Hacker News Reader

Comment by intuitionist | Hacker News Reader

The big issue here is that it’s not clear that you need a blockchain for any of it. Economically you’re launching a Twitter clone with the intention of spending $1B on customer and content acquisition. The rest is details of how that’s budgeted.

The smaller problem is in those details—the scheme described is highly vulnerable to attacks like the following: I create a network of private bot accounts who post high-frequency generated content, and a follower account who follows them all and installs the browser extension. Whenever I want I can print money (well, Twitcoin) by uploading these fake hornet tweets to the new platform. I can think of ways to defeat this specific attack relatively easily, but I’m not sure it’s easy to defeat all such attacks.

View on HN

If you're paying users to scrape, how do you ensure that they are indeed scraping and not just making up spam? You've got a catch-22 here - you need people to scrape but at the same time you need an authoritative source of the scraped data to ensure people aren't cheating, but then if you've got that source then you don't need those people to begin with.

Interaction with the real-world (outside the blockchain) is where the vast majority of these bullshitchain projects fall apart.

> how do you ensure that they are indeed scraping and not just making up spam?

For TLS 1.1 (and supposedly "TLS 1.2+" since https://github.com/tlsnotary/PageSigner/commit/4e12573): https://tlsnotary.org/pagesigner

For TLS 1.3 there's also TLS-N from https://eprint.iacr.org/2017/578.pdf but it requires a cooperating server:

"TLS-N generates non-interactive proofs about the content of a TLS session that can be efficiently verified by third parties and blockchain based smart contracts. As such, TLS-N increases the accountability for content provided on the web and enables a practical and decentralized blockchain oracle for web content. TLS-N is compatible with TLS 1.3 and adds a minor overhead to a typical TLS session. When a proof is generated, parts of the TLS session (e.g., passwords, cookies) can be hidden for privacy reasons, while the remaining content can be verified."

You could bypass this kind of verification by routing your spam through Twitter. Send spam to twitter, scrape spam from Twitter for twitcoin.

If your plan is to scrape all of twitter, the question is if that matters among the years of existing spam.

Probably not. If it does start to matter it should start becoming an issue for twitter too and they'd probably have to deal with it.

> If you're paying users to scrape, how do you ensure that they are indeed scraping and not just making up spam?

Ironically the fact that they are paying 10x for original "tweets" posted through their interface makes it so that nobody who can think would try fake scraping. You are 10x better of just blasting your crap directly at their interface as "original" "content". :)

But why is the scraping so crucial? The history is probably less important to most users than the reach of their posts, so the mirroring on Twitter seems more crucial (Twitter could probably prevent that if they wished, though).

I think blockchain is actually central to the scheme. Like this:

The ostensible scheme can't possibly actually work (it requires Twitter to not do anything to stop it -- of course, Twitter would work to stop it to the extent it ever actually threatens them).

What can work, though, is the narrative of the scheme. What you need is a mechanism to incentivize people to promote the narrative of the scheme.

That's where the crypto comes in. It's a viral mechanism to franchise selling the scheme. Anyone can buy in, promote the narrative, and get others to buy in. It doesn't really matter how many of the people have a deep belief in the original narrative or not. Get people's attention, get them to believe there is more money in it if they put in a little money, and once the money starts moving around, there will be opportunities to grab some, or perhaps a lot of it.

if you're relying on the narrative, you're going to run into the same problem as any of the other twitter clones that have tried to siphon away twitter's audience by targeting a specific demographic (e.g. parler) - you end up with only that group, and the prolific posters who bring the most value to the platform do so because they're chasing the broad audience that twitter provides.

the crypto bros don't want to perform for an audience that's only other crypto bros, they want to reach the people who aren't part of their circle, and twitter provides that. same for the right-wingers - if they can't reach the suburban wine moms, there's no point in posting.

> It's a viral mechanism to franchise selling the scheme. Anyone can buy in, promote the narrative, and get others to buy in.

This is a great definition of cryptocurrency. It nicely frames how they become pyramidal.

>The big issue here is that it’s not clear that you need a blockchain for any of it.

Doesn't the same apply to many blockchain applications ?

Yes.

Blockchain has extremely narrow potential in a decentralized social network. Blockchain was designed to prevent double-spending. In a social network, the only analagous purpose might be to reserve a name, or to locate the servers (bypass DNS).

It would be ridiculous to use a blockchain to authenticate posts, upvotes, friend connections, etc, since these should all be done with digital signatures alone

Yeah, I think adding users post, comments will add too much of friction to the app. I'm trying to build a social media on Polygon chain but haven't figured out do I store the content on chain or not.

The gist of the idea is that:

- if you provide an essentially identical service

- you make it seamless to switch

- and you incentivize users to switch

Then you may be able to aggressively siphon users from your target.

OP’s example is using a blockchain as the distribution mechanism for the incentives.

> you incentivize users to switch

Lot of startups do this but have pretty high CaC. In the Sushi/Uniswap case, users were already crypto-native. In the twitter case, the majority of users are not. There's just going to be a bigger barrier than in the former case. Sure, I think if you spend money some people will switch, but it's much less clear to me how much money you'd need.

> The big issue here is that it’s not clear that you need a blockchain for any of it.

This could be said about 99.9% of all blockchain products and companies.

Seems like you need the blockchain there to guarantee that your outrageous copyright violation becomes unpunishable.

Basically this strategy is "steal Twitter and fund the effort with another crypto Ponzi".

That's just one of the big issues.

Another is that Twitter is 6000 transactions per second, whereas Ethereum only manages about 15tps - 1/400th as much - and BTC 7tps.

I think this is way more significant than many people think. Twitter scale information distribution is a hard problem, and tying the platform to a blockchain makes the engineering side of this impossible. You'd want your blockchain to run independently from the social media platform.

You could add the crypto layer to Mastodon, and by the time scale becomes an issue, you haven't irretrievably crippled yourself and you build it from there. Plus federation and so on.

I would think scraping Twitter would come with legal hassle, so a lawsuit could be worse than any technicalities.

Perhaps a way to migrate this is to only allow tweets < a certain creation date. Would this make TWITCOIN have a hard limit similar to how Bitcoin has a Hard limit ?

That's going to be 0.000000032 Twitcoin. If you want a a coin to succeed you need to make the amount people pay easier to fathom. This goes especially for people that come from a fiat world.

i think a better way is to forgo the browser extension and simply pull a users timeline via the twitter API. Only grab tweets within the past 30mins to an hour (this interval is arbitrary but can most likely be optimized) as to prevent spam

Reply on news.ycombinator.com