[RFC] Using license as permissive as Uni v3 and more permissive than Curve

As far as I know, all Pickle code is open source and it is entirely appropriate to fork us and use our intellectual property for commercial purposes.

We have just committed to expand significantly and spend $2.5m, largely on intellectual property.

I propose that for Pickle code we use license the same as Uniswap V3 license – it limits use of the source code in a commercial or production setting for up to two years, at which point it will convert to a GPL license into perpetuity.

In other words, I propose that we become as permissive as Uniswap V3 and more permissive than Curve, which has a permanent license. Both are widely considered great DeFi poster children.

“Ok, so how does that work in real life?”

Let me describe examples of Uni V2 and Curve. It’s criminal to fork Cruve, it’s ok to fork Uni V2.

We are all impressed with Curve, CVX and YFI. But do you know how important was the license in securing Curve’s dominant market position?

Curve gets forked by Swerve Finance. Launches a big campaign against them, Swerve goes bust.
Curve gets forked by Saddle Finance. Launches a big campaign against them, Saddle gets removed from CoinGecko, Saddle’s upside is limited, no CEX or VC wants to associate with them.
Saber - low effort curve fork on Solana gets $4b TVL. It’s such low effort that code says “some number from curve” with devs not understanding what it is and what it does. Curve threatens legal action, Saber will not get CEX listings or VC funding.

Curve licenses a fork on BSC and takes revenue from there. Curve launches on Polygon and Arbitrum and takes revenue.

Meanwhile Uni V2 gets forked by Sushi and loses large market share on Ethereum.
Quickswap fork prohibits expansion to Polygon.
Trader Joe prohibits expansion to Avax
Uni V2 gets forked left and right.

Another example how licensing the code leads to better financial returns: HEX.com. Forking HEX is illegal. It is possible that in the Adoption Amplifier crowdsale, Richard kept buying HEX from himself for ETH that he recycled through CEXes, similar to what he praised in EOS sale. HEX now has market cap of 570b unlocked float and 60b staked, with HEX trading at $0.4 each, with total valuation of $255b (see hexinfo). Mcap of ETH is 410b and BTC 896b. All thanks to lack of fair forks.

My main fear is as follows: A competent group forks Pickle, deploys it somewhere, maybe on another chain, gets a bit of traction and then gets a big investment from one of the leading DeFi VC groups. That VC group starts to favor the fork through VC portfolio connections and eventually manages to outcompete us. For example, the fork gets liquidity mining from Balancer, Sushi, partnership with Uni, YFI/CVX-like derivatives, CEX listings but we don’t.

“What are you smoking, this has never happened and will never happen!”

We don’t have strong VC backing. We’re not a portfolio company of a big VC group.

Some VCs did not get on the Uniswap round and I believe funded “Chef Nomi” group to vampire attack, creating a “fair launch” competitor - Sushi. Extreme volatility and drama confused retail and created big buy opportunities.

Optimism is a portfolio company of the VC group owning Uniswap, as a result they have refused to give Sushi the same early access as they gave to Uniswap.

Three arrows capital and Polychain could not participate in Polygon’s success, so they have funded the very similar competitor – Avalanche to outcompete Polygon. Avax is largely the same as Polygon, with a slightly better bridge, but the capital creation now ends in the pockets of 3AC, Polychain and partners.

Portfolio companies within the same VC group have strong partnerships, for example Balancer does liquidity mining for MTA, AVAX and SOL. Ever wondered why Coinbase listed MTA much earlier than CRV? If our fork gets all the VC support and steers portfolio projects away from us, we might miss out on partnerships, liquidity mining, and listing

If you are an optimist, you expect YFI, CVX and other lockening wars on DILL. veToken and Frax are the first ones that come to mind. These wars were forced to happen on Curve because Curve murdered all the copetition, despite competition being VC backed. If you want to have lockening wars on DILL, you might want to make sure that we don’t have to compete with low effort forks supported by VC which has sniffed an opportunity.

Jake Chervinsky
@jchervinsky:
1/ One interesting piece of Uniswap v3 is its use of Business Source License (BSL) 1.1, which restricts production use for two years.
I see people asking if @Uniswap can really use BSL 1.1 against v3 forks. Answer: yes, mostly, at a cost.

1 Like

Unfortunately, to my knowledge, pickle is not incorporated anywhere. This means we have no official person or designee to own the copyright. If we took a risk and assigned the copyright to a single specific person, we would be opening that person himself up to lawsuits. And even worse, as owner of the copyrights, that individual could bail and essentially own the whole protocol.

Furthermore, the code is forked from other existing code, and to what extent we (whoever “we” are) even own the copyright is very much debatable.

1 Like

Can we put: “Pickle DAO” as the copyright owner, maybe add the PICKLE address and never enforce that? :slight_smile:

As far as I know, as long as we make changes to the original code, it’s new intellectual property, regardless of whether we have forked something or not.

I would never want any real lawsuits. Curve defended themselves without any lawsuits.

HEX coin does not have and owner, organization or a specific copyright owner. It has been enough to defend HEX from potential low effort forks gaining traction. What is worse, if Richard ever admits to own the origin address, it will cause him more harm than good. In fact, the guy feels so rich, that he’s forking Ethereum https://twitter.com/RichardHeartWin/status/1390886255240560642. I think he forked ETH so that HEX looks more alive with more transactions and Pulsechain is now almost ready :grinning_face_with_smiling_eyes:.
The whole HEX story is insane, but the bottom line is that if his ponzi got a fair launch fork, he would not get a ride to a 255b valuation…

The thing is, curve and uniswap are both incorporated, as crunchbase and as uniswap. This means they have home offices, lawyers, and a jurisdiction in which they could sue if they chose to. This gives their legal threats a lot more power… You can threaten to sue if you have all of the above. If you don’t, your threats are like those of an armless midget threatening to punch you in the face.

Pickle is not incorporated, has no home jurisdiction, and thus no laws or nations to which our copyright claims Automatically apply. (this is my understanding and does not make it 100% true). It would be different if we wanted to put the copyright in the names of specific people, but then those people would assume legal liability for the platform in one way or another.

I believe our code is currently licensed as MIT license. This allows business uses. But the real question is which parts of the code you would prefer to see secured. The contracts, the UI, tsukemono, etc.

I never want anyone to be sued. No provisions should be made to ever enforce that. I think it’s OK to disallow low effort forks for 2 years and then make it free for all, just like uniswap.

I’ll explain in a story:
Richard Heart has contracted devs to write HEX, a ponzi that will make late buyers lose money. Uniswap denied HEX being on the V2 UI. Coingecko refused to list. CEXes refused to list. When it started pumping, people were not happy. Then all the sybil accounts got Uniswap airdrop.
There were efforts to do a fair launch HEX fork with origin address replaced with black hole 0x000000 or fork/ETH AMM pool. But in the end, the opinion was that the copyright infringing fork will never fly, despite nobody every suing anyone. There are theories that Richard owns the origin address and for the legal reasons he should not disclose that. If that’s the case, he would never sue anyone over infringing HEX copyright. Still, being a copyright infringer would be enough to turn away CEXes, VC and some users.

This is a reply to Rawb on Discord:

I read your posts but my point was that curve has a company that makes the copyright claim a credible threat and actually does discourage low effort forks. We, on the other hand, do not provide a credible threat and thus won’t actually discourage low effort forks.

HEX has never been incorporated. All they did was write: “All rights are reserved. IT IS NOT LICENSED FOR COPYING.”

This has successfully defended a $255b coin from low effort forks (that is 50% of ETH Mcap and 25% of BTC Mcap).

Curve finance did not license the work to Curve corporation. The Curve contract is licensed to “curve DOT fi”. (I can’t post more than 2 links)

I advocate for changing the current MIT license on github to the following and I am sure it will discourage low effort forks:


(c) pickle.finance, 2021

Licensor: Pickle DAO (www.pickle.finance)
Licensed Work: Source code submitted to pickle-finance · GitHub
Change Date: Two years from submission of the code to GitHub
Change License: GNU General Public License v2.0 or later

All intellectual property contained in pickle-fiance GitHub repository has been published for informational purposes only. No license, right of reproduction, distribution or other right is granted or implied until the change date two years from submission to GitHub. For the first two years, rights are reserved by the copyright holder. Two years after submission to GitHub, the license changes to GNU General Public License v2.0 or later

The Pickle DAO may make an Additional Use Grant permitting limited
production use.

THE SOFTWARE IS PROVIDED “AS IS”, WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.


Notice:

This license is not an Open Source license. The Licensed Work will be made available under an Open Source License two years from submission to GitHub.


If Pickle DAO ever needs to enforce that and struggles, I can help.

We have made plans to expand and grow, let’s prepare for the success. Yes, we’re not incorporated, but we may incorporate in the future.
FATF will issue guidance in Oct/Nov, as early as in few months Pickle DAO might be forced to get a legal entity an register as a Virtual Asset Service Provider (VASP).

While I personally don’t agree with this particular proposal, I do want to provide some supporting evidence that unincorporated entities can still have a copyright.

This link is one article of many about a slightly relevant information to how a DAO might enforce its trademark/copyright:

2 Likes

Controversial idea, very little downside and potential to avoid existential threat.

Shortening the change period to 1 year or 6 months may be less controversial but just as effective.

Yearn vaults are GPL so we can’t apply that license to the forked yearn vaults. The license would have to include an exclusion for forked Yearn vaults which would be under GPL.

Pickle UI on GitHub does not have a license description so it’s “all rights reserved” by default. I think it is Pickle’s original work so it is possible to explicitly license that. Low effort fork would need both the contracts and UI so it would work to an extent even without licensing contracts.

If the “let everyone fork this” attitude prevails, let’s at least change MIT to GPL, so that Morgan Stanley does not fork Pickle contracts, applying a restrictive license. Change to GPL would require all derivative work to be GPL as well.

I can see that the main driver for this idea is to avoid competition against protocols with VC “protection” and composability, starting from low effort forks. After we deploy to all good chains and pump to $20 we could consider seeking an investment from a leading VC firm to get more composability with ~100 portfolio companies and attention of affiliated security guys. We already are affiliated with Yearn, though.

2 Likes

Adding and exclusion for forked yearn vaults should be simple :slight_smile:

So there are private repos that could not easily just be forked. If someone forked the UI and the contracts, it would definitely give them a good start. They would also though need an API, currently all Pickle UI derives it’s numbers from the API. This is required for the UI to function properly. In addition the Harvest bot is also in a private repo. This harvest functionality would be needed in order to execute a pickle clone properly. While it may sound like just calling the “Harvest” and “Earn” functions would not be particularly difficult, we are always coming up with new ways to reduce or prevent MEV. A Pickle clone would need to recreate this as well to stay competitive.

I think what we do currently would help competitors some, but not as much as what you would think. Contracts are the big ones, but those will be open source no matter what as they are published on the blockchain. The blockchain is permanent, so even if someone forked it, and we told them to shut it down, it is too late. It is already on the blockchain. We could shut down their website? But people could still use the contract itself. I believe this is a rabbit hole that we already have sufficient safeguards against without going overboard.

We’re also going through an exercise to remove 90% of the core logic out of the -ui repository and up into the api. I’m not sure how long this will take, but, it could mean that the only way for the UI to function in the future would be via an API where everything is a black box and not easily copied at all.

2 Likes

I love that. Sounds like a good protection from low effort forks.

How about changing the license for contracts? If the DAO is not too happy with restrictive licenses, changing MIT to GPL sounds like a no-brainer.

2 Likes