Critique of the blockchain contra blockchain

by on December 27, 2017 at 12:45 am in Economics, Web/Tech | Permalink

Here is an excellent Kai Stinchcombe essay, piling together many of the extant criticisms of blockchains, here is one excerpt:

There are four additional problems with a blockchain-driven approach. First, you’re relying on single-point encryption — your own private keys — rather than a more sophisticated system that might involve two-factor authorization, intrusion detection, volume limits, firewalls, remote IP tracking, and the ability to disconnect the system in an emergency. Second, price tradeoffs are entirely implausible — the bitcoin blockchain has consumed almost a billion dollars worth of electricity to hash an amount of data equivalent to about a sixth of what I get for my ten dollar a month dropbox subscription. Fourth, systematically choosing where and how much to replicate data is an advantage in the long run — the blockchain’s defaults on data replication just aren’t that smart. And finally, Dropbox and Box.com and Google and Microsoft and Apple and Amazon and everyone else provide a set of valuable other features that you don’t actually want to go develop on your own. Analogous to Visa, the problem isn’t storing data, it’s managing permissions, un-sharing what you shared before, getting an easy-to-view document history, syncing it on multiple devices, and so on.

Overstated, in my view, but worth a ponder nonetheless.  How many years does blockchain get before we start being unimpressed?  Ten?  Thirty?

1 clockwork_prior December 27, 2017 at 1:01 am

Without getting too detailed, ‘your own private keys’ and ‘intrusion detection, volume limits, firewalls, remote IP tracking, and the ability to disconnect the system in an emergency’ are two different things, and not really linked. Especially firewalls, which have become ubiquitous, regardless of what thinks of their efficacy at the lowest common denominator level.

Overstated is not the problem with that passage.

Reply

2 ʕ•ᴥ•ʔ December 27, 2017 at 8:57 am

They divide in two classes. Self-management with private keys, for all but the most elite and hyperactive users, precludes intrusion detection, volume limits, firewalls, remote IP tracking, and the ability to disconnect the system in an emergency.

On the other hand, we expect Fidelity to do that for us, as well as to qualify for Federal insurance (SPIC).

Reply

3 ladderff December 27, 2017 at 1:07 pm

Your repeated comments about SPIC and FDIC suggest that you think global financial arrangements are more or less sound, rather than completely insane and terminally morbid.

There are two problems with this view. 1) While it puts you in fine company, it’s wrong. 2) Whether it’s wrong or right, holding it means you deny that the real motivation for bitcoin is valid—which may explain why so many people are not sufficiently motivated to avoid getting the facts wrong. (e.g. someone on here had to be told that bitcoin storage is basically free and easy).

The journalistic/academic commentary on this phenomenon continues to be awful on average. The linked essay is a great example, replete with “gotchas!” and also written with the above-discussed view baked in to all its assertions about bitcoin—along with an evident desire to score pointless points against some libertarian golem. Anyone with a decent ear should read it and judge it incredible. That Cowen could refer to it as merely ‘overstated’ rather than glib and eristic tells us that his interest in this issue is also quite shallow.

Reply

4 ʕ•ᴥ•ʔ December 27, 2017 at 1:42 pm

In other news, I don’t have food or guns in my bunker.

Reply

5 ladderff December 27, 2017 at 3:29 pm

As you like. The vast majority of bitcoin developers and bitcoin users likewise have no food or guns in their bunkers. Meanwhile, you can see, maybe, what this attitude of yours does to your credibility: bitcoin proponents have no trouble conceding that contingent on the view that there’s nothing particularly the matter with the status quo, bitcoin is a waste of time and effort. You and the essayist and others, on the other hand, are too busy being smug to say anything accurate about it. With contingent thinking thus out of reach, we get point-missing asides about FDIC.

6 msgkings December 27, 2017 at 3:33 pm

His point is, if bitcoin is just another doomsday prepper technique, it probably doesn’t deserve the current valuation. A doomer will of course always reply, that’s what you think. And that’s the end of the discussion. Either you think you need to prepare for that kind of thing, or you don’t.

7 MMK December 27, 2017 at 2:49 pm

I don’t even know what two-factor authorization is and I’m a security engineer. I assume the author meant two-factor authentication.

Reply

8 Matt December 28, 2017 at 9:03 am

Yeah, I was a little curious about that. It really sounds like a security observer throwing buzzwords at the wall without really understanding them.

Reply

9 msgkings December 27, 2017 at 2:15 am

Yes, excellent article. Maybe bitcoin is the cold fusion of our time?

Reply

10 msgkings December 27, 2017 at 2:16 am

Or more aptly, blockchain is the cold fusion of our time

Reply

11 Brian (Moomin) December 27, 2017 at 3:13 am

Disclosure: I’m a “true believer” and too much of my wealth is in various cryptoassets, around 50%.

This article suffers from a severe lack of imagination and makes some claims which could be easily disproven by reading _any_ relevant literature. I’ll admit that if it takes 10 more years for blockchains to change anything in the world we can call the experiment a failure but there are plenty of cool applications just on the horizon, maybe 2 years from fruition.

I’ll start with the part I agree with: Blockchains are /not/ consumer-friendly. Transaction reversibility, something Bitcoin doesn’t natively provide, is indeed _the_ feature of PayPal. Consumers also certainly prefer airline miles to whatever libertarians mean by “sound money”. But consumers aren’t the only thing which matter in this world? Let’s take the example where this article fails hardest: storage-on-the-blockchain. Tyler has conveniently already quoted it for me:

> Second, price tradeoffs are entirely implausible — the bitcoin blockchain has consumed almost a billion dollars worth of electricity to hash an amount of data equivalent to about a sixth of what I get for my ten dollar a month dropbox subscription.

Starting here because it’s easiest. Nobody who knows what they’re talking about is dumb enough to suggest that storing bulk data directly on a blockchain is a good idea. Blockchains are replicated to every participant in the network. Current technology is kind of primitive and requires this but because of the costs you should only give this treatment to data which absolutely _needs_ to be consensed upon.

Before you write an article bashing blockchains, read the Filecoin whitepaper? Filecoin puts hashes (short unforgeable identifiers) of documents into a blockchain and sets up storage and retrieval markets. The actual data is _not_ stored in any blockchain, that would be dumb. Instead Filecoin creates markets you can use to ask to have files stored and to retrieve files. Miners place bids on your asks and you select the one you want to do business with. The bockchain stores just enough data to verify everybody stays honest and actually stores the files and actually retrieves the correct file and actually pays for their usage.

The promise is massively reduced storage costs. There are a bunch of datacenters out there with a bunch of servers with half-full hard drives. Those hard drives are underutilized because of high transaction costs. If I want to store some files it would cost a lot for me to go looking for people with underutilized servers and draw up contracts which take into account how available those servers are. In practice we all just use Amazon S3. Filecoin, if it works, will create a marketplace which allows anybody in the world to offer some space and anybody in the world to ask to use that space. This will lead to a _much_ more efficient outcome, which I bet leads to an order of magnitude reduction in data storage prices.

> First, you’re relying on single-point encryption — your own private keys — rather than a more sophisticated system that might involve two-factor authorization, intrusion detection, volume limits, firewalls, remote IP tracking, and the ability to disconnect the system in an emergency.

> And finally, Dropbox and Box.com and Google and Microsoft and Apple and Amazon and everyone else provide a set of valuable other features that you don’t actually want to go develop on your own. Analogous to Visa, the problem isn’t storing data, it’s managing permissions, un-sharing what you shared before, getting an easy-to-view document history, syncing it on multiple devices, and so on.

These next two points go back to what I said earlier: anyone claiming that bockchain is currently a good solution for consumer-facing applications is an idiot. It might be some day, but for now it’s in more of the infrastructure “make core components more efficient” phase. Amazon S3 does not provide “permissions, un-sharing what you shared before, getting an easy-to-view document history, syncing it on multiple devices”. But guess where all those files in Dropbox were actually stored until 2015? Dropbox was built on top of this unfriendly service. Most of Kai’s other critiques of blockchains fail for the same reason. No, consumers will not like these sharp edges. Businesses are smart and can wrap these new technologies but that takes time, probably over 10 years.

Filecoin might fail (protocols are more difficult than startups so chances are Filecoin will fail). Many of the other projects I’m watching might fail as well, but if they do they’ll fail for reasons Kai Stinchcombe is entirely ignorant of. The engineers excited about blockchains aren’t idiots, and aren’t making trivial mistakes which outsiders can imagine by just sitting in a chair.

While I’m here, here’s my case for cryptothings in general:

Bitcoin was an existence proof that untrusted shared state is possible. If you had described some of the properties of Bitcoin to any computer scientist in 2008 they would have claimed it was impossible. A bunch of people who don’t trust each other coming to agreement over some set of values? (They call it the Byzantine Generals Problem) There’s some very nice math which proves that it’s impossible. But… Bitcoin has shown that if you infuse some economics into the problem… this is actually possible! It’s a true innovation, and a real, tangible step function in the capabilities of humanity. Now we just have to figure out how much that innovation is really good for. If there’re no more applications in 10 years then I guess it didn’t win us very much. But it has the promise of lowering all kinds of transaction costs.

You can now perform some limited kinds of escrow with anybody in the world without needing a trusted intermediary. You can standardize more kinds of contracts and put them onto global markets. You can bring arbitrary amounts of money from Syria to London simply by remembering a few words. It’s all very exciting 🙂

Reply

12 Axa December 27, 2017 at 7:55 am

Storage costs? In the engineering company where I work the cost of hard drives, tapes and electricity is minimal when compared to IT department salaries. The goal is to have a 20 year backup and people that knows how to fix things when shit hits the fan. Reducing storage costs with blockchains is a solution without a problem.

2 years is 8 quarters, 10 years 120 months. Economic cycles are about 10 years. If you invest with 10 year horizon there’s a lot of other possibilities. Blockchain is just a dish in the menu.

Reply

13 Anonymous December 27, 2017 at 10:48 pm

This is one of the best comments on here in a while. Can disagree with parts of it but thanks for posting!

Reply

14 Anon December 28, 2017 at 7:57 pm

Curious which parts you disagree with

Reply

15 carlospln December 27, 2017 at 3:52 am

“How many years does blockchain get before we start being unimpressed?”

A: the square root of the interval of time during which you possessed a perma-erection regarding 3-D printing.

Reply

16 clockwork_prior December 27, 2017 at 4:39 am

The amusing thing is that 3D is actually starting to become integrated into advanced manufacturing – but strangely, for all his cheerleading of CEOs, Prof. Cowen seems remarkably uninterested in how things actually work in manufacturing.

For example, this – ‘3D Systems has manufactured more than 600,000 medical device implants for our partners. From the very first FDA cleared 3D printed titanium implant to the development of the next generation of designs and 3D printing technology, 3D Systems has a proven track record and is the partner of choice for design and manufacturing. We use our state-of-the-art ProX® DMP 320 platform to deliver best-in-class material properties and surface resolution for metal parts. Contact us today to discuss your orthopedic or cardiovascular medical device needs in titanium, stainless steels, or cobalt chromium.’ https://www.3dsystems.com/medical-device-manufacturing

But then, Prof. Cowen would likely be concerned at the cost of using such technology if everybody in the U.S. was covered by health insurance, so the less said, the better, right?

Reply

17 ʕ•ᴥ•ʔ December 27, 2017 at 9:04 am

Seems odd to bash Cowen for being right, then.

Technological revolutions aren’t neat. They can work out without being what we expect.

Bitcoin as MakerBot?

Reply

18 carlospln December 28, 2017 at 1:00 am

You definitely have a point about Prof. Cowen being right, as I will readily concede:

“We always overestimate the change that will occur in the next two years and underestimate the change that will occur in the next ten. Don’t let yourself be lulled into inaction”.
Bill Gates

Reply

19 Miguel Madeira December 27, 2017 at 5:58 am

“ the bitcoin blockchain has consumed almost a billion dollars worth of electricity”

In the bitcoin, it was really the blockchain that consumes much electricity, or it is the peculiar system that the BTC uses to put more “coins” in circulation (the competition to be the first to find the value that gives you right to a new block)?

Reply

20 Ricardo December 27, 2017 at 9:57 am

You can’t separate the two so cleanly. If it becomes cheap to add data to the blockchain, it also becomes cheap to delete or replace data in the blockchain which would cause the whole system to fall apart. The idea of the blockchain as a decentralized, tamper-proof database only holds when you have people constantly adding data to the blockchain and when adding those pieces of data is very expensive and time-consuming.

Reply

21 Axa December 27, 2017 at 7:59 am

Bitcoin is 8 years old. Blockchain ideas a bit more. Thus, 10 years after people is still looking for a problem which could be solved optimally with blockchains.

Reply

22 Bill December 27, 2017 at 9:14 am

Actually, stock clearing, bank transfers and clearing are areas where there will be more blockchain application, according to a course I attended for lawyers. Expect to see the government in the middle of financial transactions to identify senders and recipients for approved and legal currency transactions or clearings. You can listen to the course at westlegaledcenter.com but will have to pay for it. In dollars, and not bitcoins.

Reply

23 Ricardo December 27, 2017 at 10:13 am

I’m skeptical. Yes, you could use blockchains for these but what, exactly, are the advantages over the current systems we have in place? Blockchain technology does offer the advantages(?) of decentralized, pseudonymous transactions but it does so at relatively high cost and slow clearing times. In the traditional financial sector, if something goes wrong, the legal and regulatory system along with institutions such as the SIPC and FDIC help spread risk around. The blockchain is more or less tailor-made to facilitate gray market or black market transactions when parties don’t care and don’t want to know each other’s identities and are willing to accept substantially more risk in exchange for the benefits of anonymity.

Reply

24 ʕ•ᴥ•ʔ December 27, 2017 at 10:34 am

I am skeptical as well. Strong encryption is important, and some transaction log may be useful, but as soon as you talk about institutions “in the middle” you have really reduced the whole thing to a message protocol.

Do I care what a bank transfer looks like? Not really.

Reply

25 Ray Lopez December 27, 2017 at 11:34 am

That’s true.

Bonus trivia: apparently, some say, the Bangladesh central bank heist of last year that was processed through the Philippines was due to North Korean hackers, working with various nationalities. They took advantage of the reversible money transfer system of today. Imagine what they could do with a pseudo-anonymous system like Bitcoin.

26 Ricardo December 27, 2017 at 4:33 pm

“They took advantage of the reversible money transfer system of today. Imagine what they could do with a pseudo-anonymous system like Bitcoin”

We don’t have to imagine: https://www.forbes.com/sites/laurashin/2016/12/20/hackers-have-stolen-millions-of-dollars-in-bitcoin-using-only-phone-numbers/

27 Bill December 27, 2017 at 11:54 am

Ricardo, According to the presentation, stock clearing through intermediaries takes several days, as do international bank transactions, whereas blockchain cuts that down to virtually nothing, which is why you will see more of it. There are other features that will change as well: companies maintain through intermediaries and agents the names of holders of their stock; blockchain makes proof of ownership virtual and eliminates intermediaries, which could mean that you could have direct voting on shareholder actions.

Reply

28 Bill December 27, 2017 at 11:57 am

Ricardo, Here’s an example re corporate governance and blockchain technology: https://academic.oup.com/rof/article/21/1/7/2888422

29 Ricardo December 27, 2017 at 4:20 pm

Have a look at the Real Time Payments platform and other similar systems that are being rolled out around the world. For that matter, credit and ATM cards have offered real-time clearing for years. I can go to rural Cambodia today and withdraw hundreds of dollars out of my bank account with a nearly instant clearing time, no blockchains involved. The fact that it may take a few days for an international wire transfer to clear there is just a function of legacy systems that are already being upgraded.

30 Bill December 27, 2017 at 7:00 pm

Ricardo, I believe what I heard from the lawyers and finance persons who clearly said that international banking transactions and stock transactions take the time they claimed they take and that it will be instantaneous. Here is the link to the site that hosted the presentation: westlegaledcenter.com The course is on blockchain technology and legal issues.

31 Ray Lopez December 27, 2017 at 10:39 pm

@Bill, why would instantaneous money transfers that are not reversible be a good thing? Look at the Forbes article that Ricardo linked to. Instantaneous, irreversible money transfers = bad. Slow, reversible money transfers (if caught in time, before the money is transferred) = good. Amex for a while (and maybe still does) allowed you to pay for something like $100k USD via your credit card, instantly. I always thought that was a bad thing. Better for slow transfers (unless it’s a routine credit card transaction of say $2k USD) so you can catch the bad guys in time, before the e-money is lost. Same for central banks balancing. Old “batch” transactions handled at the end of the day, allow checks to clear in three days, is fine with me.

32 Axa December 28, 2017 at 3:34 am

SWIFT transfers cost 30 USD and are same day from Europe to America and next day from America to Europe. Same price nomatter the amount sent and predictable clearing times.

Ps, instantaneous is quite a stretch when compared to the reality of 250K bitcoin unconfirmed transactions at any time.

33 Evan Van Ness December 28, 2017 at 3:34 am

Saying that “blockchain has no use case after 10 years” in 2017 is like saying “computer communication has no use case after 10 years” in 1979 because ARPANET hadn’t gone mainstream.

Reply

34 JP December 30, 2017 at 5:23 am

–> “(And, worth noting, for those seven transactions a second Bitcoin is already estimated to use 35 times as much energy as Visa. If you brought Bitcoin’s transaction volume up to Visa’s it would be using as much electricity as the rest of the world put together.)”

The last sentence is not true, the electricity spent is used in mining new blocks, which is independent of transaction throughput. The mining is purely profit driven, as the block reward diminishes, so does the incentives to mine new blocks.

Reply

35 Ali Lakhany January 12, 2018 at 1:42 am

One thing that we should understand crystal clear, it is that we just CANT make far in Crypto world without having proper awareness of it. I have learned it all very well; it is what has helped me MOST. And thanks to Duecoin, I am in really good shape to gain. It’s ever near the exchanges, it is one of the rare choice that you just CANT afford on missing out. duecoin.io

Reply

Leave a Comment

Previous post:

Next post: