I have to say that I think the problem isn't that people are forced or not to issue credit. It's that they will feel obligated. Loans between friends and family can cause a lot of tension as it is. Adding a system such that formalize this system of spreading ill will is probably going to lead to a lot of pissed off people.

"You can lend to me, Ripple's going to keep me honest."

"It's been a whole extra year Bob, don't make me mark this debt as unfulfilled in Ripple."

If you connect lender and borrower through mutual social connections I can only imagine it getting messier.

A lot of why bitcoin works is that the permanence of transactions keeps people cautious. Reputation is everything and people who reliably maintain their end of the bargain get noticed and get business. When build credit into a system while putting forward social proximity as a substitute for demonstrated reliability people's already shitty risk calculations are going to take them to dark places.

Shame bankruptcy under the right conditions would keep them from being required to pay. Remember that bankruptcy laws made Donald Trump a very rich person.

Not really.

There are super computers, but generally they are already committed to tasks other than fucking with bitcoins and floating point operations also don't have much to do with SHA-2 hashing. There isn't a single supercomputer that could 51% the bitcoin network on its own. Traditional and even newer GPGPU supercomputers are actually a really shitty way to attempt a 51% attack.

Thanks, yeah it just never appeared in hotmail, but gmail worked.

Doesn't seem to be sending emails with login information...

Actually as a standard published by the US government, SHA-256 is in the public domain. All government publications released to the public automatically enter the public domain.

This seems like an interesting idea for allowing margins trading.

From what I understand double spending after even a single confirmation would be nearly impossible, but MtGox is willing to do big transactions. Transactions involving five digit numbers of bitcoins and six or more digit numbers of dollars. When transactions get that big, sometimes you have to take the safest route. Waiting for six confirmations may make things absurdly safe on their end, but they have to do it because any security failings can hit their reputation hard.

Not necessarily. If I had a five figure number of bitcoins and wanted to buy a house, I'd probably just sell enough of them in one big lump after perusing the bid orders. Even expecting the price of bitcoins to appreciate faster than the mortgage would build interest it would be worth it to not have the debt.

Overall the dump was a good thing for bitcoin. That five digit lump can now be spread around the bitcoin economy and the exchange took it like a champ rebounding like it did. It shows this isn't like the bubble that happened earlier in bitcoin history and with half of all bitcoins minted, some of these older lumps of bitcoin are going to drop into the market from time to time.What would be stupid is if everyone waits till bitcoin hits another record high to do so, because that would be catastrophic to bitcoin's long term health.

Ah, I just looked at the core count and assumed.

Pretty much you'd have to reprogram the FPGA, but that isn't a trivial task. Even for the simpler SHA-512 used by bitcoin it took time before people had it working.

Scrypt needs a lot of memory, but you can alleviate that to an extent by throwing more computing power at it. The thing is in the absence of the extra memory scrypt is designed to degrade in a way that becomes disproportionally more expensive in terms of additional calculations needed. You'd need a very expensive FPGA with a sizeable amount of onboard memory to make it worth even trying to begin programming it for scrypt.

A better bet if you are serious about LTC mining performance would probably be one of the new Tesla K20's from NVIDIA now that they've introduced an architecture to replace Fermi that doesn't suck for crypto. The problem is whether you are interested in committing $4,000 in hardware to LTC.

Thanks

I managed to install the client from the November 29th source package. Initially qmake and make failed so I managed to put it together with the following steps:

• Installed Qt-creator IDE
• Opened the project file in Qt-Creator and hit the run button
• Saw errors and installed libraries as Qt reported them missing
• Repeated this process for openSSL (libssl), Berkeley (libdb, libberkeleydb) and a metric ton of boost libraries.
• Some unnecessary libraries may be included in previous line
• Hit the run button again, got just under five thousand errors
• Hit the build button anyway
• Build returns only five errors and a functional binary that started downloading blocks as soon as I fired it up

This should work for anyone else running 64 bit Ubuntu 12.04 LTS.

It seems like an interesting alt-coin experiment.

You should be able to build it from source http://bitcoinarmory.com/index.php/building-armory-from-source

Just remember that it needs the reference bitcoin-qt client to do anything.

It's still the dominant cryptocurrency by orders of magnitude. A lot of the alt-coins have collapsed rather quickly (see solidcoin), but litecoin has persisted even if it still trades for a small fraction of what BTC does.

Actually, I think the intention of the mainline developers is to lower the transaction fee over time as the value of bitcoin rises relative to other currencies.

I think a great way to encourage people running nodes is by making it easy to connect a feature rich and network light mobile wallet to you own personal electrum-type server. That way your phone, tablet, or frequently disconnected because you drag it everywhere laptop can have an instantly available wallet without trusting third party servers.

The electrum server and stratum are freely available, but it would be nice if some of the bigger bitcoin news and information sites put together a guide for the mobile bitcoin user doing business in person.

Even the big schools on some level do offer something in the way of a compact environment. On most sufficiently large campus if you did engineering, that portion of your studies would be contained within the College of engineering for example. For your general education courses and such you would still be tossed into the sea of the larger university.

There are medium sized regional schools though that tend to have the strong engineering programs that liberal arts colleges generally lack. One example would be Missouri Science and Technology in Rolla, but the downside of studying there is being in Rolla Missouri. If you are looking at state schools though all but the most neglected campuses usually boast fairly solid engineering programs. Private universities in general tend to be a crapshoot for engineering though and even the large, selective, and well endowed can be shamed by mid tier states school in engineering if nothing else. Some of them might throw some serious money around trying to get you to enroll, and then you can take the money you saved and go to an prestige graduate program.

If you want to do computer science or computer engineering though a SLAC is a generally safe bet for the full course of study. I'm guessing you don't though because it seems all the prospective engineering always want to do mechanical with the exception for the three weirdos in every class who want to do civil out of the gate (the rest of the civil engineers come fleeing from the job market and apprenticeship requirements in architecture).

I'm skeptical of the 3+2 programs that involve a liberal arts college and another university's engineering school. It just seems way too much like the less expensive community college transfer option.

You mentioned Claremont McKenna, and if you apply there why not Harvey Mudd?

If anyone wants a low water mark for the APUs, I fired up some miners on my E350 powered laptop. 950KH/s on the CPU portion and 11.5MH/s on the GPU side. I didn't run it more than a couple minutes for obvious reasons. At least it can push Fallout 3 on Medium setting on the screen's full resolution using Wine on Ubuntu 12.04

At one point I do believe the bitcoin client allowed unsigned transactions aimed at IP addresses and those were removed from this list of things allowed fairly soon due to man-in-the-middle. I'm fairly sure anything that reintroduced unsigned transactions, much less forced them, would no longer be bitcoin.

I don't know about easier, but the initial outlay seems like it ought to be less expensive to get the first few batches done on a 90nm process at least.

It's important to remember that what made FPGA miners attainable was the dramatic savings the vendors got by ordering bulk chips that were already one or two generations old. I think it was a roughly 10x reduction in the per chip price going with the chips they did versus the newest chips. Once they had them, all it took was programing them to specialize in hashing.

When it comes to ASICs, like most things bitcoin I'll probably get into them a bit late. It will probably be three or four months after the first products have been confirmed to ship. I'm not afraid of them never shipping so much as I'm more comfortable letting people run some into the ground, before I commit to purchasing a single purchase chip. The FPGAs have reputations for lifespan as well as resale value outside of the bitcoin community. Whoever ships an ASIC first is shipping uniquely new hardware, and they are going to make their customers very happy as long as the hardware performs and holds up.