|
Just a shot on the dark: try finding a smallest possible USB drive and format it FAT16, not the usual FAT32.
|
|
|
|
No downtime to report.
Are you allowed to report how much you pay per month to Barret Lyon and his friends? Edit: Oh, noes, it looks like Barret had sold Prolexic. |
|
|
|
I think critics of C++, cannot code C++.
The problem isn't coding, its maintaining the code. The quality of C++ implementations leaves a lot to be desired. And those desires will probably never be fulfilled. Lets revisit the issue when you integrate your boost-using libbitcoin with some other system written with a different release of boost or that just requires to be compiled with a different set of flags. You will probably end up like many other C++ projects: maintaining a private boost tree where every instance of the identifier "boost" was replaced with "god_please_forgive_me_i_will_never_do_it_again". But for now have fun while you still have your hair un-pulled. |
|
|
|
Am I missing anything on the technical side? Is this a worthwhile idea?
On the technical side the detail you missed is that currently the block timestamp has almost two-hours of leeway. And Luke-Jr at Eligius is actually actively using this leeway for some internal optimization and is adamant about keeping doing it that way. On the long-term viability side there's a bit of defect in the sense that the meter-maid collection-rounds work as a sort of crime deterrant. Car-owners will rationally avoid parking spots where the attendant was replaced by a photodetectors. |
|
|
|
I am trying to understand the SelectCoins function (at least at a high level):
It computes an approximate solution to an one-dimensional knapsack problem. In textbooks it is typically posed as finding a maximum lower than the size of the knapsack (to minimize the empty space). Here it is inverted: the goal is to find minimum higher than the size of the knapsack (to minimize the change required). Whatever you do, don't reinvent the wheel, there's a rich literature on this problem. |
|
|
|
I did some benchmarking of client startup.
Did your profiling tool allocated some time to I/O wait? Seems like you've enumerated a CPU-time profile, not a wall-time profile. Thanks. |
|
|
|
There are many other scenarios.
Eg. Dun & Broadstreet will start selling ratings for each address: pay 10BTC and you'll know whether the address is "green" or "red".  |
|
|
|
God damn it.
Take it easy. It is highly unlikely that Oracle will go after you. Unlike Silk Road there's no criminal activity involved, this is at worst a small commercial dispute that would most likely still fit under the small-claims rules. I speak about this as somebody who "illegally distributed" some information from the Oracle Technology Network. When I was contacted by Oracle legal they wanted us to add the proper (TM),(R)&(C) signs after their trademarks. Those guys are smart, they don't go after the pikers and they understand how building up a software business takes time and many beta releases before actually going commercial. It took us several years to go from "illegal distributors" to "gold partner". It is quite incorrect to make direct comparisons between Microsoft and Oracle as far as licensing goes. With Oracle you can still register for free and download fully functional non-expiring top-of-the-line database system. All you have to do is click the check-mark that says "I will use it for R&D purposes". I'll finish by emphasizing that my post is about the forum, not about the Solidcoin software and its distribution. |
|
|
|
"use special-case bullshit address DevNull0DevNull0DevNull0DevNull0000000 for coin burning"
If I remember correctly, the only EC public key assured not to have a private key is all zeros (or some transformation of it needs to be all zeros). My cryptography books are in storage, so I won't be able to contibute. But it would be nice to have a citation for a mathematically guaranted black-hole address. There was somebody on this forum posting the elliptic curve crypto code in the programming language J which allowed the use of arbitrarily small finite fields. In a small field one could simply do an exhaustive search for the subfield with no solutions. And then generalize it for the NIST fields and curves. Alternatively, you could generate a valid key pair on a computer disconnected from the network, write down the public key, burn the whole computer and post the video of the whole stunt on youtube. Probably more people would believe that such a key is a black-hole than when you would publish a mathematical proof. |
|
|
|
Guys, here's a context diff for few % faster implementation of scrypt.c: *** scrypt.c.bak 2011-10-12 21:52:49.166078144 -0700
--- scrypt.c 2011-10-12 22:53:33.717402266 -0700
***************
*** 31,75 ****
--- 31,94 ----
#include <stdlib.h>
#include <stdint.h>
#include <string.h>
+ #ifdef __GNUC__
+ #include <byteswap.h>
+ #endif
static inline uint32_t
be32dec(const void *pp)
{
+ #ifdef __GNUC__
+ return bswap_32(*(uint32_t const *)pp);
+ #else
const uint8_t *p = (uint8_t const *)pp;
return ((uint32_t)(p[3]) + ((uint32_t)(p[2]) << 8) +
((uint32_t)(p[1]) << 16) + ((uint32_t)(p[0]) << 24));
+ #endif
}
static inline void
be32enc(void *pp, uint32_t x)
{
+ #ifdef __GNUC__
+ *(uint32_t*)pp = bswap_32(x);
+ #else
uint8_t * p = (uint8_t *)pp;
p[3] = x & 0xff;
p[2] = (x >> 8) & 0xff;
p[1] = (x >> 16) & 0xff;
p[0] = (x >> 24) & 0xff;
+ #endif
}
static inline uint32_t
le32dec(const void *pp)
{
+ #ifdef __GNUC__
+ return *(uint32_t const *)pp;
+ #else
const uint8_t *p = (uint8_t const *)pp;
return ((uint32_t)(p[0]) + ((uint32_t)(p[1]) << 8) +
((uint32_t)(p[2]) << 16) + ((uint32_t)(p[3]) << 24));
+ #endif
}
static inline void
le32enc(void *pp, uint32_t x)
{
+ #ifdef __GNUC__
+ *(uint32_t*)pp = x;
+ #else
uint8_t * p = (uint8_t *)pp;
p[0] = x & 0xff;
p[1] = (x >> 8) & 0xff;
p[2] = (x >> 16) & 0xff;
p[3] = (x >> 24) & 0xff;
+ #endif
}
You can use this code for other SCRYPT based coins as well. There's quite a bit of opportunity for further optimization, especially on the 64-bit systems, but I just don't have time to do it now. If somebody uses a Visual Studio compilers for Windows, the equivalent of "bswap_32" is "_byteswap_ulong". Have fun. |
|
|
|
Bitcoin shouldn't have gone public, it should have remained as a closed experiment similar to how the internet began between academic institutions, even Facebook started along those lines. Gradually more people that actually have an interest and understanding of the technology would have got involved until it was ready for a public release. Sadly it's too late now, it was released into the wild before it was ready and it's reputation is tarnished as a result.
Very nice and succint summary. I would just add that the Internet, unlike Bitcoin, didn't have the hot mining-lottery-tickets decaying with half-life of 4 years. So the ignition mechanism was built-in from the inception. |
|
|
|
|
10) use the correct network byte order everywhere in the protocol
|
|
|
|
Question for technical people. Would it be a bad idea if I told people to add "addnode=<myip>" to their litecoin.conf file so that they will immediately talk to my client for the genesis block? Would that overwhelm my client and take my client down? Or does bitcoin handle that well?
By default Satoshi client accepts 125 incoming connections. The flag is "-maxconnections". Then there's kernel per-process limit "getconf OPEN_MAX" which is 1024. That's all if you running Linux. On Windows you'll have to run a Windows Server to accept many incoming connections. I didn't read you code, I'm assuming you did a competent job modifying it. |
|
|
|
Have you considered the possibility that the law is complete and utter bullshit?
In general I think that US securities regulation is decently good. I particularly like the portions about fraudulent sale practices. Edited to add: The other thing that I find neatest is that non-trading entities have by law free access to the full market information just for the cost of the pipe. Many universities take advantage of it. This is really outstanding feature of the American regulation. In my opinion bitcoin by itself should be suitable for trading under the OTC BB rules in the USA. But currently it is impossible to detangle "bitcoin by itself" from the mass of false information about it. It is really hard to separate honest mistakes from intentional fraud. The underlying source code is complex and very little is known about its behaviour under stress. Courts and lawyers will have interesting tasks ahead of them. |
|
|
|
db.log is empty
This is most important. It means that none of the databases got corrupted, most importantly wallet.dat. ************************
UNKNOWN EXCEPTION
bitcoin in CMyApp::OnUnhandledException()
Now this I something I'm completely unfamiliar. Otherwise the log file looks pretty much like normal exit out of bitcoin. Since you running your Ubuntu behind some consumer-grade UPnP router, that would be the next item to investigate. Can you check if all the port mapping and NAT sessions are sensible? Or just reboot the router without investigating. It is a known cure for many networking problems with the cheap hardware. |
|
|
|
|
Maybe you could start by fishing the "strange errors" out of the "debug.log", "db.log" and maybe a scrollback buffer of your console?
|
|
|
|
So, 2112 how do you define money?
By the way, the link you posted by Nagle talks about GBLSE, which is a centralized stock exchange. I don't see how it is related to this discussion.
It doesn't matter how I define money. This is for the courts to decide. The link into the GLBSE thread is a very clear and important example that the definition of "security" under the criminal law is way much broader that under the contract law. And it is a beautiful example how the "we didn't issue a security" defense was useless in the court. The judges and jury will always focus on the inherent deception in the sales and promotional practices. In the Bowdoin case there were additional counts of mail and wire fraud which greatly strenghtened the prosecution case. Anyway, I'm not writing this to argue. The point is to provide the interested readers with somewhat competent perspective from non-lawyers on how to avoid getting snagged into an expensive litigation. I'm hoping that the inteligent readers will understand the arguments and do a further research on their own. When the large portion of the Bitcoin community is engaged in some form of crookery it is easy to get inadvertently caught in some "accesory to fraud" litigation. |
|
|
|
What underwriter? GLBSE doesn't have underwriters.
GLBSE is kind of an agglomeration of underwriter, sales agent and exchange. In the normal finance those functions would be spread amongst separate entities. GLBSE provides some of each, with the additional mix-in of possible conflits of interest. The issuer is the one responsible for the issue.
This is the typical situation. But sometimes a security that is issued and sold completely legally becomes a subject of litigation purely because of subsequent illegal sales practices over which they had no control. There's some case law about incorrectly issued ADRs for which the entire culpability was placed on the underwriter and sales agent in the USA, not on the original issuer. There were also cases related to reverse-buyout where the pre-buyout directors were found innocent and the entire culpability was placed on the post-buyout directors and executives. For over 10-years now I'm involved in a privately-run enterprise and no longer have any contact with the lawyers that I worked with in the past in the publicly-traded company. I thus apologise for not providing a proper legal citations. Its a complex and very interesting subject, where the quick one-liners are raraly a full answer. |
|
|
|
So, per the very definition of money, bitcoin is plainly just that, money.
You are quoting very good sources for the beginners. May I suggest that you dive a little bit deeper: follow the links I gave to the John Nagle's post, and follow his links to the appropriate legal sources. This back-and-forth is kind of pointless, leads to no greater understanding for anybody. With the beginner-level argumentation one can make a proof that cigarettes are money. And in fact they were in use as such in some prisons and internment camps. Ultimately, this kind of arguments for and against will end up as an evidence in some future trial about illegal promotional and sales activities of bitcoin. Quite obviously you are a beginner, for example don't meet the definition of "accredited investor". You also never have been involved in any securitues litigation, and probably not in any litigation whatsoever. Please treat this forum as an opportunity to learn the wider social context of this nice innovative whatchamacallit called bitcoin. Repeatedly arguing the same point over-and-over with somebody like me is just a waste of your time. You can spend it better on a real learning and education. Web is full of the resources to learn, but all the good ones require focus and dedication, they aren't a single page summaries like the ones you referred to. |
|
|
|
Is the idea not worth developing?
I think it is, but not at this stage in the Bitcoin lifetime. Lets wait one or two knees in the mining-lottery-ticket curve. I was thinking a bit about placing trust not in peers but in exchanges. Exchanges are the part of the ecosystem that has the most stable cash flow. Lets say you regularly deal with MtGox. It would mean that you are willing to listen to MtGox if they broadcast a block-chain checkpoint. Something like that. Because they have a stable(-st) cash they should have the resources to cross-validate competing branches of the block-chain. |
|
|
|
|
Show Posts
