Bitcoin Forum
May 26, 2024, 11:27:28 AM *
News: Latest Bitcoin Core release: 27.0 [Torrent]
 
  Home Help Search Login Register More  
  Show Posts
Pages: « 1 2 3 4 5 6 7 8 9 10 11 12 13 14 [15] 16 17 18 19 »
281  Economy / Service Discussion / Re: Butterfly Labs CEO 25 Million USD Mail Fraud — A Concise Summary of Evidence on: September 21, 2012, 02:45:14 AM
Wow is it possible that GigaVPS could get scammed twice? Once by pirate and once by BFL?



Simple answer: Of course  it is possible.
282  Economy / Services / Re: Gigamining / Teramining on: September 21, 2012, 02:44:00 AM
Hey GigaVPS
What do you think about that:
https://bitcointalk.org/index.php?topic=110805.0?

Are you familiar with that?
Are you concerned about that, if not why?
I'm curious about your opinion as a shareholder.

283  Economy / Service Discussion / Re: Butterfly Labs CEO 25 Million USD Mail Fraud — A Concise Summary of Evidence on: September 21, 2012, 02:38:16 AM
Wouldn't it be funny if they were able to scam everyone out of their upgrade money, then on top of that get many of the people to send functional FPGAs back to them so they can run off with the $$$ and a couple TH/s of mining gear to keep the $$$ flowing in afterwards?   Grin  Grin  Grin
This is a good indicator. Once ASICs come out the FPGAs have much less value - so why would they want them back? Well, because if you get them back but ASICs never ship then you have a huge mining rig with excellent value as ASICs haven't yet pushed up the difficulty.

So not only will you be smarting from not getting your ASICs but you'll have to put up with knowing that the bitcoins you're using are being mined by your very own rigs in the fraudsters hands. Which explains why they didn't mind shipping FPGAs in the first place as they knew they actually having you pay for their own super rig.


Good point.
Is there any good idea what to do with this amount of fpga rigs?
Sell to some government for cracking passwords etc.?
284  Economy / Service Discussion / Re: Re: Butterfly Labs CEO 25 Million USD Mail Fraud — A Concise Summary of Evidence on: September 21, 2012, 01:45:23 AM
This reminds me of that Woody Allen movie where they open a cookie shop next to a bank to set up cover while they tried to tunnel into the vault. Meanwhile the cookie shop takes off as a successful business and they are stuck in a quandary as to whether they should rob the bank or not...
What title it is?
I want to watch it. Smiley


Besides, by your logic if we actually run away with the idea that this is a really elaborate con- why wouldn't they just use ASIC's as a setup for a even BIGGER scam, like ASICS v.2, or use ASIC v.2 for the even bigger scam ASIC v.3? Certainly after multiple iterations of putting out confirmed, verified, working, high-quality equipment the amount of trust the community would have in them would be through the roof. Just imaging the pre-order volume for ASIC v.5!

Answer is simple.
There is hundreds of people here even on this forum who knows that ASIC is expensive but possible, but there isn't any other possible path to get sth. faster or with higher performance with nowadays technology than ASIC (only path I can imagine is math/algorithmic breaking cryptography used).
Please correct me if I'm wrong
285  Economy / Service Discussion / Re: Re: Butterfly Labs CEO 25 Million USD Mail Fraud — A Concise Summary of Evidence on: September 21, 2012, 01:32:11 AM
but: i am only good at getting scammed, not at detecting scams. so: who knows...

Just a digression.

I really like that quote.

Hmm if my logic is ok then if you use 100% occasions to get scammed then you are quite good scam detector.
It would be sufficiency condition for no-scam-business to get not get involved by you when opportunity given.

But I really like that quote.
286  Bitcoin / Bitcoin Discussion / Re: Bitcoin-Qt/bitcoind version 0.7.0 released on: September 20, 2012, 09:09:10 PM
Thanks for new version.
Smiley
287  Economy / Economics / Re: How much BTC do you need to live at your level rest of your live without working on: September 17, 2012, 07:35:15 PM
I want about $100k/year. This works out to about 10,000btc/year. Pirate pays 6.5% per week compounded weekly, so I just need 10000/(1.065)^52 or about 379 bitcoins to retire.

Oh noes, my retirement plan has failed. I guess I'll just live on Social Security...

Not true actual due to pirate default.

Pirate default bankrupted Social Security?!?

No. Smiley
288  Economy / Economics / Re: How much BTC do you need to live at your level rest of your live without working on: September 17, 2012, 05:38:25 PM
[Warning: poster is batshit crazy]

I think bitcoin is hastening a technological singularity, so saving it for the "rest of my life" seems impossible to determine right now. I'll work until my services are no longer required, at which point I'll have to calculate what speed I can run at while living off deflation indefinitely.

[/crazy]

I understand how nuts this sounds, so when pitching bitcoin to people I kinda don't mention what I plan on doing with mine. Tongue

I like it. Smiley

Still, could you try to roughly estimate some numbers? Smiley
Just give a guess.

289  Economy / Economics / Re: How much BTC do you need to live at your level rest of your live without working on: September 17, 2012, 01:19:44 PM
As far as i can tell bitcoin doesnt hold actually wealth but is based on the worth of other currencies around the world.

As far as i can tell bitcoin doesnt hold actually wealth

Hmmm... if I have 10000 I can buy good car.
It's wealth for me.

As all fiat it has no gold or any other security, it's power comes from  technology, and prices comes from Bitcoin economy.
But...  what the hell it's connected with topic?
290  Bitcoin / Development & Technical Discussion / Re: Benefits of multisig usage? on: September 16, 2012, 05:23:18 PM
We can ad time limit to get agreement between them, if not cash will go to charity, ad some cash guarantee deposit of seller.
I have lots of ideas.

Just to pre-empt you, since you're asking about this now but the concepts have been discussed for 2 years now, start with what's already been discussed.  First, read through the examples on the Bitcoin Contracts page.  There's lot of examples mixing multi-sig with locktime, etc.  Also, for the specific buyer-seller escrow case, read through the thread that I started with Gavin to discuss exactly that -- create ways for two-party escrow without risk of coins being lost forever.

The buyer-seller problem is complicated because the situation is not symmetric, and dealing with the asymmetries requires some care to not give either party an advantage to being a dick.  I'd appreciate if you read and responded in those threads with your ideas, so that progress can continue ironing them out (but of course, read them first Smiley).




THX for links, I'll do it at night.
Smiley
291  Bitcoin / Development & Technical Discussion / Re: Benefits of multisig usage? on: September 16, 2012, 02:31:53 PM
Escrow is a great one:
Alice wants to buy a burger (shipped by priority mail Tongue) from Bob, but they don't trust each other, and neither one wants to send first. They both trust Eugene, though. Alice creates a 1-of-2 transaction which can pay to Bob once signed by either Alice or Eugene. The three scenarios:
1. Alice creates the transaction; Bob sends burger. Alice signs the transaction and Bob gets his money.
2. Alice creates the transaction; Bob doesn't send the burger. Eugene sees that Bob is a scammer and doesn't sign the transaction; no money changes hands.
3. Alice creates the transaction; Bob sends the burger. Alice refuses to pay. Once Eugene is satisfied with Bob's proof that he sent the burger, Eugene signs the transaction. Bob gets paid.

It's possible to do this with a 2-of-2 transaction between buyer and seller.  Then both parties have to find an agreeable resolution before anyone gets the money.  Thus, neither party has any incentive to try scamming the other.  However, there's a risk that the coins are locked forever if there is no resolution, so I had started a thread to discuss how it might be done without a third-party.  It's complicated, but it works if you include "risk deposits."  I think most of the complexity can be hidden under-the-hood, though. 

In most cases, you should just use a third-party.  It's very cheap for third-parties to operate because they never really "handle" the money themselves.  But one of the beauties of Bitcoin is that you can have the bitcoin network itself act as your "trusted third-party" in cases where privacy is critical, or the two parties can't agree on a trustworthy third-party.

We can ad time limit to get agreement between them, if not cash will go to charity, ad some cash guarantee deposit of seller.
I have lots of ideas.
292  Bitcoin / Development & Technical Discussion / Re: Benefits of multisig usage? on: September 16, 2012, 12:20:11 AM

 But I expect that the most common use-case will be for regular users to split their private keys between two devices (such as primary computer and smartphone), such that both devices need to be compromised for the attacker to get the coins (and the user will have to access both devices to use it).


What if one loses his smartphone? Nowadays how does it works with the double authentication in this case?

I just want to keep it quite simple, in this case he will just lost others money so they have to trust others, but we can increase number of trust persons CDE to larger one CDEFGHIJ.. and give a rules like two can lost their insurance level etc making it more and more complicated.

Benefit from that is that me could make use of my amount of money without taking risk, or manage risk without additional cost of risk put in banking system (or lower risk replacing banking system with more transparent bitcoin system).

But It's not thread about this it was just example to get some tech question.
I'm still looking for answer over forum.
293  Bitcoin / Development & Technical Discussion / Re: Benefits of multisig usage? on: September 15, 2012, 11:48:07 PM
And another Q:
In addition to previous scenario, could we obligate C, D, E to keep 500BTC till Friday on their specific addresses, and if they won't do that if one of them will default others will auto-transfer their 500BTC to me.
Do we have to engage additional institutions to track C, D, E wallets value?

This scenario lest me to lend money without takeing risk, makeing my profit shure in given time, and all risk is dived for C, D, E.

What you would say for that auto dept mechanism?

It's off topic. I'm just looking for answ. for tech. questions, if idea is worth discussion - I'll start new thread.
294  Bitcoin / Development & Technical Discussion / Re: Benefits of multisig usage? on: September 15, 2012, 11:37:53 PM
Simply put:  regular bitcoins only need to be signed by one address (private key) in order to be spent.  If coins are encumbered in a multi-signature transaction, it requires multiple signatures -- perhaps multiple, different, geographically separated computers.  Or multiple people.  Perhaps 2 out of 3 owners of a company will need to supply signatures to send the coins.

There's a very rich set of functionality that can be enabled through multi-sig.  Escrows, contracts, I can't even fathom all of them myself.  But the key is that there is no longer a single point of vulnerability for multi-signature-required coins.  An attacker will have to compromise multiple computers/people/nodes/servers in order to steal those coins.

EDIT: there's other features of multi-sig that might actually make it easier to spend [allow any one of multiple people to access them], or produce escrow such that defending against an attacker is not exactly the intent.  But I expect that the most common use-case will be for regular users to split their private keys between two devices (such as primary computer and smartphone), such that both devices need to be compromised for the attacker to get the coins (and the user will have to access both devices to use it).

Unfortunately, all this comes with a lot of extra complexity.  But it's up to application developers (like me), to try to make it useful for non-Bitcoin-experts.  And I look forward to digging into it after Armory becomes beta.

So, soft is not ready, I know it, but technically it is possible now right?
Is it possible to make transactions like this time or block dependent?
Example transaction:
Person A borrows from me 1000 BTC but as insurance he have persons C, D, E who pays 20 BTC each to ensure me that person A is worth my trust.
Person A have to send me back 1010BTC (with interest) before next Friday (block number), and if he will not, automatically C, D, E will lost their 60BTC and I will get that.
If A will send me BTC back in time C, D, E will get their money back.
Is it technically possible now?

If time relation is not possible, we could engage some third party as time responsible and trustworhy company with some satoshi-s profit.
295  Bitcoin / Bitcoin Discussion / Re: How to determine real value of 1 bitcoin? on: September 15, 2012, 05:24:25 PM
Only way is to chose some bunch of goods (lots of them and representative) and count value like inflation is calculated.

http://en.wikipedia.org/wiki/Inflation
296  Bitcoin / Development & Technical Discussion / Benefits of multisig usage? on: September 15, 2012, 05:15:59 PM
Could someone list benefits of multisig usage?
What exactly we could do?
297  Economy / Services / Re: [WTB] Your signature space on: September 14, 2012, 09:17:17 PM
OK thx both of you. Smiley
298  Local / Polski / Re: Polski! on: September 14, 2012, 08:20:31 PM
Kręcę się tu już jakiś czas i dopiero teraz tu trafiłem, pozdrawiam rodaków.

Entuzjasta z Warszawy.
299  Economy / Services / Re: [WTB] Your signature space on: September 14, 2012, 07:39:37 PM
Did anyone get a single satoshi from that?
Is there official prices per post or sth.?
300  Economy / Services / Re: Would you invest in an illegal business on: September 14, 2012, 07:36:13 PM
Short, yes I would if profit would be enough.
Pages: « 1 2 3 4 5 6 7 8 9 10 11 12 13 14 [15] 16 17 18 19 »
Powered by MySQL Powered by PHP Powered by SMF 1.1.19 | SMF © 2006-2009, Simple Machines Valid XHTML 1.0! Valid CSS!