You know why I know you lost the argument. One word in your post.

Edit:


Again I dispute your claim that I am incorrect on the matter; however you have saved me time by collecting this material in one place and for this I do thank you.

Incorrect. You claim to win in the end when in fact the opposite is the case.

Edit: As demonstrated in your response here where you did nothing to support your claim:


Where for example did I say that Monero is secure against an attack with 1000x its hash rate? Or any POW coin for that matter?

I am not wasting any more time on you. What you see is what you get.

Quoted for posterity. I will deal with this later.

Edit:

If you do not want to waste your time may I suggest:
1) Stop proposing attacks that make zero economic sense
2) Stop insulting people.

The question in my mind is will this prevent the fork proposed by the devs of Ethereum?

The attacker in the 51% case is mining otherwise perfectly valid blocks but at a higher attack difficulty rather than the correct optimal difficulty in the blocks. There is no actual change in the protocol. The trade-off is that the 51% attack blocks could be easily detected by the network in exchange for a much lower cost of attack. In either case once the attack ends the difficulty starts to fall since time ticks by with no blocks being mined. There is a key difference between Bitcoin or a Bitcoin clone and Monero in that in Bitcoin the difficulty adjustment is discrete approximately every 2 weeks while in Monero it is continuous, This means that by timing the end of the attack correctly the attacker could keep the difficulty constant in Bitcoin for 2 weeks, while in Monero the difficulty starts to fall approximately 2 min after the end of the attack.

http://www.coindesk.com/data/bitcoin-mining-difficulty-time/
http://www.coinwarz.com/difficulty-charts/monero-difficulty-chart

Collusion between honest miners and devs could mitigate the attack but in both cases it involves ignoring the difficulty in the blocks in some way, effectively a hard fork. One can mitigate against this attack by having the difficulty algorithm drop the difficulty at a faster rate if no blocks are found after a given amount of time, where this time is statistically much larger than the normal block time.

Still I do not see the economics here. One can do a lot more damage with a 51% attack than this.

So does the 100x attack since the attacker has to maintain the attack for a period longer than the number to blocks used by the difficulty algorithm to calculate the change in difficulty. So my point is why would the attacker not just fork the coin and set the difficulty to some insane high value?

Edit: I am still waiting for the answer to: For how many blocks do you want to maintain the 100x attack for?

In the adaptive blocksize limit case as in this case the attacks you formulated required a greater hashpower than the hashpower required to mount a 51% attack on the coin; however I must say than in the blocksize limit case the attack you proposed did not require 100x the hashpower of the coin. Can the difficulty adjustment algorithm in Monero be improved? Of course it can but that does not give any validity to an attack requiring 100x the hashpower of the coin since with a small fraction of such hashpower one can fork the coin and replace the difficulty adjustment algorithm by one that is completely broken by design.

As for our differences on the relative merits of proprietary vs free libre open source software development or between copyleft and non copyleft FLOSS licenses it is best to simply agree to disagree.

Edit: In 2001 Steve Ballmer the then CEO of Microsoft called Linux a cancer because of the Copyleft GPL 2 license. http://www.theregister.co.uk/2001/06/02/ballmer_linux_is_a_cancer/ It is fair to say that 15 years later this slow growing cancer is having an impact.

I do not want sound sarcastic, but this is starting to sound to me as attempting to find the most inefficient way to attack a POW coin.

For how long do you want to keep up your 99.9% attack for? Like for how many blocks?

Incorrect. This is mounting a 51% attack for approximately one day in order to double the blocktime for approximately two days. I do find it interesting though when people suggest attacks that are way more expensive than a 51% attack and have relatively insignificant impact.

If the proof of work algorithms are different there is nothing new here. Manipulating the difficulty in the manner suggested in the article is a horribly inefficient way to attack a POW coin, if the objective is to create havoc in the coin, since it requires much more hashpower than a simple 51% attack. The simple economics of this is that the greater the market price of a coin, the more costly such a brute force attack would be and consequently the least the chance of such an attack. There is really nothing new here. As for bear raids again the more liquid an asset is the less likely a bear raid will succeed.

If the proof of work algorithms are the same then there can be little or no cost in the ASIC case I mentioned above, because the equipment is surplus and obsolete and there is no other use for it. This is why we have seen this phenomenon in some small SHA 256 coins not because the coins were specifically targeted for an attack but rather because people were trying to extract additional value out of obsolete ASIC Bitcoin mining equipment.

Edit 1:  Many in the early days of Bitcoin were afraid that some large bank would attempt such an attack on Bitcoin in order to "nip the competition in the bud", it did not happen. My take is this did not happen because the large players rarely take much notice of the small players until it is too late. If Monero were to grow large enough to pose a serious threat to Bitcoin, and that will take way more than a 1 billion USD market cap, then it will be also way more difficult to attack.
Edit 2: The DAO and MTGox were attacked because there were very weak and full of vulnerabilities. Ethereum could easily go down with the DAO, because the Ethereum devs panicked under pressure and are inducing the community to panic. They should simple do nothing  and let the DAO fail rather put the entire credibility of Ethereum at risk. Bitcoin did not fork over the MTGox troubles and that is why Bitcoin is still around. As for Monero it could not fork to pull these kinds of shenanigans even it it wanted to, because one would not know which transactions to censor.

This difficulty attack requires a much higher percentage of the hashpower than a 51% attack.

The article does make a valid point. If a coin uses the same mining algorithm as an "ASIC Goliath coin" that has gone through more than one generation of ASICs, then it can be attacked using the now obsolete first generation ASICs from "ASIC Goliath coin" that would be otherwise worthless junk. The example of Dogecoin in the article would only work if Litecoin goes through more than one generation of ASICs and Dogecoin were not merged mined with Litecoin. Extrapolating  this attack to Monero fails because:
1) Monero uses CryptoNight not SHA 256, Ethash, Scrypt etc.
3) CryptoNight is ASIC resistant. There are no ASICs for CryptoNight let alone multiple generations of ASICs
2) Monero has by far the greatest hashrate of all the CryptoNight coins.

There is a critical distinction between been able to hard fork a coin and been able to use a hard fork in order to freeze funds. The real threat here is the ability to selectively freeze funds and / or reverse transactions since this is what happened in Ethereum with the DAO. To do this a developer or core team etc needs two things:
1) They must convince the community (miners, users etc) to go along with the fork. It is one thing to propose a hard fork that is clearly beneficial to the community. It is a very different matter to act as a Judge Jury and Executioner, by freezing funds and / or reversing transactions. The jury is still out on this in the case of Ethereum. Will the Ethereum community approve the fork?
2) They must know which funds to freeze and which transactions to reverse. It is here where Monero and Dash (In the case of Dash only, If the funds are mixed) provide a critical second level of protection, since this information is simply not available to anyone due to the design of the coin. This is a very important benefit of private, anonymous and fungible coins.

Time will tell, and I am also unsure if they will get the hashrate. They will need a percentage of the hashrate comparable to that of Namecoin in order to be secure. otherwise they could get attacked for the reasons you cited above. Of course if it fails or even if it works then one can learn from it and build something better.

Edit: For starters they also need  a coin that can scale as the main chain for this to work and Bitcoin does not meet this requirement.

It is not that simple. Namecoin has 35% of Bitcoin's hashrate and generates only 0.063% of the miner reward revenue. 1 NMC = 0.00063 XBT. This is actually comparable to 10% of the current Bitcoin fee revenue as a percentage of the overall miner reward. Furthermore, My take is that in the case of a pegged sidechain there is actually an even lower cost to merge mine because one does not have to sell a different coin in an exchange. For many miners the hassle of exchanging the NMC for XBT is likely not worth the trouble; however in the case of a sidechain with a two way peg this exchange issue is avoided. In the case of a pegged sidechain say generating the say even as low as 0.5% of the main chain revenue one could expect and even greater proportion of the main chain's hashrate. Yes one would need 17.5 of the Bitcoin hashrate to launch a 51% attack against Namecoin; however for a Bitcoin pool such an attack has potentially a much higher downside, in the negative public relation such an attack could generate alone, than the comparably minuscule reward such an attack could produce..

This issue of double spend attacks could for example be dealt with by treating transactions between chains as coinbase transactions (120 blocks rather than 6 blocks).

I do agree that is requires a fair amount further research, but I am not prepared to simply dismiss it based on just the comments in the articles. Namecoin does provide some very compelling evidence over many years that merged mining can work even when there is a marked difference in the revenue generated by the main chain and the merged mined alt-coin or side chain.

Actually the long term trend in the market capitalization chart of Monero says that funds are moving into Monero rather than out of Monero. https://coinmarketcap.com/currencies/monero/#charts.

I suspect the hard drive. I have no problems running Monero on a computer with a rotary hard drive provided it is one of the larger ones. I run Monero on an 8 year old computer with a 2 TB rotary hard drive without any problems. I suspect one can go down to 1 TB or even 500 GB on a rotary hard drive and still be ok; however I have not tested this.

A side chain is the proper way to do this, and it can be merged mined with the main chain. Scaling as we both know is a very serious issue so one has to start with a base coin that can scale on the main chain properly even before adding smart contracts. Then address the scaling issues of the smart contracts themselves.