Cryptocurrencies

[Iskaral Pust]

5 hours ago, Morin said:

As a small sidenote: The Enigma was never really considered to be uncrackable. German mathematicians knew very well how it could be attacked. This is bit of myth, caused by the immense difficulties the Allied forces had in decyphering Enigma encoded messages.

Enigma was sufficiently difficult to crack with available computing power and the number and intended length of messages.    It was a balance between code strength and practical requirements to have it distributed throughout the navy.  Enigma was a system more than just a code: underlying code, machines for encoding/decoding, one-time day pads, etc.  Even when the underlying code was known (the Allies seized an Enigma machine in May 1941) they still struggled to decrypt many days of messages.  It was the sloppiness of longer messages and especially repeated phrases (“Heil Hitler”) that reduced the computational permutations that allowed Bletchley Park to routinely decrypt the daily volume of messages using a custom-built decryption computer.   

[Yukle]

On 1/15/2018 at 0:28 PM, Iskaral Pust said:

Enigma was sufficiently difficult to crack with available computing power and the number and intended length of messages.    It was a balance between code strength and practical requirements to have it distributed throughout the navy.  Enigma was a system more than just a code: underlying code, machines for encoding/decoding, one-time day pads, etc.  Even when the underlying code was known (the Allies seized an Enigma machine in May 1941) they still struggled to decrypt many days of messages.  It was the sloppiness of longer messages and especially repeated phrases (“Heil Hitler”) that reduced the computational permutations that allowed Bletchley Park to routinely decrypt the daily volume of messages using a custom-built decryption computer.   

That's accurately summarised.

Furthermore, the Allies continued to have difficulties with naval encryption. German naval commanders weren't convinced that their codes were secure and wondered how the Allies just to happened to hit *only* the most crucial targets. Were they compromised? Were there too many spies slipping through? It makes sense: when you know your enemy's secrets, you never let them know, so perhaps the Allies were selectively playing their hands while always knowing just what was going on.

Just to be sure, though, naval commanders increased the number of rotors that the naval messengers had to use. Even this small change added so many more permutations that Allies couldn't effectively break naval codes in most cases within the timeframe of any information being relevant.

They still managed it now and again, but it's more realistic to think of cracking the Enigma as the Allies picking intercepted messages more or less at random and then devoting their expensive hardware at cracking it in the hopes that they'd have something useful.

One of the side-effects of always looking for key phrases such as "weather report" was that, while they managed to often crack codes with a short cut, they also ended up with a lot of decrypted meaningless weather reports before being able to make headway into the important stuff. And, of course, they had no idea what was important and what wasn't until after they'd decrypted it. A whole day of successful code-breaking could end with nothing much useful learned.

[maarsen]

7 hours ago, Yukle said:

That's accurately summarised.

Furthermore, the Allies continued to have difficulties with naval encryption. German naval commanders weren't convinced that their codes were secure and wondered how the Allies just to happened to hit *only* the most crucial targets. Were they compromised? Were there too many spies slipping through? It makes sense: when you know your enemy's secrets, you never let them know, so perhaps the Allies were selectively playing their hands while always knowing just what was going on.

Just to be sure, though, naval commanders increased the number of rotors that the naval messengers had to use. Even this small change added so many more permutations that Allies couldn't effectively break naval codes in most cases within the timeframe of any information being relevant.

They still managed it now and again, but it's more realistic to think of cracking the Enigma as the Allies picking intercepted messages more or less at random and then devoting their expensive hardware at cracking it in the hopes that they'd have something useful.

One of the side-effects of always looking for key phrases such as "weather report" was that, while they managed to often crack codes with a short cut, they also ended up with a lot of decrypted meaningless weather reports before being able to make headway into the important stuff. And, of course, they had no idea what was important and what wasn't until after they'd decrypted it. A whole day of successful code-breaking could end with nothing much useful learned.

Shark, the code name for German Enigma messages, was cracked in 1942 and so completely, that all messages could be read. The German navy used a fourth rotor to make their encryption much harder to read. It was still cracked. Code books giving the detail of the next days rotor settings, taken from weather forecasting boats saw to that.

The British had an Enigma machine, courtesy of the Poles, after the fall of France. Using that, and deduction, they had built their own machine. The rotors used and the position of the letters on the rotors were the complication, not having a machine. 

[Morin]

2 hours ago, maarsen said:

Shark, the code name for German Enigma messages, was cracked in 1942 and so completely, that all messages could be read. The German navy used a fourth rotor to make their encryption much harder to read. It was still cracked. Code books giving the detail of the next days rotor settings, taken from weather forecasting boats saw to that.

The British had an Enigma machine, courtesy of the Poles, after the fall of France. Using that, and deduction, they had built their own machine. The rotors used and the position of the letters on the rotors were the complication, not having a machine. 

The inventors of the Enigma already knew or at least suspected that their configurations weren't secure. The original patent even calls for ten rotors.

The attack used on the Enigma, which we today would call a known-plaintext attack, where you would leverage knowledge of what the encryption of certain plaintexts looked like to decrypt further ciphertexts, was however considered impractical by the german military. Examples like weather reports and repeated phrases ("Heil Hitler") proved that notion wrong.

The British break was in many ways inefficient. The Germans had themselves come up with a much better idea to attack their own system during the war. I think it's really interesting, that the huge advances in computer science that accompanied the Allied decryption efforts, were due to a somewhat "bad" solution to a problem.

[maarsen]

2 minutes ago, Morin said:

The inventors of the Enigma already knew or at least suspected that their configurations weren't secure. The original patent even calls for ten rotors.

The attack used on the Enigma, which we today would call a known-plaintext attack, where you would leverage knowledge of what the encryption of certain plaintexts looked like to decrypt further ciphertexts, was however considered impractical by the german military. Examples like weather reports and repeated phrases ("Heil Hitler") proved that notion wrong.

The British break was in many ways inefficient. The Germans had themselves come up with a much better idea to attack their own system during the war. I think it's really interesting, that the huge advances in computer science that accompanied the Allied decryption efforts, were due to a somewhat "bad" solution to a problem.

the British break was based on work done by the Poles even before the war started. There were a large selection of rotors to choose from, but 3 were installed, 4 for the navy. if the British were going down a sub-optimal path, they were led that way by the Poles. The idea of the Enigma is really quite good, if done correctly. The American armed forces used a modified version of a coding machine based on the Enigma, up till the mid 70's, for battlefield coding and decoding. Microchips started to become cheap and reliable about that time and creating a virtual Enigma type machine using software becomes viable then. 10 rotors makes for an extremely cumbersome package in a mechanical machine but could be done much more easily using a virtual machine.

The original point of all this is that encryption can be broken due to operator error and/or technological advances. Using elliptical functions or factoring multiples of very large prime numbers is based on the fact that both are really hard to calculate even with lots of computing power. A new idea or theorem involving elliptical functions, such as using them to solve Fermat's last theorem, can lead to a breakthrough in codebreaking. Finding if a large number is prime is also really hard, as is finding prime factors of a large number. Even so, if a proof to the Riemann hypothesis is found, proving primality may become trivial and there goes public key cryptography.

Now we have the revelation that Intel and AMD cpu chips have a pair of fatal flaws that allow someone to bypass the encryption entirely and read passwords directly. None of this makes me feel really secure about bitcoin and encryption. If I am aware of all this stuff, I am sure the NSA is even more aware and has been for much longer.

[Morin]

56 minutes ago, maarsen said:

the British break was based on work done by the Poles even before the war started. There were a large selection of rotors to choose from, but 3 were installed, 4 for the navy. if the British were going down a sub-optimal path, they were led that way by the Poles. The idea of the Enigma is really quite good, if done correctly. The American armed forces used a modified version of a coding machine based on the Enigma, up till the mid 70's, for battlefield coding and decoding. Microchips started to become cheap and reliable about that time and creating a virtual Enigma type machine using software becomes viable then. 10 rotors makes for an extremely cumbersome package in a mechanical machine but could be done much more easily using a virtual machine.

The original point of all this is that encryption can be broken due to operator error and/or technological advances. Using elliptical functions or factoring multiples of very large prime numbers is based on the fact that both are really hard to calculate even with lots of computing power. A new idea or theorem involving elliptical functions, such as using them to solve Fermat's last theorem, can lead to a breakthrough in codebreaking. Finding if a large number is prime is also really hard, as is finding prime factors of a large number. Even so, if a proof to the Riemann hypothesis is found, proving primality may become trivial and there goes public key cryptography.

Now we have the revelation that Intel and AMD cpu chips have a pair of fatal flaws that allow someone to bypass the encryption entirely and read passwords directly. None of this makes me feel really secure about bitcoin and encryption. If I am aware of all this stuff, I am sure the NSA is even more aware and has been for much longer.

Primality testing is really not hard, particularly compared to factoring and the Riemann hypothesis has pretty much no impact on public key cryptography.  You are of course right, that we might see a major algorithmic breakthrough that invalidates a portion of cryptography.

But this is not really a problem of just bitcoin. Large portions of the modern digitalized banking system also use cryptographic measures.

I think there are many issues with bitcoin, but untrustwortiness of the the cryptography is really not one of them. If you could break the security of bitcoin, online banking would be a much bigger (and potentially easier) target.

[Yukle]

On 1/21/2018 at 4:06 AM, Morin said:

If you could break the security of bitcoin, online banking would be a much bigger (and potentially easier) target.

This I disagree with, since any bank has security features to prevent these problems running away. A bank can freeze its trading, alter its security and then resume in a limited form. It can recall suspect transactions and pass on suspicious evidence to police. Its naive to think that banks aren't prepared for massive breaches. That's not to say that they don't happen, but think about it: how frequently do you think they are being attacked? Probably continuously. If you are a big bank then there is never a time when you're not under attack. And yet they seem to hold in almost all cases, and when they don't, they still have contingencies to fall back on.

[Morin]

10 hours ago, Yukle said:

This I disagree with, since any bank has security features to prevent these problems running away. A bank can freeze its trading, alter its security and then resume in a limited form. It can recall suspect transactions and pass on suspicious evidence to police. Its naive to think that banks aren't prepared for massive breaches. That's not to say that they don't happen, but think about it: how frequently do you think they are being attacked? Probably continuously. If you are a big bank then there is never a time when you're not under attack. And yet they seem to hold in almost all cases, and when they don't, they still have contingencies to fall back on.

I do not know how frequent attacks in general on banks are. I am not aware of a single cryptographic attack that was ever carried out against a bank. There really haven't been any major attacks on cryptographic protocols in the last decades. Given how prevalent hash functions and digital signatures are in It security I am not sure that there is really a good contingency plan for something this major.

In my personal experience, every IT security department I have ever talked to essentially said a sudden cryptographic failure would a catastrophe of such scale, they consider it far out of scope to plan for it. 

[Iskaral Pust]

Coincheck announced today that it was hacked and lost $400m in bitcoin.  They have the dubious honor of now being possibly the largest single bitcoin theft.  Bitcoin fell as low as $9,200 (half of it's peak value just six weeks ago), before rebounding to $9,900.  That's according to the prices on Bloomberg.

I'm starting to feel guilty because in late December, some friends* of ours asked me about their soon-to-be son-in-law's bitcoin investment worth ~$30,000 (who can tell if that was accurate considering how much the price was moving).  After I explained it, they were shocked and wanted to know if they should pressure him to sell them now (to protect their daughter's financial interest).  I said it was a speculative bubble and very likely to lose a lot of money, but as a parent I wouldn't try to direct a soon-to-be son-in-law what to do with his money.  I still think it's right for them to not get involved, but I don't want to get blamed for his losses. 

*Not the friends who asked around the same time if they should sell all their stocks in their 401k and invest in bitcoin instead.  Stocks are up ~10% since then.

[Mexal]

Bitcoin is actually up to 11.4k and it's starting to break some trend lines which may lead to a bullish sentiment. Will be interesting to see if there is some volume behind it. If so, target is probably around 13-14k.

To your point above, I agree with your advice to tell your friends to stay out of it. I'd be really annoyed if someone told me what I should do with my money if I didn't solicit the advice, especially if I didn't feel they had any understanding of the market. You're right, it's a bubble, but to lose everything, the value would essentially have to go to zero and the likelihood of that happening this year is almost zero. So if he can take his 30k investment, make some return on it, which is likely greater than anything he'd make in the stock market right now, then get out, it's still a good trade.

As an aside, this isn't abnormal in Crypto. In Jan & March 2017, it dropped significantly in value before rebounding. From June to August 2017, it did the same thing. In Sept, it did the same thing before going on a run to 20k in December. We'll see what happens but there is definite money to be made here and if your friend's son in law is smart about it, he can do quite well. With Robinhood now adding crypto to it's trading app and 500k+ applications of people waiting to get access, there will be another influx of cash. And really, in August the price got as low as 1,800. The fact that it got up to 20k is one thing but it being stuck in the 10-11k range is still 6x return if you bought in August so while it did drop 50% in value from Dec - Jan, it still went up 600% from August to now. Buying in at the top (17-19k), just like the stock market, is a good way to lose money.

Everything I have in crypto is profit and it's all money I can afford to lose. I'm under no illusion that BTC itself is kind of meh and that at some point, this might completely blow up. At the same time, I bought into projects that I believe in, ones that have shown a strong track record for communication, innovation and delivery and ones that I think have a legitimate future (Stratis, POWR, Aion, ETH). I have very little actual BTC and what I do have is essentially there to buy short term trades at the bottom before a catalyst, make a quick 20-30% and get out to buy other projects I believe in. I'll bank profits every month or two and that way I'll be protected against a complete crash. Not sure why people wouldn't want to do that.

[Shryke]

This remains the best thing to come out of this environmental murdering cryptocurrency bullshit:

http://www.techweez.com/2018/01/26/ponzicoin-crypto-scheme/

Quote

Back in 2014, a crafty developer created a cryptocurrency and aptly named it PonziCoin. The idea was that it was to operate like a proper Pyramid scheme, where early investors would need to invite more people to invest in this coin in order for them to make money. Well, things didn’t go as planned as he waited for people to invest in the said cryptocurrency and he made off with approximately $7000, which is around $2.2M in today’s value, without giving any payouts to anyone.

Quote

 

San Francisco based developer, Rishab Hegde, “jokingly” built a cryptocurrency based on Ethereum and named it PonziCoin – an exact copycat of what happened back in 2014.

Rishab Hedge went ahead to warn the investors on the coin that it was a Ponzi Scheme, “The world’s first legitimate Ponzi scheme,” reads the coin’s landing page.

 

Quote

 

Now here’s the major news; People actually invested in the Ponzi scheme. Maybe with the hopes of being the early investors and cashing out before everything collapses. After around 8 hours, PonziCoin had attracted attention and the platform had collected around 250 Ether coins ( valued at more than $25,000).

Mr Rishab seems to have gotten cold feet due to the attention and he decided to pull the plug on PonziCoin, leaving investors out it the cold and possibly making away with their money since none of the investors got a payout on their investment

 

People literally fell for a scam openly describing itself as a ponzi scheme. Twice.

[DireWolfSpirit]

On December 22, 2017 at 9:20 PM, Ghjhero said:

I read an article in the BCC a few days ago that mentioned with as volatile as the price of Bitcoin is, it is hardly worth reporting on the price as that isn't anywhere near as important as other aspects of BTC such as the ever increasing transaction fees. Two days ago I moved the roughly $60 worth of BTC I have off of Coinbase and into a wallet only to discover it now takes me more than half my BTC to move it should I want to send it to an address in the future. So I'm kinda screwed and don't really have much choice but to let it sit there until this mess is given the attention it needs.

Thought this was interesting.

[Yukle]

I've seen more and more in the news that cryptocurrencies can't be easily taxed (if at all). I'm really not keen on a method that allows yet more hoarding of wealth into the hands of the very few. I would much prefer that retailers simply rejected taking them as currency.