Ooo I like this.  Quadratic speculation!

The bitcoin trademark holder?  Who holds the trademark for "internet"?  What about trademarks in different countries?

As for the bitcoin.org domain owner, what about the bitcoin.com domain?  What about bitcoin.net, bitcoin.gov, bitcoin.xxx, or bitcoin.bit? not to mention bitcoin.jp, bitcoin.de, bitcoin.us, bitcoin.co.uk, ... et cetera?  If we're only worried about "established" bitcoin sites then who gets to own bitcointalk.org, bitcoincharts.com, and bitcoinwatch.com?

If these are real problems then I have a few more:

• Who gets to be the Bitcoin CEO?
• Who owns the patents on the basic concepts behind Bitcoin?
• Who owns the copyright on the blockchain?

To me at least, the formation of an organisation to protect and promote Bitcoin (at least during the bootstrapping decades) very much seems to be an optional good rather than a necessary evil.

To compress so much fail into such a short piece of text is simply artful.

I would support this and think it's good for Bitcoin's future.  To me, Bitcoin is a child and needs protection and support.  In time, Bitcoin will mature to adulthood and, provided it was raised properly, there should be nothing any one person or organisation can do to control it.

I would suggest that such an organisation should have a constitution which undergoes public scrutiny before being established.

Ouch!  I didn't realise this stales situation was so sensitive to the end users configuration.  My current round data is:

Current Speed	Valid Shares	Invalid Shares	Efficiency
720.00 MHash/s	8,161	175	97.90%

So I've only dropped about 2%.  I'm using phoenix with an aggression of 12 for what it's worth.

I'm perfectly happy with receiving only bitcoins.


Interesting.  The thing I noticed was that after merged mining came online, there were significant rejects when moving to a new Bitcoin block.  The Namecoin longpolls don't seem to generate any significant rejects for me.  There was a predictable 10 second period of rejects when moving to a new Bitcoin block (this has now shrunk to only 6 seconds it seems but I have few data points to go on).

It's a bit upsetting to lose my 99.9% efficiency average but I'll live.

While the rejects are certainly a problem I would much rather use a 98% efficient merged mining pool than a 99.6% efficient normal Bitcoin pool.

Good news.

I'd like to send shadders some BTC for the hard work, do you have a link for me?

There appears to be some kind of delay but this is probably a clock sync issue.  I'm more concerned by the gaps of rejects I seem to now be getting.  Here's some data (09:45 -> 10:45 UTC, 2011-10-25):
{
My miner's state:
  09:46:19 - Disconnected (Presumably you are brining merged mining online here)
  09:48:01 - Connected
  09:48:01 - NMC 24299
  09:51:03 - New job
  09:57:15 - New job
    (all shares here are rejected)
  09:57:25 - BTC 150559
  10:03:25 - New job
  10:05:20 - New job
  10:05:21 - NMC 24301
  10:07:25 - New job
    (all shares here are rejected)
  10:07:36 - BTC 150560
  10:08:07 - New job
  10:15:23 - New job
  10:18:38 - New job
  10:20:00 - New job
  10:24:40 - New job
  10:24:40 - NMC 24306
  10:33:11 - New job
  10:36:05 - New job
    (all shares here are rejected)
  10:36:15 - BTC 150562
  10:37:50 - New job
  10:39:49 - New job
  10:39:53 - New job
  10:39:54 - NMC 24309

Bitcoin block explorer:
  09:50:23, 150559
  10:04:36, 150560
  10:23:32, 150561
  10:32:47, 150562
  10:39:22, 150563

Namecoin block explorer:
  09:45:45, 24300
  10:00:55, 24301
  10:03:27, 24302
  10:07:29, 24303
  10:14:26, 24304
  10:16:42, 24305
  10:18:39, 24306
  10:28:47, 24307
  10:36:06, 24308
  10:37:52, 24309
  10:40:58, 24310
}

Is this all as it should be?

Also, when you do start applying DGM to the Namecoin blocks, will there be namecoins in my account to withdraw or will they be automatically traded for bitcoins.  I'd guess there are a fair few miners who don't care much for namecoins and so some options for payout would be nice:

1) Donate all earned namecoins to the pool.
2) Withdraw earned namecoins to a namecoin wallet.
3) Have the pool convert the namecoins to bitcoins and added to my Bitcoin balance (with the pool taking a fixed fee for this service - 5% say).

I'm glad to hear the issue is known and am looking forward to a working merged mining system.

I'm slighly confused by the new work pushes I'm currently receiving.  Some new jobs come with a BTC block number and they correlate quite well with the blocks reported by the Bitcoin block explorer.  However, I can't find a clean correspondence between new jobs and Namecoin blocks and this worries me.  Is this normal?  I really would expect to receive a new job for every new Namecoin block.

For 21:00 -> 22:00 UTC (2011-10-24)

My received new jobs:
  03:38 (BTC 150495)
  08:43 (BTC 150496)
  18:43
  23:07 (BTC 150497)
  33:07
  43:08
  46:48 (BTC 150498)
  56:48
  57:31 (BTC 150499)

Bitcoin block explorer:
  02:57, 150495
  08:00, 150496
  22:24, 150497
  44:08, 150498
  56:31, 150499

Namecoin block explorer:
  09:51, 24197
  16:29, 24198
  24:25, 24199
  32:00, 24200
  34:29, 24201
  40:03, 24202
  45:29, 24203
  45:59, 24204
  50:10, 24205
  55:45, 24206

Thoughts?

We've just had a 65% increase in Namecoin difficulty and consequently the income bonus from using merged mining has plummeted to little more than 20%.

Sorry to hear you were ill.  I'm glad to see that the pool is behaving properly now (blocks are appearing just as with the other merged mining pools).

It's regrettable that the hashrate is down but that at least means the miners here will be enjoying the promotions for longer.

This is not a problem, it is a fact about the reward system.

This is true but completely irrelevant to the issue of pool hopping.

GoldenBoar, your recent comments betray your lack of understanding and it's clear to me that you won't comprehend the mathematics without some careful thought and study.  If you genuinely want to learn then feel free to PM me.

I'm sorry Mike for hijacking your thread, I got carried away.  I'll stop now.

No I don't.  What is wrong with it?


If you start mining now then your chances of your shares being in the range (X-N, N] are exactly the same as they would be at the beginning of a round.  The expected round length for the current round is precisely 2,140,994 + 1,468,195.430 = 3,609,189.43 shares (basic corollary of the lack-of-memory property) and it gets longer with every single non-block-generating share submitted.  Suggesting that the round is close to an end because many shares have been submitted is like suggesting that a heads is due after flipping a fair coin five times and getting tails each time.

You are more than welcome to try to "reverse pool hop" the PPLNS pool here, you will be helping the miners out by lowering the variance slightly and you'll get a fair reward for your shares in the long run.

Lol, sorry about that.


The value of your shares changes as more information about the near future of the pool is revealed.  This has nothing to do with whether or not you continue to mine at the pool yourself.  If I submit a share it will lose value as non-block-generating shares are submitted and gain value when blocks are found; N pool shares after my submission the true value of my share will be fully revealed.  Exactly the same happens at proportional pools and, indeed, with most reward systems.  A share is behaving very much like the scratch card I described which has a 1/4 chance of being a winner ($4 prize).  The value of the scratch card is $1 (expected value) but, when it is scratched to reveal no prize its value decreases $0.  The risk of this decrease in value remains whether or not you decide to buy more cards before scratching the first.

The only two sound ways of mining I'm aware of which don't decrease the value of your share as future non-block-generating shares are submitted (i.e. have zero share variance) are PPS and PPLNS with N = 1 (effectively solo mining).  These are both completely hop-proof and don't favour 24/7 mining over intermittent mining.  PPS requires a fee to be sustainable (at 0% the pool balance will follow a random walk and will become arbitrarily large and negative at points in the future).  This problem cannot be avoided (SMPPS tries to avoid it by placing the random walk along the time axis - this is more subtle but equally flawed).


I agree that the mechanic for hopping is similar to that for a proportional pool.  However, a proportional pool can be effectively hopped with information from the past whereas, a PPLNS pool can only be exploited using information from the future.  I agree that you don't need to know exactly when the round will end but the lack-of-memory property of the distribution of future blocks w.r.t. shares tells us that the current round duration gives us absolutely no information on the expected number of blocks in the next N shares.

The description you quoted from Eligius is spot on; this is the basic idea behind PPLNS.  A real implementation will be slightly more complicated because it must account for changes in N and/or in difficulty and must also be careful about what happens when the pool starts and ends.

Yes, if you stop mining and none of your shares end up within the (X-N, N] range then you get no reward.  If there is one block in the range (X-N, N] you get 50/N BTC (assuming no transaction fee rewards and 0% fee), if there are two blocks in the range (X-N, N] you get 100/N BTC, and so on.

Let's suppose we have a PPLNS pool with N = difficulty/2 (and assume difficulty is fixed for simplicity).  The PPLNS pool here is a good real world example.  The average number of shares to a round here is 2*N (because share difficulty is 1 here).  Some rounds are much shorter and others are much longer but most are longer than N.  It is not possible to hop this pool in the way you describe without some kind of time machine.  To "reverse pool hop" you need to know when a round is going to end.

What do you mean by "and I stop mining before N"?

If you decide to stop for gaming then your shares don't disappear.  Each share will generate funds for you for any and all blocks which are found by the pool during the pool's next N shares (whether or not you mine for this period is irrelevant).  If the pool finds no blocks during the N shares following your submission then your share has failed to generate any BTC for you at all (again, whether or not you mined at the pool during this time is irrelevant).

PPLNS means that when a block is found, the reward is split uniformly between the most recently submitted N shares.  Such a reward system neither rewards nor punishes intermittent miners in terms of average income.

In PPLNS, N is normally either constant or tied to difficulty, not to "round shares".  If N were "round shares" then the reward system would suffer the usual pool hopping problem faced by proportional miners.  When N is 1 the reward system is identical to solo mining (which also neither punishes nor rewards intermittent miners in terms of average income).

Absolutely, this reward system is vulnerable to "reverse pool hopping".  Shares at the end of a round are more valuable than shares at the beginning in general.  If a hopper hops into the pool for the last N shares of each round then they are going to earn much more than a 24/7 miner or random intermittent miner.  However, this strategy requires the ability to predict the future.

I don't see how PPLNS is "geared for the 24/7 miner" at all.  Share variance is higher than with other methods but, with N as large as difficulty/2, the effect is not so bad.  (Approximately: It is like comparing a $1 bill with a scratch card which has 1/4 chance of being worth $4.)

Certainly, the expected income from either 24/7 or intermittent mining at 0% PPLNS is identical to that for 0% PPS.  Average 0% PPS payout with no operator risk is one of the reasons I like this reward system.

Shares submitted to a Prop or PPLNS pool become worth less and less with time regardless of whether or not you continue to mine.  The average income per hour will be the same whether you mine 24/7 or mine intermittently (at random).  If you don't want your shares to lose value with time then PPS is practically the way to go (SMPPS when the pool balance is positive is just as good).

SMPPS is dangerous to use 24/7 because the reward system necessarily creates arbitrarily large pyramid schemes and, eventually, one will be big enough to kill the pool and any miners caught in it will lose big time.

I disagree.  Most people select an avatar which is much more cheery than they are in real life and I doubt this is an exception.

The BTC promotion applies to all of the BTC/NMC pools.  The NMC promotion only applies to the PPLNS pool.

What is it about the PPLNS pool which means you can't use it?