Imperfect Union: How ConstitutionDAO didn't raise enough money | Hacker News Reader

Imperfect Union: How ConstitutionDAO didn't raise enough money

fwb.help

27 pointsconanxin53 comments

> There were other issues standing in the way of the DAO’s egalitarian ambitions. As VICE’s Motherboard points out, the median ConstitutionDAO donation was $206.26. Those small-dollar donors are seeing most or all of their refunds getting eaten up by Ethereum network gas fees, which remain static regardless of the size of the transaction.

This is kind of insane to me. People were shilling ConstitutionDAO _pretty_ hard on here back when this happened. My biggest concern was this was a scam, or that people would use the opportunity to scam in some fashion (still a reasonable theory). Instead, it turns out their supposedly superior financial system can't even handle refunds? Come on.

The scam is the Ethereum ecosystem itself. It succeeded once again in pulling in a few more millions of real money from retail investors who won’t see their $200 donations again.

Where did the money go? It provided liquidity to the top of the pyramid, of course.

> investors [...] donations

one of those is the wrong word.

I generally disagree that willing participants who gave their money without expectation of future return are getting fleeced in this particular project.

You don't need to look far to find rugpulls https://twitter.com/RugDocIO regularaly tweets out ongoing rugpulls.

Interestingly though after a technical change in where gas/txn fees go (EIP-1559) any payment of fees actually does go to all holders of ETH proportionally due to fees fueling deflation. Of course even still this is not like a traditional pyramid scheme, but more similar to stock buy backs being fueled by each individual sale that a company makes.

Gas fees are high because there's so much demand for limited block space. This is being addressed by moving things to rollups, and later by expanding rollup capacity with shards. Some stuff is in production on rollups already, but everything's in flux because the tech is still improving at a rapid pace.

Rollups take transaction capacity from a couple dozen transactions per second to a couple thousand, and sharding will multiply rollup capacity by a factor of twenty initially, and significantly more later as hardware improves.

> This is being addressed by moving things to rollups, and later by expanding rollup capacity with shards.

First, it may be worth considering that non-crypto people don't know or care about any of this jargon. Anyway, I hope it works. To me it just sounds like deferring trust to smaller but faster institutions/coins with proven track records. Doesn't seem all that different from the current financial system. At any rate, there are more issues than just gas fees.

Rollups are a way to compress the storage space required by a transaction. They store the compressed versions on chain, and there are cryptographic guarantees that everything is correct and that users can withdraw funds to regular on-chain accounts.

Shards expand the storage available, by making it so that individual nodes can store just a portion of the total stored data. Each shard gets the full security of the main chain.

The issue with refunds was that gas costs were high and many contributions were small, so that issue at least would be solved as scaling improves.

Thank you for clarifying the specifics. Sounds unnecessarily complex, to say the least.

If you have a simpler way to achieve 100,000 tx/sec in a large decentralized network with strong security, then you should probably publish it. The alternatives I've seen so far achieve scaling by sacrificing decentralization.

I have no such need for a large decentralized network responsible for handling my money as I have access to regulated (albeit not enough) financial institutions in the US and USD.

IPFS/BitTorrent/Matrix is cool though. Can any of this research help in those areas?

The networking technology from IPFS/etc will likely be helpful. Something a little different about Ethereum shards is that it needs stronger guarantees on data availability. On the other hand it's less ambitious in terms of total storage capacity.

I tend to get heavily downvoted here whenever I suggest that Ethereum might be useful for something, so I tend to keep my comments strictly technical. But I'll take a chance and say that Uniswap, for example, is a pretty interesting financial innovation with no equivalent in legacy finance. And just the other day, the International Emissions Trading Association published a statement saying:

> Digital tools hold the potential to revolutionise monitoring, reporting and verification, and they offer new means for highly secure, transparent and globally accessible registry infrastructure. There are also innovations in asset formation, such as issuing carbon credits in the form of digital credits or native tokens secured on a public blockchain.

https://www.ieta.org/page-18192/12286504

I appreciate the accurate summary of the technology and scaling solutions you provided.

It would be refreshing to see this community actually have a dialogue on the tech.

I'm sure it can improve in the future, but it seems that at the moment Ethereum is very unsuited for small transaction style fundraising, like this. The level of transaction fees is orders of magnitude higher than other options (heck in some countries the fees to move money to another bank account is 0)

Side note, the term “gas fee” is a nice example of how crypto projects pick needlessly obtuse vocabulary. It’s a transaction fee. There’s no “gas” involved. Why call it that?

Perhaps I’m jaded but my feeling is that using plain terms would reveal the emptiness of the space. A non technical person can see that a $200 “transaction fee” is a rip off.

Jargon is inevitable.

Have you ever listened to an investment banker in their native habitat?

Gas (as in a car) is a useful analogy to the function of this fee in the protocol.

Transaction fee implies something else all together.

Something else like what? It's a transaction and you pay a fee, what's wrong with this explanation?

Since you asked, here's where the terminology comes from.

The Ethereum VM has a bunch of different opcodes, some physically more costly than others. To account for this, each opcode has its own "gas cost." Storing a value costs much more gas than adding two values.

The gas price is the current price in ETH for one unit of gas. The gas price fluctuates with demand.

The transaction fee is the total ETH paid for your transaction. If you have a complex transaction that executes a lot of expensive opcodes, you will have a higher transaction fee than someone doing something simpler, even though you're both paying the same gas price.

Sounds like the quote from the article is using the term "gas fee" to refer to the overall transaction cost. At least I don't see any reason why someone would specifically car about the gas fee portion of the overall fee in the context of losing most of the money in the refund process.

Users focus a lot on the gas price since it's a consistent indicator of the cost to transact, so they've tended to use the terms "gas fee" and "transaction fee" interchangeably. Your transaction fee is used entirely to buy gas, which is spent in running your transaction.

On the other hand, there's SpiceDAO who spent three million dollars to buy a copy of Dune sketchbook (whose starting price was 25k), and pretend it gives them the right to produce new IP based on Dune.

So what does that tell you about decentralized decision making other than "people are stupid"?

Did they pretend and/or think this? The widely shared tweet I have seen simply stated that they want to develop IP based on Dune. It did not say that permission to do so is derived from the book.

> (whose starting price was 25k)

That is how auctions work. The only concern with such setups is that information about the amount of money available to such a DAO bidder is public knowledge and thus can be exploited.

The tweet is this: https://twitter.com/TheSpiceDAO/status/1482404318347153413

And I'll quote it in it's entirety for posterity:

--- start quote ---

We won the auction for €2.66M. Now our mission is to:

1. Make the book public (to the extent permitted by law)

2. Produce an original animated limited series inspired by the book and sell it to a streaming service

3. Support derivative projects from the community

--- end quote ---

Since then they've backtracked saying that they are looking to develop "completely original IP", but that raises the question "why spend money on the Dune book then?".

Ah yes. And don't forget their Q&A series of tweets. I'm linking to the best one: https://twitter.com/TheSpiceDAO/status/1484285677646729219?s...

--- start quote ---

Q: Are you aware that purchasing the book does not give you its copyright?

A: Yes. After two months of outreach, conversations with former business partners and consultations with legal counsel we were not able to reach an agreement with any of the rights holders...

--- end quote ---

And yet, they still were "hell bent on buying the book at any cost".

Now that they realized they have nothing but an expensive book on their hand, they are "developing a new animated series" that they have no money for. You can see the hilarious exchange here: https://forum.spicedao.xyz/t/original-series-package-strateg...

"all you need to approach a streaming service for a sale that would finance the production is a package and not a finished product"

and

"We were quoted $500K USD for a writer’s room early on and quickly abandoned the thought"

They spent 2.6 million dollars on book. This would've comfortably covered a huge chink of an original animated series.

It seems to me that the communication both in the original tweet and the subsequent communication is consistent:

- They did not think it gives them to copyright. - They want to create /original/ IP merely /inspired/ by the book. - These are two separate ventures.

I have no position on whether their decision to spend 2.6 million dollars on it is smart or not; people can ultimately buy what they want, and if people fundraised 2.6 million to buy a book, and the decision makers therefore had the authority to spend as much, I see no problem.

The only reason they should have spent $2.6m at auction is, if there was another bidder taking them that high.

Short of someone literally bidding against them with the foreknowledge of their maximum bid, it seems very unlikely that anyone else would have gone close to that number from the $35k expected price.

A more sane approach if they realised that someone was bidding the up would be to duck out of that auction and by any of the other copies of the book that exist, likely for faaar less than $2.6m

Due to the nature of the DAO, their maximum bid was known beforehand, and I don’t know if there was a mechanism to bail out at that point, and with short notice.

Edit: Except the person/people tasked with doing the actual bidding making an executive decision to, but would that have been within their mandate?

But why buy the book, for any amount? It was already available scanned to everyone for free online. The purchase literally didn’t further any of their stated goals, unless they thought it would give them rights to the IP.

It seems clear that the purchase of the book was a purpose in and of itself.

Why would people come together to do that? Owning a copy of the book gives them no rights or possibilities to do anything they couldn’t before, except keeping it at someone’s house or do viewings of it in meatspace. Is the latter the idea?

To correct some information, the 25k-35k euros was the estimated price even inflated for the success of the film. The book is rare, but not unique.

You don't pay 100x est to own a physical copy of a non-unique book, a scan of which is available in full online, for 100x without thinking there is some additional benefit to winning.

Even if you remove the 'they thought it would give them IP rights' stupidity defense you're left with blowing 100x estimated value of other people's money stupidity situation, which is just as bad.

You're just swapping one stupidity that should discredit them from running any kind of DAO for another that should do the same.

> So what does that tell you about decentralized decision making other than "people are stupid"?

That people are also greedy, and prone to motivated reasoning.

Are they still planning to mint NFTs of the book and then burn it to increase the NFT price?

I’m shocked people believe such a simple story … they didn’t mistakenly pay 30x the value of the asset based on a legal misunderstanding … it’s money laundering.

For this to be money laundering (and not a donation) they need to be on the other side of the trade.

Is this not trivial to do by simply participating in the DAO with anonymized wallets?

Article does not support the assertion in the title. To be clear, I'm a cryptoskeptic. But by the article's account, ConstitutionDAO never "lost its way." They just didn't raise enough money to accomplish their goal. It seems like the founders acted with integrity throughout and did not make any moves that could even clearly be labeled as a mistake.

Perhaps the choice of the cryptocurrency that they used and showing how much they raised publicly was their 'mistake' then.

Maybe. Those are not clearly indicated as mistakes either (i.e. the article does not demonstrate or even argue that another choice would more likely than not have led to a better outcome).

And the article conveys some justification for the choices that did get made. For example, they couldn't use certain "second tier" Ethereum features that would have reduced gas fees, because they aren't approved in states like New York? So maybe you argue that such features should have been used, but then you're optimizing for a different outcome -- the outcome where you don't win the auction. With the benefit of hindsight, sure, but the leaders didn't know before the auction that they would lose. So it was certainly a reasonable choice to take the approach that seemed most likely to raise the most money.

> Maybe...

Definitely.

> showing how much they raised publicly was their 'mistake' then.

I’ve seen countless people mention this as an issue, but the only time this would really matter is if you win, but the other bidders used the information to bid up the sale price. The fact is the other bidder had unlimited resources and desire to win the auction and so would have outbid the ConsitutionDAO even if they didn’t have knowledge of their budget, it’s not as though Ken Griffin would have given up before his maximum budget because he didn’t know ConstitutionDAO’s total funding.

Ok, we've put that in the title above.

The article mentions high gas fees impacting refunds, and I was wondering what the extent of that impact was. The article doesn’t answer that question, but it does link to a tweet which links to a page of stats about ConstitutionDAO that does: https://dune.xyz/ilemi/ConstitutionDAO-Funding-Tracker

55.78% of funds returned! Truly an incredible accomplishment, and not in a positive way.

(edit: On second thought, does the 44% of original funds remaining include people who haven’t tried getting refunds? Curious if there’s an estimate of ETH lost purely to transaction fees.)

> not in a positive way

Presumably it's positive for the people collecting the fees, so that's OK then.

Which is negative for those of us who want to buy a frigging graphics card at a reasonable price.

Post eip-1559 most fees are burnt instead of paid to miners.

Here's hoping that eth's merge happens soon

Yea, if you don’t request a refund, you don’t get it so yes, that % includes people who haven’t requested a refund.

Also, this token became a “meme coin” in the weeks that followed the auction loss, returning something like 10x or more on Uniswap. More sophisticated users traded their tokens there, rather than going through the official refund process.

Another reason people wouldn't request a refund would be that they would actually lose money by doing so if their original spend was under 200$ because of "gas fees".

I might have the query wrong, but I am seeing 0.8% of redeemed fees were lost to transaction fees. I'm sure someone could also generate something that looks better into the effects of redeemtion size & whether people are just not redeeming because too much would be lost in txn fees.

Anyway here's the query:

``` WITH burnt as ( SELECT SUM(gas_used * (gas_price + priority_fee_per_gas))/1e18 as eth_burnt FROM ethereum."transactions" WHERE SUBSTRING("data",1,4) = '\x3015a5b5' --redeem AND "to" = '\xd569d3cce55b71a8a3f3c418c329a66e5f714431' AND SUBSTRING("data", 37,32) = '\x0000000000000000000000000000000000000000000000000000000000000024' --constitutiondao projectid ), ), returned as ( SELECT SUM(bytea2numeric(SUBSTRING("data", 101,32))/1e18) as returned FROM ethereum."transactions" WHERE SUBSTRING("data",1,4) = '\x3015a5b5' --redeem AND "to" = '\xd569d3cce55b71a8a3f3c418c329a66e5f714431' AND SUBSTRING("data", 37,32) = '\x0000000000000000000000000000000000000000000000000000000000000024' --constitutiondao projectid AND "success" )

SELECT (SELECT * FROM burnt)/(SELECT * FROM returned)*100 ```

To be used on dune analytics: https://dune.xyz/home

DAOs are just another convoluted scheme to try to make people buy cryptocurrency.

"The Problem With NFTs" explains it very well https://www.youtube.com/watch?v=YQ_xWvX1n9g

This is tangential (and just an observation on poking around their homepage to find out what fwb.help is), but if landing pages are going to swing back toward having 90s-styled tickers, then is there a reason we don't just bring back the <marquee> tag in some form? (Or standardize a modern equivalent, like <ticker> or something?)

Asking as someone who actually used these in my nascent dabblings in HTML and is really confused that modern webdev has moved to replacing them all with non-semantic (and manually-styled) <div> tags. I'm sure there's a reason that I missed sometime between my junior high message board days and when I actually picked this up as a profession, but whenever I get an immediate glance at one it's always kind of seemed like kind of a regression. (At the least, having a dedicated standard for this again would allow for element-specific CSS options, like what <li> has for instance.)

It’s very ironic that of all the people for ConstitutionDAO to lose the auction too, it happened to be Ken Griffin.

Reply on news.ycombinator.com