elective-stereophonic
elective-stereophonic
Show Posts - mthcl  
Please login or register.

Login with username, password and session length
Advanced search  

News:

Latest Stable Nxt Client: Nxt 1.12.2

Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.

Messages - mthcl

Pages: [1] 2 3 ... 21
1
Nxt General Discussion / Re: Price speculation
« on: February 11, 2018, 12:32:27 am »
...  when Come-From-Beyond and BCNext were running the show ...

this "and" sounds interesting   ;)

2
Nxt General Discussion / Re: Sergey Ivancheglo is BCnext
« on: July 22, 2017, 01:11:02 am »
c'mon, of course he's BCNext. Well, rather, BCNext is him   :)

3
Merci beaucoup, Ludom!




4
Transparent Forging / Re: Math of Nxt forging
« on: December 22, 2016, 09:51:10 pm »

6
Nxt General Discussion / Re: Price speculation
« on: December 22, 2016, 07:48:18 pm »
http://www.ledgerjournal.org/ojs/index.php/ledger/issue/view/2/showToc - I've published that "math of nxt forging" paper in an academic journal

7
Nxt Community News and Announcements / Re: IOTA - JINN
« on: January 31, 2016, 11:35:46 am »
HI, you should update the opening post to make obvious that the Jinn to Iota exchange is finished  .

Yesterday,  I did send Jinn to the NXT address and entered the iota address in http://collect.iotatoken.com/ . I had no message on the way saying that the collect is finished.

What is going to happen to my Jinn token ? Will they be changed into iota or refund to my NXT account ? ( nxt account : NXT-J6BE-JL6S-EY39-F7PMD )

Nicolas
See https://bitcointalk.org/index.php?topic=1262688.msg13605469#msg13605469

You have to contact Come-from-Beyond (on bitcointalk, not here).

8
Consensus Research / Re: Interactive Proof-of-Stake
« on: January 16, 2016, 11:44:25 pm »
Anyhow, in a week or so the BT adjustment algorithm will change, and it will become virtually impossible to play games with it.

Hi, could you please clarify why the new adjustment algorithm makes grinding impossible? The only improvement I see is that BT is not adjusted every block. It is adjusted every 2 blocks, so indeed an improvement, but not really impossible I think. Most probably I don't understand something. Thanks.
As kushti observed just above, in the current situation the forger may delay a block a little to influence the BT adjustment. Since currently the BT can change a lot (two times!) in just one step, every possibility to play with it can be considered a danger. With the new algorithm delaying one block makes a very small effect on the BT, because of the averaging.

9
Русский (Russian) / Re: Форки и развитие NXT
« on: January 16, 2016, 08:49:07 pm »
Интересно, не хочет ли кто-нибудь сделать форк некста с алгоритмом Transparent Forging-а, описанным в разделе 8 в https://www.docdroid.net/ecmz/forging0-5-2.pdf.html ?   А то нанять кого-либо у меня нет денег, а последнюю мою программу я вообще написал больше двадцати лет назад. В принципе, на основе этого алгоритма можно сделать "надёжные" instant transactions, что ценно. 

10
Consensus Research / Re: Interactive Proof-of-Stake
« on: January 15, 2016, 12:14:50 pm »
Anyhow, in a week or so the BT adjustment algorithm will change, and it will become virtually impossible to play games with it.

11
Consensus Research / Re: Interactive Proof-of-Stake
« on: January 13, 2016, 09:10:00 pm »
OK, assume R=16, l=10, for definiteness. Imagine that there there is someone who controls, say, 10% of the stake (or, maybe, several big holders collude so that they do control 10%). Assume also that other accounts are not so big, say, each controls less than 0.1%.  Then, the 10% holder would be able to forge really a lot of blocks in a row, by dividing his stake into 100 equal parts (with overwhelming probability at least one of his accounts gets the maximal m=R).

What do you think of this attack?  In general, the situation that many small holders cannot forge anything together is worrying, I think...

I'll check both IPoS / Nxt against that, but intuitively, in this case a system is probably vulnerable. I don't worry much about such a scenario though. Economy is about Pareto distribution of wealth. Paper ( http://arxiv.org/abs/1308.3892 ) shows the Bitcoin is about stretched exponential distribution. I can't imagine a system with a single 10+% account and others hold < 0.1%.
Not necessarily single account, but maybe a collusion  of several big holders. Anyhow, that's all speculations, for now.

Could you clarify also this:
Quote from: mthcl
As a side note, I don't understand why you write "However, it is possible to iterate over delta" when discussing the Nxt algorithm on p.3.  Delta is in the right-hand side of (2), which is monotone, and it is not hashed. What advantage could it bring to the attacker then?


12
Consensus Research / Re: Interactive Proof-of-Stake
« on: January 12, 2016, 10:33:56 pm »

Yes, small-stake accounts as well as big stakeholder generate a disproportionally low number of tickets. So for a big stakeholder, aside of attacks, there's the economic incentive to split stake into middle-class accounts.

For attack with stake-splitting, now best number of accounts for 33% stakeholder is about 180(R=16, l=10)(old number with b^m was about 96). Bigger swarm reduces chances to generate a better chain than network's.
OK, assume R=16, l=10, for definiteness. Imagine that there there is someone who controls, say, 10% of the stake (or, maybe, several big holders collude so that they do control 10%). Assume also that other accounts are not so big, say, each controls less than 0.1%.  Then, the 10% holder would be able to forge really a lot of blocks in a row, by dividing his stake into 100 equal parts (with overwhelming probability at least one of his accounts gets the maximal m=R).

What do you think of this attack?  In general, the situation that many small holders cannot forge anything together is worrying, I think...

13
Consensus Research / Re: Interactive Proof-of-Stake
« on: January 11, 2016, 09:27:06 pm »
Well, even with R=8 there is a big problem with the "hidden chain" attack performed by the guy with max balance. Assume, for example, that the richest guy has 5% of the stake, and others less than 1%. Then, even if he forges alone, he'll get sometimes very heavy weights (5⁸=390625), one his block will easily overweight a very long blockchain created by all others.

Oh, that's true. I've fixed ticket's score formula with m * log2 b. Happily, simulations show the updated formula works better against attacks, so adversarial power IPoS is claimed to be safe against is raised to 1/3(33.33%). I've added you to Acknowledgement section  :)  Uploading fixed paper to the Arxiv...

It's m times binary logarithm of b, correct?  Then there should be kind of "best splitting strategy" to maximize the forging chances (with logarithms, it's clear that getting the highest possible value of m is the best strategy). It would be interesting to do the calculations...  But it is important to observe that if everybody uses some kind of splitting strategy, then this l parameter (the number of blocks you must skip before forging the next one) is essentially unimportant (if you have a lot of small accounts, you won't "feel" this restriction).

Also, do I understand correctly that with this algorithm "small" accounts will never forge?  I mean, assume that a rich guy has balance B, and splits it into (say) 10R*l equal accounts. Then, any account that has less than B/(10R*l) has almost no chance to forge?..


14
Consensus Research / Re: Interactive Proof-of-Stake
« on: January 10, 2016, 11:32:29 am »
As a side note, I don't understand why you write "However, it is possible to iterate over delta" when discussing the Nxt algorithm on p.3.  Delta is in the right-hand side of (2), which is monotone, and it is not hashed. What advantage could it bring to the attacker then?

15
Consensus Research / Re: Interactive Proof-of-Stake
« on: January 09, 2016, 08:47:58 pm »
Well, even with R=8 there is a big problem with the "hidden chain" attack performed by the guy with max balance. Assume, for example, that the richest guy has 5% of the stake, and others less than 1%. Then, even if he forges alone, he'll get sometimes very heavy weights (5⁸=390625), one his block will easily overweight a very long blockchain created by all others.

16
Consensus Research / Re: Interactive Proof-of-Stake
« on: January 09, 2016, 07:14:50 pm »
I'm through the paper; but let me first ask the following to remove any doubt: do you really mean that the score of a ticket is b^m (b to power m), and not simply bm? I thought it was a misprint, but it happens two times, on pp. 4 and 5...

Alright, b^m (b to power m). I also tried b*(m^k), with k=8 seems more or less ok. With k=1, big stakeholders have too much advantage.
And what value of R you're thinking of? 16, as in the paper?

17
Consensus Research / Re: Interactive Proof-of-Stake
« on: January 09, 2016, 05:46:28 pm »
I'm through the paper; but let me first ask the following to remove any doubt: do you really mean that the score of a ticket is b^m (b to power m), and not simply bm? I thought it was a misprint, but it happens two times, on pp. 4 and 5...

18
Consensus Research / Re: Interactive Proof-of-Stake
« on: January 05, 2016, 02:03:22 pm »
Thanks for sharing! Will read...

20
Transparent Forging / Re: Transparent Forging - the latest info
« on: November 13, 2015, 07:21:06 pm »
No, the solution of my paper was not mentioned by CfB, but I recall he had a favorable impression about it. In any case,
 (1) I know for sure that CfB is very occupied now, so don't expect too much from him;
 (2) as JLP said, all this is not urgent. Of course, it would be nice to have the solution of my paper tested; but I cannot do it myself (I wrote last program around 20 years ago). Anyone wants to create a Nxt clone with this TF algorithm?..

Pages: [1] 2 3 ... 21
elective-stereophonic
elective-stereophonic
assembly
assembly