Decentralized exchanges have their limitations. For instance, they can't properly interface with fiat money, which always requires a trusted third party within the banking system.

The current crop of DEX can also be more properly described as "non-custodial" -- they don't hold user funds but they use centralized architecture. Etherdelta had its DNS registry updated by attackers, who set up a phishing site on their domain and extracted user private keys. That's one example of how vulnerable this semi-decentralized model is.

Overall, though, I'm guessing decentralized exchanges will eventually play a very important role. When we can trade on a completely decentralized architecture with decentralized stablecoins, it'll be a real game changer as well. We're probably several years away from that point.

Did I walk into a time warp back to 2015? What a throwback! I thought this was a necro-post at first.

Get with the times, OP. The 1 MB block size limit is no more. It's now a 4 MB block weight limit, ever since Segwit was activated in 2017:


The limit will probably be increased further some years down the road, but there's really no urgency about it. I recently made a bunch of transactions where I paid only 1 satoshi/byte -- that's less than $0.01 per transaction. Worst comes to worst, people will just need to pay higher fees.

Judging by the facts that came out of the Coinbase/IRS case, not many people are paying their cryptocurrency-related taxes anyway. If they're not paying on capital gains, they definitely shouldn't be deducting for capital losses.

The allowed deduction is pretty minimal, anyway. Even though you can carry forward capital losses from year to year, the yearly limit on capital loss deductions is $3,000. So if you lost $3,000 or more in 2018, that's the most you can deduct from your taxable income for the year.

I think decentralized exchanges will eventually make small, illiquid altcoin markets like Cryptopia completely obsolete. Binance is incredibly profitable, but they obviously see the writing on the wall considering they are building a DEX.

However, I wonder about the limitations of decentralized order execution -- especially regarding network latency and unreliable orderbooks. It could take many years before DEX technologies can even remotely compete with centralized exchanges, which matters because real-time execution and order book depth are really important considerations for traders. All the current DEX exchanges are also built on centralized server and domain infrastructure, so they can be targeted by governments. Who knows when we'll see a DEX built on a decentralized DNS and mesh networking?

Not to mention, atomic swap capabilities are not built into most altcoins at this stage. It might be many years before we see a DEX emerge than can compete with centralized exchanges -- even altcoin-only exchanges.

It should only be a temporary freeze. Any blockchain analysis by an exchange that suggests money laundering, etc. could warrant that under their stated terms. The same might apply if coins are sent to Binance directly from darknet markets.

Here's the thing: I recall during the Wex.nz fiasco that Binance said they needed victims to file police reports regarding the stolen funds. (In their home countries? In Wex's registered country, Singapore? In Malta, where Binance is registered? Who knows.)

Absent that, they had no legal right to indefinitely freeze the assets. I have a hunch they froze funds temporarily in the direct aftermath but didn't end up really doing anything, since I never heard any news after that. This was back in October, I believe.


They're "wild west" banks -- much worse than regular banks. They're centralized, they confiscate/freeze funds..... and they are untrustworthy, opaque and secretive, hiding in jurisdictions where customers can't come after them.

What makes you say that? The CFTC approved LedgerX's physically settled markets in 2017. That those markets have been live and running smoothly for 1.5 years now only makes approval of Bakkt more likely.

People are talking about Bakkt like it's an ETF, but I don't see any indication that commodity futures markets are so difficult to launch. The recent delays probably have more to do with the US government shutdown than anything else. The entire process has grinded to a halt.

Uh, they obviously had value. They referenced $11 million in ERC-20 tokens. This is small potatoes for any exchange hack, but I'm sure the people sitting on millions of dollars in losses aren't thinking about that.


He did the same when customer funds at Wex.nz were sent en masse to Binance a few months back. Does anyone know what happened with that?

There's a big difference between Binance temporarily freezing funds and hacking victims getting their money back.

Here's my guess: $4,200

They're too small to have a drastic effect on prices. If a major exchange like Binance or Bitfinex had a big hack, it would be a different story. I remember the day Bitfinex got hacked in 2016 -- the price fell 20-30% within a day. By comparison, today's price just had a normal 3% fluctuation.

Sadly, many newer investors are asking for it, too. I've seen lots of posters around here praising regulators and hoping for cryptocurrency or token-specific regulations, either because they think it'll root out scams or pump their holdings. It just illustrates what the more mainstream investors and consumers looks like: "We're helpless idiots, please save us!" It plays right into governments' hands.

$1 million in daily volume gives you some idea -- it wasn't the biggest exchange.

With the altcoin market being so dead, I haven't touched tiny exchanges like that in over a year. I do have an account there and if this were 2017, I might have been affected.

Yet another example why you shouldn't store funds on exchanges. Just deposit, trade, and immediately withdraw.

That's the point. Money doesn't indicate any intrinsic value at all, whether we're talking about Bitcoin or dollar bills or cowry shells. Something is money only by virtue of the fact that people collectively imbue it with market value.


Sure. It's just not intrinsic. Intrinsic value regards nature -- innate characteristics, i.e. the elemental uses of gold in industry or the consumptive value of cocoa beans.

Money, as a medium of exchange, requires people to imbue it with value, by definition. That can't be intrinsic because it doesn't exist in nature.


Why is gold a good store of value? Isn't this directly tied to its value as money?

As far as being a medium of exchange, convertible gold-based notes worked just as well as fiat money. Gold-based money has a much longer history than fiat, actually.

What makes Bitcoin superior is that it's easy to transact with without using trusted intermediaries like banks.

These positions aren't at odds.

Bitcoin is transferable over a communications channel. It's portable, divisible and easy to store. This gives it better features than gold with regards to being a medium of exchange.

Decentralization is achieved through redundancy. This is inefficient and requires higher costs than any centralized scheme.

How do these statements contradict one another?


Lightning uses a P2P protocol based on distributed redundancy, just like Bitcoin. That you keep mischaracterizing it as "centralisation" is simply dishonest.

Anyway, LN network topology isn't relevant to the question of trustlessness, which is the goal of Bitcoin. If we can retain node decentralization in Bitcoin, I think using LN -- even if a more centralized topology emerges -- is the best available option.

If you had your way, a small minority of users (miners) would have blocked all users from ever having the choice to use trustless, off-chain alternatives like LN. You want to restrict the choices of all Bitcoin users and force everyone to transact on-chain. You're obviously not concerned about centralization of Bitcoin nodes, either.

It's not intrinsic, though. If nobody used Bitcoin, it would have no utility with regard to transactions. It would have no utility at all. So, it obviously has no intrinsic value, as Satoshi correctly pointed out. I just think this has no bearing on whether it can be "money."

Either way, it sounds like a temporary band-aid on an unsustainable system.

I don't see how this situation could arise anyway. Why should anyone regularly mine at a loss? Even if fee revenue is low, difficulty should always adjust downwards as unprofitable miners are pushed out of the market. There's always going to be entities with free electricity or spare CPU/GPU computing power, if it comes to it.

The bigger problem for central banks at that point is that Bitcoin would be worthless, based on the mining dynamics. The above situation exhibits extremely low demand for Bitcoin...

I've seen people say that intrinsic value is a requirement for money, but I think they're wrong. It seems that the only true requirement for money is that it's scarce. Without a sense of scarcity, all value becomes arbitrary. Past that, all that's really required is a general consensus among people to use something as money.

Satoshi made a great post about this many years ago.


I don't understand what this means. Turing completeness is just a statement about whether a system can perform any possible calculation or computer program. What does this have to do with money?

That sidesteps his point though, doesn't it? Even if you believe that on-chain scaling is superior to off-chain, we should be able to agree that Bitcoin's properties make a superior medium of exchange to gold.


Actually, it probably will. The more people that buy coins to lock them in LN channels, the more supply that is removed from the market. That should increase prices.

I don't think LN is ideal, but neither is Bitcoin's decentralized design. Decentralization is inefficient and costly. It's pretty amazing that LN could leverage Bitcoin's security while sidestepping those costs.


Altcoins that are cheaper to transact with have been available for many years now. If what you're saying were true, everyone would have exited exchanges with altcoins rather than bitcoins. Yet Bitcoin is still #1, in spite of its expensive costs. Why so?

This seems to be the important question. What tangible benefits can we expect from ensuring the viability of GPU or FPGA mining? At large scale, how different are the economics of general purpose vs. application-specific processors? In both cases, it seems like mining hardware production would still be fairly concentrated among a few giant producers. Also, if we look to historical pool hash rate distribution of, for example, Ethereum vs. Bitcoin, the former is not necessarily less concentrated. So, what's the goal here?

I think a lot of people look at the advent of ASICs in 2013 and the subsequent rise of Bitmain, and they automatically blame ASICs for how the industry developed. However, I think a lot of it has to do with industry consolidation and concentration of capital that you'd see develop in any growing market over time.

It's not surprising and it's not restricted to India. There's been lots of discussion about this among US users and traders going back several years. I personally know two people who have had bank accounts closed, presumably because of high volume activity linked to Bitcoin exchanges.

It even makes sense. Think about it: If you're a conservative bank and you have no idea how seriously Bitcoin exchanges are enforcing AML/KYC laws, would you want those customers moving loads of money through your bank? Probably not, because that opens you up to liability for AML violations. These days, banks are increasingly expected not only to know their customers, but to know their customers' customers.

It might bolster BTC's position relative to other altcoins, but I'm not even sure that's true. Even if it were, that doesn't mean it's a boon for BTC's price either. If anything, panic selling in altcoin markets usually translates to selling in the BTC market.

I also don't think the realities of the ETC attack have (or will) sink in for altcoin investors. POW coins have been getting 51% attacked for years and but they keep getting launched and people keep buying them. With altcoins, I've always thought it prudent to wait for payments to be very deep in the blockchain before accepting them as confirmed.