I just noticed you've been replying to me inside my own messages that you've quoted. I've never found anyone who didn't know how to use quotes on this forum before. I'll un-bury your previous hidden messages:


You were doing it wrong.


I'm not saying that only I can't tell when a transaction was generated, but that it is in general impossible to do so. Blockchain.info can tell you when it first saw a transaction. Sometimes that is half an hour after the transaction was first broadcast, which can be days or years after the transaction was generated.

Can you tell me when this transaction was generated?


No, you can't.


You can 'pick up' any transaction you want to and pretend it was the 'last one'. So long as the timing is nearly right, nobody can argue with you about it. I know you say your site picks fairly, but you can't prove it.
 

I would need to know how many satoshis were bet. Then I would generate the number between 1 and 10000 for each of the transactions I have pre-generated using that number of satoshis. About 10% of them would give a number in the range 1000-2000, so on average I would only have to test 10 transactions to find a 'winning' one. I could test 200 transactions and find around 20 winnings ones, and broadcast them every 0.1 seconds for 2 seconds around the target time. One of them will appear on blockchain.info close enough to the closing time to be acceptable as having been fairly selected.


I'm not supreme people. I'm trying to help you understand where you're going wrong. You seem not to want to understand.


Some more than others, yes.


You want me to prove that your game isn't provably fair? How do you propose I do that, other than what I have done already? I have described an attack that you can carry out against the players to make sure you win. The existence of such an attack, even if you never perform it, is enough to demonstrate the lack of provable fairness.

No, I am in no way connected with bustabit. I did come up with the idea for their provable fairness - their chain-of-hashes-in-reverse thing was something I suggested to them, and rather than telling me I was wrong over and over they actually listened. Can you imagine?


I think I've given you enough pointers already. If you still don't get it I don't think there's any hope for you.

Sorry.

That's what I said.


Who are you? Is this the same guy on a different account? Or are you the "brains" behind the operation?


I keep telling you, I understand it works, but it isn't provably fair. You cannot tell what time a txid was generated. Take a look at a transaction up close. It doesn't contain a timestamp.


I understand. There are multiple transactions per second. You can pick any of them, and make your own. You understand that they are created by regular people, right? There's nothing to stop you from broadcasting your own transactions just when the 30 seconds is up, and then picking one of them as "the one".


This is nothing to do with my site. My site is a dice game, and so has quite a different provably fair scheme. My mind isn't stuck anywhere. I am attempting to point out what is wrong with your system, and why it doesn't work to protect you or your players from abuse. If you re-read what I have written above you might even start to understand it.

Yes. It's an innovative new twist on the tired old lottery concept. A +EV lottery - did you ever see that before?


I would never have heard of your site if you didn't start posting in this thread using its URL as your name. Since you are accusing pevpot of cheating somehow (when that's impossible, since it's provably fair) I thought it only fair to see where you were coming from. I found that your site isn't provably fair, for the reasons I have already given.


I'm not. I used to talk to him pretty regularly, but haven't heard from him for a while now. He's a smart guy and I like his work.


I do understand it, and have pointed out what is wrong with it in my previous post. You can't just say "we pick a random txid" and think that that is provably fair. You need to find a way of picking the outcome that is repeatable and verifiable by any third party.


I understand, but it isn't pro ably fair. I did read your FAQ, but what it describes isn't provably fair.

What you are doing is no better than using random.org to pick the winner. Yes it's fair. No it isn't provable. You could pick any number you like and tell me you got it from random.org. Or you could pick 3 different numbers from random.org and use the best one. I wouldn't be able to tell that you picked it the way to claimed to pick it.


I am using the comparison of the two sites in an attempt to explain to you how pevpot is provably fair and your site isn't.

Read up on how pevpot's provable fairness works. Nobody knows who won the game for about an hour after the result is decided. The stuff about "key stretching" is required to make sure nobody is cheating. It's not just there for fun - it's necessary. You don't have anything like it. Perhaps if you could understand what purpose it serves, you would then be able to see why it is necessary, and how the lack of anything similar on your site makes it open to abuse.


You should read more and write less. If there's anything I'm saying that you don't understand, I can try to help you understand it. The way things currently are, your site is open to abuse so it is in your interest to fix it.

Can you imagine a way that you could cheat if you didn't pick the winning number until after the game is over? It's not hard. You just buy a ticket, then pick that ticket's number as the winning number. Now can you think of a way that you can prove that you aren't doing that? That's the hard bit, and that's the bit that you can't do using your system.


I've no idea who cazkooo is. (Sorry cazkooo if I should remember you). The way he speaks sounds to me like either he understands my point and sees that I am right, or knows enough about me to know that I wouldn't be talking like this if I wasn't.

Even if all three of us *were* working together, it wouldn't matter. The game is provably fair. We couldn't cheat you even if we somehow worked together.

Edit: I just looked at your trust feedback and saw that at least three different people have already tried and failed to get you to understand that your site isn't provably fair:



I'll be leaving similar feedback if you don't stop falsely claiming your site to be provably fair. Two ways to do so:

1) fix the provable fairness; make it actually provably fair

2) stop claiming it is provably fair when it isn't

I have you at a disadvantage, since I know for a fact that I *did* exactly what you are claiming nobody would do. I bet 16 BTC on the very first draw. See the post immediately above yours for (kinda rough, screenshot based) proof.


The pot has never been as low as $5. The early-bird bonus has never been less than 0.1 BTC. I know, because Just-Dice has sponsored it for that much every week after the first week thanks to a generous donation from cowbay, the winner of the first week's draw.

And it seems a little rude to say I'm not in my right mind. I got my money in good and was unlucky not to win. It was a +EV bet. You don't find those very often.


I don't think you understand. The total "satoshi" is known 30 seconds before you pick the transaction. That gives you 30 seconds to find a nice txid that will let you win. Then you pick a txid "at random". That isn't provably fair. How can you prove that you picked it at random as opposed to deliberately picking one that let you win? You can't.

The profit is never anywhere near that low. And I bet over 10 BTC the very first week.


I bet over 16 BTC on the first round. Think about it your wrong.

One of the bets was sent from my poloniex account and the other from my moneypot account. I have a copy of the pevpot "Letter of Guarantee" for both deposit addresses.

What you're probably missing here is that Ryan is known to be trustworthy. He has handled massive amounts of bitcoin at bustabit with no complaints from anyone as far as I've heard. I was happy to trust him with my 16 BTC, and am confident that I lost it fair and square.


I know for a fact that it wasn't him the first week.

So could anyone else. You also have a small chance of losing 10 BTC. If you want to take that risk, do so. It's +EV after all.


I don't see them when I visit. Scan for malware.

If you're saying you are seeing something that makes you think the site operator is cheating, please share it with us. The owner went to quite some lengths to set things up such that it is impossible for him to cheat. If you see a way he could be cheating I would love to hear about it.

I can't help but notice you have what looks like a URL for your account name. I wonder if perhaps you run a competing service. Could that maybe be colouring your vision?

I checked your so-called provably fair system and it's kind of laughable. You use the txid of the "last bitcoin transaction".


The problem is that the hash of a bitcoin transaction is entirely predictable. I can pre-generate thousands of them and have them ready for broadcasting. Then I pick a bunch that make my ticket win and broadcast them at just the right time. If I was you, I could then pick one of them and claim that it was "the last bitcoin transaction". In other words your system isn't provably fair at all. It's exploitable by players, and cheatable by yourself.

I think you should rethink the system you use before making such a half-assed complaint against a better thought out game.


Check the first draw. I bought a ton of tickets and lost. I haven't played since.

How could you possibly know that no sale of that account happened after September 2014?

Please don't present your guesses as facts. If you have evidence that a single person was in uninterrupted control of the account from September 2014 until December 2015 please present it. If you don't, please don't pretend that you do.

I wrote a Python script that lets me pipe 'clamd' style RPC commands to it. I was having trouble running up against the maximum shell command length when trying to use clamd to create and send large raw transactions.

It lets me do things like this:



The Python source is here.

Edit:

I should mention that the script comes with no warranty. It nearly cost me 3200 CLAMs earlier:



I've since fixed that bug, but there could be others.

---

The 50k sell was mostly matched with the 129 BTC buy wall. So you appear to be saying that you just bought 43000 CLAMs from yourself for 129 BTC? That seems like a silly thing to do on an exchange. Couldn't you have come to a private agreement between your various selves off-exchange? Or were you sacrificing the fees to bump CLAM up to 2nd place by daily volume?

Just kidding. We know who sold into that buy wall. It was a Just-Dice investor. I helped him make the withdrawal, and he announced his dump seconds before it happened.

It's pretty simple: master-p stopped returning coins.

Some new account claimed to be him and said he had sold his account; he signed a message to prove it - but the sale was only for 2 BTC so that makes no sense

Some other new account claimed to be him and said he had been hacked; he signed a message to prove it -  but the alleged hacker left all the stolen coins in the escrow wallet for days so that makes no sense either

I expect it's just a regular exit scam like many others.

Did I miss anything of substance?

It's been a while since I checked this thread, so sorry if this has already been answered.

What's the status of the work on making SWC available to those of us who don't run Windows?

Thanks.

I think the key element of "trust network" that you are missing is the word "network".

I never traded with you, and I don't know whether to trust you. But a guy I regularly trade with regularly trades with a guy who says we would trust you with up to $100. I only have very few contacts, but they each have people they trust, and the network fans out like that. Don't treat people's trust rating here as meaning anything at all. The last I heard my rating was negative because some guy got mad when I criticized a site he owned and so he left me unrelated negative feedback as punishment. Instead look at my feedback, so whether you trust any of the people who rated me, and make your decision accordingly.

Reading this might help.


If you're a newbie, you send first. That's the cost of being a newbie.


If the newbie uses escrow every time he trades, he is forever a newbie. If you never give him the chance to steal from you, how can you know he won't steal from you when you give him the opportunity?

You can see for yourself:

Profile -> Forum Profile Information -> Show post count by posts ->

Total time logged in: 109 days, 17 hours and 59 minutes

I think it's pretty much the opposite. In absolute terms you will generally make more running an exit scam than operating legitimately, for far less effort. When I was running Just-Dice, I was taking 10% commission on profits, which is a healthy amount. The site profited by around 20k, so I was able to take around 2k in commission. When I shut the site down the bankroll was in the region of 50k. That's 50k that I could have just kept rather than returning. By returning it, I turned down the chance of "earning" 25 times more than I had earned in the previous year of hard work. I will likely never earn that much again, so the argument that "trusted people won't scam because they earn more by being honest" doesn't work. You make more stealing than you do being honest. Which, I guess, is why we see so many people stealing.


It's conceivable that he went into the escrow business with a view to running it legitimately, and only decided to scam when it turned out not to be worth running honestly. (Or, maybe the fairy tale about hackers is true; who knows!)


I'm half considering setting up an automated escrow service for signature campaign managers to use to escrow to safeguard payments to campaign members. But seeing you and Blazed both write about how it's all too much hassle, I'm having second thoughts.

Do you think people would use a service like that if there was a small percentage fee to cover expenses?

How long do I have to get to level "Bronze I"? Or is it already too late?


Can't you just rehash each password the first time they log in under the new system? It's annoying to force us all to change our passwords. What's the reason for this? Was the old password database compromised? If not, why force us to change them? Is it OK for me to use the same password I used before? Or are you thinking it may have been compromised already on your end?

Edit: I reset my password. It told me:


Please fix it.

That site itself lists the problems with 2-party escrow. Once the buyer sends the coins, the seller can still refuse to ship the goods, and demand a cut of the escrowed coins before refunding them.


When you're one of the three people and the scammer is the other two, you're fucked. That's what's wrong with 2 of 3.


I left negative trust. I think it's important to leave it in place, at least until the account is in safe hands.


This makes no sense. How did the hacker know that you didn't have the wallet seeds written down on paper? Encrypting one copy of a wallet doesn't encrypt other copies of it.


I think he is saying that he has backed up and restored the wallet he recovered from the hacked machine, and that that is the only copy he has access to.


So you had multiple copies of your wallet, but then one of the copies started giving you problems so you figured the best thing to do was to delete all the backups? That's just crazy.


So you do have backups after all? Why not use those to recover the coins?

Be careful. Just because one of my privkeys signed the first input to a transaction, it doesn't mean I agree with the content of the CLAMspeech. The transaction could be a withdrawal from Just-Dice. When withdrawing from Just-Dice you can specify any CLAMspeech you like. The transaction's inputs will be signed by private keys in the hot wallet, but the owner of those keys doesn't endorse the content of the CLAMspeech.


I worry that we're going to end up hard-coding many different similar-but-different features into the CLAM client, and wonder if there's some way we could generalize it so we don't have to edit C++ code for the next person to dream up a use for CLAMspeech.

(Cue sighs and groans from Creative and Kef...)

Why not?

You've admitted that you are selling stolen goods. Does that mean I shouldn't warn people that you are selling stolen goods?

Quoting to reference in trust feedback.