 [codename] Bond

------
changelog:
------
* Adapted Java API instead of JSON API (Phase I).
* Support for Spanish (thanks to VanBreuk - 8118526282532613576 ), Portuguese (thanks to garcias - ) in addition to German (thanks to teletobi - 11810049982317883239), Russian, English languages.

* Support multiple account forging with timers mixed with watchlist(Phase I)
* includes nxt.jar of 0.8.8 version in lib folder
* some more multithread operations, gui bug fixed, it should be more fluid.

------
Right now, you will have to re-enter secretPhrase everytime you launch the software and continue to forge. (via Forge Button)

Phase II: (suggestions)

Would you want to have one master password (or PIN code) to encrypt and save secretPhrases?

Saving would allow you to forge even after restart with master password (or PIN code). GUI would have to encrypt and save your secretPhrases in database. Encryption will use AES(MD5(masterpassword), secretPhrase). MD5(masterpassword) itself would be saved as Java Preference (is available in Default settings menu)

In addition, Would you want to get account number with one click?

This way, user only sees account number (secretPhrase is secure and auto-generated) and with master password (or PIN code) he can retrieve secretPhrase of that account.

Please make suggestion if you find it not secure

------
Download
------
https://bitbucket.org/fmiboy/clienxt/downloads/clienxt_007.zip

------
Usage
------
Simply unzip archive and

- Windows users - double click Clienxt.jar and enjoy
- Linux&Mac users - in terminal ./Clienxt.run & and enjoy

------
Screenshots
------



------
Reports
------
report feature requests, bugs https://bitbucket.org/fmiboy/clienxt/issues?status=new&status=open


------
Support
------
Appreciate any kind of support!
Translators are very much welcome to contribute. Please donate some to our translators, as well, they are just awesome!

if you like the effort and the changes, please donate: 13792774143018875909
legendary
Activity: 1890
Merit: 1086
Ian Knowles - CIYAM Lead Developer
all parallel chains use the same Nxt Wallet?

You do realise that Nxt *doesn't have a wallet*?
legendary
Activity: 1890
Merit: 1086
Ian Knowles - CIYAM Lead Developer
That's why I keep saying you don't need 50 or 100 character password  (not even 35, actually  -- 25 is fine). . It doesn't add any more security. If there is a security hole (like for example keyloggers) even 1 million character password isn't going to help.

Agreed - unfortunately people don't *think* or bother with things like *math* here - instead they just pluck numbers like 35 or 50 "out of thin air" and make it a *rule* that must be followed.
legendary
Activity: 1512
Merit: 1004
Any suggestions about Parallel Chains? If not then I'll stick to BCNext's draft.
maximum parallel chains is ?
all parallel chains use the same Nxt Wallet?
hero member
Activity: 644
Merit: 500
Is this 1626 words dictionary enough now or should we go with a bigger one, also to allow less words?

Why?

If crypto.getRandomValues is working properly then 1626 is just perfect. If crypto.getRandomValues  isn't working then even 20,000 words won't be enough, as real entropy will be limited by the random number generator.

Some people are having difficulty comprehending that 128-bit is a very big number.

Lets assume the worst case.

Lets say in two years, 1 billion people are using wesleh'sy client and  everyone created their pass phrase using these randomly generated words.  

Now lets assume that every bitcoin miner in the world  (millions of dollars of electricity per day)  all work together to find just one of these 1 billion randomly generated pass phrase. The current hash rate on bitcoin network is 3 million GH/s.

How long would it take for the entire bitcoin network combined to find just one of 1 billion pass phrase?

Total number (aprox 128-bit)  1626^12 = 3.4×10^38

1 billion users, 3.4×10^38 / 1000000000 = 3.4 x 10^29

Now the entire bitcoin network combined works to find just one pass phrase   3.4 x 10^29 / 300000000000000000  =  1133333333333 seconds

that's 36,000 years (36 thousand years).

lets assume they get very lucky and find one pass phrase only after 1% of search, that's still  3 thousand years

And after all that work (and electricity) and 3 thousand years later,  they might discover that the account they did find only had 2 Nxt  in it.

The point is that  the problem (if there is) will be with random number generator -- not with number of words, which are just fine.

That's why I keep saying you don't need 50 or 100 character password  (not even 35, actually  -- 25 is fine). . It doesn't add any more security. If there is a security hole (like for example keyloggers) even 1 million character password isn't going to help.
sr. member
Activity: 490
Merit: 250
I don't really come from outer space.
Here's an interesting service for Bitcoin using multisignatures:

https://bitcointalksearch.org/topic/making-multibit-safe-to-use-on-computers-you-cant-trust-502830

Should multisig be implemented in Nxt, a similar but decentralized service would be a good idea to explore.
hero member
Activity: 655
Merit: 500
Jimmy Wales is discussing wikipedia potentially accepting bitcoins on reddit

http://www.reddit.com/r/Bitcoin/comments/201fa6/hello_from_jimmy_wales_of_wikipedia/


Interesting quote

Quote
What I've said for a long time continues to be true. I'm "cautiously pessimistic" about Bitcoin.
What I mean by that is that I'm loving the idea but I see a lot of real problems. One of my concerns is that the complexity of it all will hinder adoption.
I also have concerns that are perhaps born out of a lack of full understanding. What happens when BTC mining returns drop significantly lower than costs? Will transactions fees be sufficient to induce miners to continue participating?
But I'm not forecasting doom for BTC. I'm cautious. I don't know enough to say for sure.
legendary
Activity: 1890
Merit: 1086
Ian Knowles - CIYAM Lead Developer
I've completed the 1st draft of my article about forging from a probabilistic point of view: http://www.docdroid.net/9yub/forging.pdf.html

Comments, suggestions, (and donations Wink) are welcome!

Please, discuss it here: https://forums.nxtcrypto.org/viewtopic.php?f=17&t=836 (or send me a PM). Unfortunately, I won't be able to follow this thread in the next few days.

+1. Ciyam should like it a lot.

Great stuff - I hope some donations are sent for his efforts.

@mathcl - I think I found a spelling error:

...then the system should be able to fence o ff the attack by other means, which we shall not discuss in tis note.

(probably should be "this paper" instead).
hero member
Activity: 644
Merit: 500

Here's my new logic for my client http://nxtra.org/nxt-client (to be available later today, not yet uploaded)

Thoughts?

very cool adding the random word for passwd.
There are space between random words of password  or just consecutive random words without space?

Spaces
hero member
Activity: 644
Merit: 500
35 char is more than enough. Don't need to add special character

Exactly.

50 characters is even stronger than 256-bit, even though curve25519 is only 128-bit  (at best) strong.




sr. member
Activity: 247
Merit: 250
James, do you want us to just start trading it ?
legendary
Activity: 1176
Merit: 1134
Im not sure what it is doing, but it seems to be doing what it is supposed to do
Its reporting all the active peers, which are then used to calculate the most useful nodes and allocating nodecoins based on that.

current 33.33333333 0.27705624 nodecoins | sent 56.00000000

Do you see the nodecoins in your NXT AE "my assets"?

James

Yes
Code:
nodecoin 110 999'999'999 0.00001% / / / Transfer
Cool!
Now, just need more people running it to find any problems.Not sure the nodecoin allocation is quite right yet

James
legendary
Activity: 1176
Merit: 1134
Im not sure what it is doing, but it seems to be doing what it is supposed to do
Its reporting all the active peers, which are then used to calculate the most useful nodes and allocating nodecoins based on that.

current 33.33333333 0.27705624 nodecoins | sent 56.00000000

Do you see the nodecoins in your NXT AE "my assets"?

James
Jump to: