[eejjjj82]

Imagine if in 1975, the editor of your local news paper said that solar cells are the worlds least efficient energy source ever invented and they were the technology industry's greatest failure. Or in 1995 you read the same article about websites... now remove all journalistic integrity and quality from those misinformed articles and you have what this post is about.

This is an editorial fluff piece with a very poor core argument. I'm very surprised has risen this high on HN, usually the opinion pieces here have far better formulated arguments.

e.g. "Ok, so what do we even mean when we say “blockchain”.The answer is really that nobody knows, it’s not a term that has any precise definition and it really depends on who you talk to and the context. (Source: https://threader.app/thread/1363418896301228033)"

The irony being... the author also never bothers to spell out /which/ blockchain he is ranting against.

Again, he get's sidetracked on his own strawman argument of what a blockchain is...

"Now some introductions will tell you a blockchain is a data structure from computer science which consists of a linked list of blocks which are iteratively hashed so that subsequent blocks depend on the hashes of previous blocks. (7/)However the utility of a term of art for any piece of software that includes two of the most common concepts in computer science (hashing, linked data structures) is just bad classification. (Source: https://threader.app/thread/1363418896301228033)"

He entirely misses the point that the novelty of what a blockchain is, isn't a fancy linked list, it's a mechanism for distributed consensus. Where by distributed computing algorithms from Paxos -> RAFT and private implementations of distributed databases like BigTable, Dynamo, Cassandra, and now next generation iterations like Hazelcast, Redis Grid, Spanner, etc... have enabled the greatest leaps in computing scalability of the past 10 years. However these are all privately maintained databases.

The "blockchain"... whichever implementation to which you refer, must be defined in its usability as a trusted distributed database. The mechanism of the datastructure to represent it is the least interesting or novel part of the technology.

The rest of the author's "points" regarding utility and energy consumption are equally misinformed and poorly formulated.

All in all the author just doesn't seem very informed of what blockchain is beyond the sorts of talking head gossip you'd seen on CNBC. His own limited imagination into the possible uses are akin to people from 1995 who couldn't imagine why websites would be so big or that solar panels would ever be efficient enough to generate energy.

[vnxli]

Why is HN so anti crypto?

Blockchain/distributed databases seem like really cool tech with some valuable real world applications outside of just coins and stonks but I feel like we never get into that here.

Or is this just a vocal minority thing?

[rcxdude]

Because it's a technlogy with a huge amount of hype behind it but very little demonstrated applications. Especially when that hype leads to huge amount of money being thrown around by people with little familiarity with the technology. This is bound to get pushback. I agree there's some potentially interesting applications, but there seems to be little of that kind of reality coming from the crypto-sphere.

[sixQuarks]

I think it’s similar in nature to when IBM’s founder said “I think there is a world market for maybe five computers”

[ForHackernews]

Still waiting to see any of these "valuable real world applications".

I've seen some proposed vapourware nonsense about voting-on-blockchain or identity-cards-for-refugees-on-blockchain but they didn't make any sense as far as I could tell.

[tomw1808]

Most of these projects are crap, but there are some very legitimate use cases when it comes to binding businesses together. Blockchain solutions make it possible to transact easier than before.

One particular use-case that stayed with me for a long time is what DeBeers is doing with the diamond tracing. For long time diamond certificates were faked between Afrika and Antwerp, especially on border checkpoints. This is where blood diamonds were introduced into the supply chain. Participants cannot be trusted and a common ledger (a.k.a Blockchain or DLT) is a solution. With a traditional database you'd have again a "super administrator", which is exactly what you want to avoid. This has, obviously, nothing in common with a public network like Ethereum, but it's still "Blockchain" and I believe it's still a very valuable real-world use-case. https://www.tracr.com/

Paul Brody does also a lot of real-world projects with EY.

[rcxdude]

The question with this kind of application is usually "does this actually need a cryptocurrency framework? (or even benefit from it)". For example, with these diamonds: it's not like you need a distributed consensus system to create a cryptographic certificate of origin for diamonds (or any other good). Usually the challenge is in being able to connect the certificate with a specific item (which cryptocurrency tech does not help you with). Unfortunately, like with many other websites of projects in the crypto space, their website does nothing to explain how they are actually solving the problem, and just throws out buzzwords.

[tomw1808]

What you say doesn't make sense to me.

A blockchain is a tool, like any other. All the properties of a blockchain make it the _ideal_ tool for this use case. It has all the functions literally built in to create and store and interact with a cryptographic certificate (if you will).

Imho it's the right tool for the job. The only reason to dismiss it as the tool of choice is a personal belief. You don't like, then simply don't use it, but objectively the "buzzwords" are the tech itself, it's just how it is.

Just because they are not telling you exactly how they do it, doesn't mean they don't make a good job at it. Apple and Google also don't tell you exactly how their 5G stack looks and works internally, all you get is a nice reactive UI to interact with Apps which are connected to the Internet to surf HN while sitting on the toilet...

[rcxdude]

The maintenance of consensus in blockchains is expensive, both in terms of resources used in proof of work (though other less wasteful consensus rules are possible), and in terms of technical complexity. It also has certain tradeoffs in terms of immutability which aren't always desirable. For example, what happens if ownership of the diamond token on the blockchain and ownership of the physical diamond get seperated somehow? From the looks of things (again, technical details are extremely sparse on the website so I'm forced to speculate), this is a new technical development not leaning on any particular existing piece of code, so by choosing blockchain they have chosen to take on extra work to implement it (and blockchain consensus is not easy to implement). If they were to e.g. implement it on top of etherium somehow then I could buy they are riding on something 'built-in' (though they would then be paying the cost of etherium 'gas'). This is why I am distrusting of systems which aim to use blockchain when they don't need to (and certainly the existence of systems which work but they could work just as well or better independent of blockchain doesn't point to blockchain being a useful technology).

> Just because they are not telling you exactly how they do it, doesn't mean they don't make a good job at it

It also doesn't mean they are doing a good job. Absent visible results (I have this for 5G), a technical explaination (with 5G I can read far more detail on this than I ever would want to know just from public documents even without all the internal details), or even a description of how this is intended to be used or exactly what guarantees it provides, how am I supposed to take this as evidence it works (let alone is a natural solution to the problem)? It's basically "just trust us, bro", ironic for a technology which is supposed to work in a trustless manner.

>The only reason to dismiss it as the tool of choice is a personal belief

I do think blockchains have potential, but it is a technology which is only sensible in a very narrow set of circumstances, because it is a solution with a lot of extremes owing to it attempting to solve an extreme problem (global trustless consensus).

[vcavallo]

crypto is all wrapped up with other things that the powerful, vocal, hateful slice of the internet like to trash in public (while blindly supporting one another on anything that feels connected).

[slfnflctd]

To me, the biggest question seems to be whether distributed trust and/or consensus is best served by blockchains, or if we've already established safer and better ways of doing this in prior systems that have been around longer. For now, industry is overwhelmingly sticking with the latter. After all, we've been contemplating and refining trust & consensus systems for the entire history of civilization. It all comes down to relationships and reputation in the end-- and even blockchain tech itself seems to need to deal in those terms (at least at higher levels).

I wouldn't rule out there being certain use cases where blockchain may turn out to be the best way to go, but at the moment that doesn't appear guaranteed to be a huge or lucrative field by itself (and yes, you can say the same about solar in 1975 or websites in 1995, but that proves nothing as there were many more ideas with similar uncertainty around them that failed alongside those). Meanwhile, the intense amount of speculation going on just smells wrong to a lot of us in light of the dearth of clear applications or industry uptake. Personally I'm skeptical, but undecided. It's mainly the speculation that gives me the willies.

I agree the author should have discussed these issues more.

[erickj]

There is absolutely snake-oil and speculation slithering amount the beneficial projects.

Far too many people see this as a quick way to dup others and make a quick buck, because that has been profitable so far. However that in itself doesn't diminish that decentralized distributed consensus is a new class of tool that did not exist before Bitcoin. And while Bitcoin is a very inefficient implementation as we're learning, there are now new protocols with far better efficiency guarantees than proof-of-work

[ForHackernews]

You've had 10+ years to come up with something useful to do with a blockchain!

Ten years after the web opened for commercial use we had Google, eBay, Amazon, Facebook.

Ten years after Napster we had BitTorrent, Kazaa, a dozen others sued into irrelevance.

Ten years after the first smartphone we had...everything we do with smartphones today.

Like c'mon, if this technology is so revolutionary and transformative when are we going to see it transform something, anything, beyond illegal drug sales and ransomware!

[noxer]

We have DEXs and it didn't take 10 years only 5 or so. There is plenty more stuff only possible with DLTs. By the time mainstream uses its products they wont know it runs on DLTs just like people using tiktok dont know it runs on TCP/IP. yet TCP/IP was once a stupid idea no one needed either. You couldn't even call someone with it.

[markburns]

Well perhaps we’re 10 years since ARPANET. And the World Wide Web hasn’t been invented just yet.

[erickj]

you need to recognize that this is work being done by collectives of people scattered across the globe.

the coordination of this effort is more akin to countries negotiating a treaty than operations inside of a silicon valley garage startup.

also, 10 years is really not a very long time.

[drexlspivey]

You are surprised that a piece bashing cryptocurrency gained traction in HN? You must be new here.

[sjg007]

People did write that solar cells were useless and they also wrote that the web was a joke.

[erickj]

Yes... but I'm helping those that weren't alive until 1995 know that these things happened. There are many among us.