Wonder who this solominer is? 88.6.216.9

[DeepBit]

There is no block chain to secure without transactions, and there are no transactions without the block chain.

Technically nothing stops blockchain from working without TXes. It may become worthless but still working.

[FreeMoney]

I still think the most important thing is that they have not figured out how to circumvent the proof of work...  they will eventually include transactions when you can make some btc's from it..  currently it is still mostly block reward!

Why do we know that?
We know that they solve about 30 blocks/day.

For all we know they might have found a semi-analytical/statistical algorithm that just takes a few computers and we calculate their hashing power only by solved blocks. AFAIK there's no conclusive evidence that it takes them 1.4GH/s to solve 30 blocks a day. That's just based on the current difficulty. All we know is that it still takes them at least some work to do it, otherwise they would be faster by now.

Umm, so with what 5 computers they can do 30 blocks a day, but they don't want to spring for 5 more?

In this bizarro universe they make $5k/day and have a super awesome more efficient way to find blocks, but aren't investing that money in it. O K.

[DeathAndTaxes]

I still think the most important thing is that they have not figured out how to circumvent the proof of work...  they will eventually include transactions when you can make some btc's from it..  currently it is still mostly block reward!

Why do we know that?
We know that they solve about 30 blocks/day.

For all we know they might have found a semi-analytical/statistical algorithm that just takes a few computers and we calculate their hashing power only by solved blocks. AFAIK there's no conclusive evidence that it takes them 1.4GH/s to solve 30 blocks a day. That's just based on the current difficulty. All we know is that it still takes them at least some work to do it, otherwise they would be faster by now.

Umm, so with what 5 computers they can do 30 blocks a day, but they don't want to spring for 5 more?

In this bizarro universe they make $5k/day and have a super awesome more efficient way to find blocks, but aren't investing that money in it. O K.

Maybe they take the saying "Mo Money, Mo Problems" to heart and trying to avoid becoming too rich.

[muyuu]

The voluntary fee scheme is something I saw as a potential source of problems since I first read about it. As you say, miners can't make the network pay a fees but they have the right to exclude transactions. This can lead to an uncertainty over what fee will make my transaction go through that ultimately hurts the whole system. I believe we will have to abandon this charity shop style of operation and switch to a "serious business" style of operation, where people know expected prices and are given some guarantees of getting what they paid for, in regards to fees and transactions.

I don't think building it into the protocol is necessary.  Improving the client might make more sense.  It could use historical pricing so provide users options.

Send with no fee (90% chance of confirmation within 4 hours)
Send with 0.01 fee (90% chance of confirmation within 2 hours)
Send with 0.025 fee (90% chance of confirmation within 40 minutes)

The issue I would fear with having rules in the protocol is it would be easy to game (both by miners and by large transaction entities).

I don't know, since I barely ever use the standard client to make my transactions I'm a bit biased towards thinking it wouldn't be sufficient to change common behaviours within the network. Most of the time I pay via withdrawing from my exchange (possibly the most anonymous easy way to do it) or I receive rewards from mining directly. I also run Armory. I think I've only made a grand total of 1 or 2 transactions ever from the official "Satoshi" client, and with standarisation, online wallets, speculation platforms, exchanges etc I'd expect the number of transactions made directly from the "official" Satoshi/Gavin client to be just a small percentage of all transactions. The fact that it's so bloody unfriendly and craps your HD prevents me from installing it in more than 1 computer and I bet I'm not alone.

Building a general set of conditions into the protocol would not be necessary. Just by having most of the big mining pools agreeing to a protocol for fees, a standard, predictable behaviour based on fees could be established.

For instance, if 80% of the mining power agreed not to process free (or < 0.001 *) transactions in less than 1 hour, then they would hardly get through any sooner.

* beware with setting a fee based on current exchange rates... that would need periodic revision. 0.001 for all we know could be a large sum for a fee in just a couple of years.

[P4man]

To get this back on track...

Thin foil hat scenario; deepbit is having  terrible luck (down -33%), so do some of the other pools Im looking at (bitminter for one, its also down by 30% at least). At the same time, mystery miner seems to be increasing his output (5 out of the last 20 blocks!), and the overall network hashrate is going slightly down, not up as youd expect if this is LargeCoin.

Is it somehow conceivable mystery miner is "stealing" blocks? Sniffing network packets? Bitminter and deepbit are hosted by the same company I think.

[rjk]

To get this back on track...

Thin foil hat scenario; deepbit is having  terrible luck (down -33%), so do some of the other pools Im looking at (bitminter for one, its also down by 30% at least). At the same time, mystery miner seems to be increasing his output (5 out of the last 20 blocks!), and the overall network hashrate is going slightly down, not up as youd expect if this is LargeCoin.

Is it somehow conceivable mystery miner is "stealing" blocks? Sniffing network packets? Bitminter and deepbit are hosted by the same company I think.

Wow, no I doubt it. And according to Sipa's graphs, the net speed is indeed increasing.

[P4man]

[rjk]

http://bitcoin.sipa.be/speed-lin-2k.png

Oh sorry. I'd only looked at it a day or 2 ago when it looked like it was going up. Seems to have dropped since then.

[Glasswalker]

Theoretically if you could intercept traffic going in and out of the pool's primary hosting provider, you could slurp up all the shares being submitted, but that wouldn't do you any good would it? the getwork that the pool is providing the miners is based on the pool's preferred block header, which has their address and such in it. Even if said attacker could "steal" the block, it would still pay out to said pool...

I may be mistaken in this, but that's how I see it from my understanding of how mining works.

[rjk]

Theoretically if you could intercept traffic going in and out of the pool's primary hosting provider, you could slurp up all the shares being submitted, but that wouldn't do you any good would it? the getwork that the pool is providing the miners is based on the pool's preferred block header, which has their address and such in it. Even if said attacker could "steal" the block, it would still pay out to said pool...

I may be mistaken in this, but that's how I see it from my understanding of how mining works.

Correct. However, if you can intercept shares in such a manner, you could also just provide your own work with your own bitcoind. Most users wouldn't be able to tell the difference, except that they would not be getting a payout of any kind from the pool.

[DeathAndTaxes]

Theoretically if you could intercept traffic going in and out of the pool's primary hosting provider, you could slurp up all the shares being submitted, but that wouldn't do you any good would it? the getwork that the pool is providing the miners is based on the pool's preferred block header, which has their address and such in it. Even if said attacker could "steal" the block, it would still pay out to said pool...

I may be mistaken in this, but that's how I see it from my understanding of how mining works.

That is correct.  The reward address is a part of the block thus a part of the merkle root.  The hash which solves the block only solves that EXACT BLOCK with every single transaction including the coinbase transaction which pays the pool/miner(s).

[PulsedMedia]

oh ffs, any forced TX fee will be top the detriment of BTC. Let the market decide!
Market forces are incredibly strong, you guys are not giving it enoug credit.

Forced fee of 0.01BTC today would be 5$ cents, ok that's not much for MOST of us, but how about africans? Indians? In Africa 36.2% of people live with under 1$ a day! GDP per capita is just 1200$ while in US it is 48147$, so that 0.01BTC works about 40 times higher for africans relative to their income, now would you pay 2$ per transaction? (Adjusted by living standards would probably be more like 5$)
Whoops, you just closed down 922 million people market from BTC by introducing TX fees! :O

TX fees will happen, no matter what, when block rewards go down, and more people see value in BTC. Bitcoin is still very small and niche. How many Bitcoin users are there currently? Not that many!
Market cap for all BTC too is very low still.

People will start paying TX fees overtime, introducing now any rules to stop 0 tx blocks, forced TX fees etc. would hinder bitcoin, maybe even to the point it never becomes widely adopted *oops*

The protocol is not broken, there is nothing to fix, 0 TX blocks are not an exploit or an attack.
Want TX Fees? By all means increase the required TX for Your blocks.

Eventually ppl will start including more TX fees. Last 24hrs was 6730 transactions which is extremely low when thinking globally. Now when we start reaching 100k transactions a day, which is still very low, if 0.1% includes a TX fee of 0.01, that is 1BTC a day - still very small by any standard. When we are close to having all 21million BTC in circulation, and active userbase of 100million, we are talking about something like 50million+ transactions a month. At this time even if just 25% miners adopt a minimum of 0.001BTC fee, people will be incentivized to gve TX fees, say 2% choose to include them at the minimum level that is 1000BTC a month already, still for smaller user base than PP has!

Average PP account has about 40$ on it, so at this point BTC has to be *at least* 19.04$ each (see my post in economics), but as BTC people tend to hoard them as investment etc. and no sane person keeps much money in PP, we can safely quadruple this to 160$, meaning each BTC is atleast 80$ worth -> 80k $ worth of TX fees a month with still very small adoption!
And to be honest, i see more like people having BIG balances as their savings etc. So we could probably increase that still *10 fold*.

If Bitcoin becomes a true success, to which there is currently no objections, we are going to see more like an user base of *2 billion*.
With same ratios of 2 transaction every other day per active user average, with 2% paying TX fees of 0.001BTC, that all of sudden becomes: 20 000BTC or 3 200 000$ a month in TX fees. More than 100k $ a day, more than 4k $ an hour, something like 700$ per block!

And you want to be greedy now, and slow that day from happening? (Greedyness is one of the biggest reasons new and inexperienced companies fail, hell, even i have fallen to that bear trap with a very popular site making hundreds of revenue daily!)

FYI, transaction fees has to be voluntary and arbitrary, so we can allow people a way to transfer cheaply very low quantities also (there is currently some forced fee for very tiny transactions tho, isn't it there?) for micropayments etc. other things which cannot be done with today's money gateways.

Think about a situation where BTC replaces Paypal ... And that is just the beginning!
TX Fees will happen, people will see value in that, especially if the benefits of including TX fees are heavily advertised by miners: Be included faster, support the strength of the cluster etc.

[muyuu]

Relax, man.

What we are witnessing now are market forces leaning towards not giving a flying one about transactions and just reaping the rewards of mining, because fees are typically so low right now.

Block rewards are going down soonish and transaction numbers are nowhere enough now to have a negligible fee per transaction. This is the simple reality.

We will find out the hard way. This particular miner is already setting a significant trend in the network.

[P4man]

Correct. However, if you can intercept shares in such a manner, you could also just provide your own work with your own bitcoind. Most users wouldn't be able to tell the difference, except that they would not be getting a payout of any kind from the pool.

Miners would notice they are not seeing as many shares on their stat page as they see in their miner. Within hours someone would raise hell. Now its just blamed on luck.

Mind you, it could very well be bad luck and coincidence, but deepbit 24hr luck is now down by almost 40% 45% now. Is that a record? Bitminter hasnt found a lucky block in ages. And mystery miner doesnt seem to have any effect on global hashrate. I dont claim to know how it would or could work, but it looks suspicious to me.

[PulsedMedia]

Slush luck: 78%, 112%, 101%
1day, 7day, 30day
but it's been a really bouncy couple days.

[P4man]

Slush luck: 78%, 112%, 101%
1day, 7day, 30day
but it's been a really bouncy couple days.

Slush is 3x smaller. Anyway, obviously -45% can (and  therefore, eventually will) happen, its not a statistical impossibility, its just seems to coincide with this mystery miner appearing. Add to that bitminter is seeing similar bad luck on the same host, and it begins to smell fishy.

[eleuthria]

Slush luck: 78%, 112%, 101%
1day, 7day, 30day
but it's been a really bouncy couple days.

Slush is 3x smaller. Anyway, obviously -45% can (and  therefore, eventually will) happen, its not a statistical impossibility, its just seems to coincide with this mystery miner appearing. Add to that bitminter is seeing similar bad luck on the same host, and it begins to smell fishy.

BTC Guild luck has been pretty abysmal the last two days, but I'd still not call it fishy.  Every pool has good days and bad days, we've just happened to have a few pools get their bad luck at the same time.

[P4man]

BTC Guild luck has been pretty abysmal the last two days, but I'd still not call it fishy.  Every pool has good days and bad days, we've just happened to have a few pools get their bad luck at the same time.

Yeah by itself its entirely plausible its just bad luck and I wouldnt read anything in to it, if it werent coinciding with this mystery miner.  Even so Im not exactly convinced they are related, but its an intriguing (im?)possibility.

[DeepBit]

I 'm checking my logs carefully, but don't see anything except for bad luck (and I have very good logs).
Shares are accepted, blocks are generated and approved by the network as they should.

Funny thing is that MM used my relays for publishing her blocks and I was afraid that people will look at the pie chart and start accusing me of approaching 51% again :)

[P4man]

I 'm checking my logs carefully, but don't see anything except for bad luck (and I have very good logs).
Shares are accepted, blocks are generated and approved by the network as they should.

But what if. Assume MM has root access to your server. Could you think of a way that would allow him to build blocks with shares from your miners, without causing anything weird in your logs?