Radix deep dive…Part 4 of 4
Community
Spreading itself across all the relevant social media platforms, Radix is making a serious effort to build a community around its project. With an active Telegram channel, Discord server, Twitter account, subreddit, and podcast (The DeFi Download), Radix keeps its community informed on all updates surrounding the project, in addition to providing insightful content about industry topics in general. The community itself is strong among individuals, with node-runners building reputations on Discord to eventually solicit delegated stake. Community members even contribute to the research arm of Radix by running nodes on the current Cassandra testnet with founder Dan Hughes live on his Twitch channel.
Radix is building an extremely active developer community focused around its Scrypto programming language, developer component catalog, and future dApp ecosystem. After an extensive review of the Radix whitepaper and queries via its search engine, the one thing Radix failed to fully elaborate on is its governance structure. It appears that developers will be at the core of making changes, similar to on-chain governance of blockchain protocols where developers first propose a change through a code update, then each node votes on whether to accept or reject the proposal. On a side-note: Although the forums and communication channels are great ways to find answers about the project, be aware that much of the social media has no shortage of cult-like members voicing their undying support for the project.
Did Radix actually process 1.4 million TPS?
In a controlled and open test environment, yes. This is how many projects demonstrate their transaction throughput. All of Radix’s trial code for this scalability test can be found on its Github. It also explains in a two-part series on its blog how it built and deployed the test. The first part is a high-level read, digestible enough for those with a good understanding of blockchain and distributed ledger concepts. The second part is a deeply technical explanation that dives into the code and logistics of running such a test.
If you’d like to save some time and not navigate out of this site, here is a brief summary of the salient points: Radix replayed Bitcoin’s entire 10-year transaction history (up to June 2019, when the test was conducted) in less than 30 minutes. Because Radix is not a blockchain and Bitcoin groups transactions into blocks, the Radix team converted Bitcoin’s dataset to Radix transactional entities known as Atoms, a dataset it could process. It conducted the test utilizing 1,187 Google Cloud nodes distributed across 17 countries.
Back in 2019 (before a larger than 1 billion dollar market cap), Radix explained that its fundamental limitation is “how much money we are willing to spend on running these tests.” (In other words, it thinks it can handle more, it just has to keep paying to do more tests to prove it.) In addition to providing the test code, Radix’s Github also contains the tooling that allows anybody to set up a test network in Google Cloud to verify the strength of Radix’s tech for themselves.
In short, Radix legitimately did what it says it did. Although I don’t hold a Ph.D. in computer science, nor am I pretending to be smart enough to understand everything that Radix explained it did for this test, I am privy to the power of peer review. The power of highly intelligent trolls and internet warriors fighting the good fight to expose false claims and forever uphold the purveyors of such venomous duplicity. Hence, if Radix lied, some credentialed expert would have called them out within the last two and a half years, boosting their reputation for disproving the claims of a billion-dollar entity. But no one has, the test code is public, and Radix is a top 70 project by market cap. Only one question remains: Will Radix deliver the same throughput when its final, sharded mainnet version is released outside of the testnet environment?
Centralization and Mainnet Genesis
The Radix mainnet was centralized for all of two weeks. The mainnet initially launched under a two-week “bootstrapping” period where Radix-controlled nodes secured the network, giving the community time to fairly start swapping EXRD for XRD, staking assets, connecting nodes, and registering validators. After that initial two weeks, known as Epoch 1, the Radix Protocol became responsible for automatically choosing the top 100 validators — based on the amount of delegated stake — to secure the network for each following epoch thereof.
That set of 100 validators then participates in consensus until the end of the epoch, and then the cycle repeats. When the fully sharded Cerberus hits mainnet under the Xi’an release, there will be no limit to how many validators can participate. Because neither Radix nor any of its underlying entities have control over the validator selection process or who can use the network, Radix is both fully decentralized and permissionless.
The Critique
If you made it this far, you’ve probably thought more than once about investing in EXRD for Radix’s growth potential, or VPNing — if you’re in the US — into Bitfinex and buying some XRD to send to the Radix Desktop Wallet to earn those juicy 11–12% staking rewards. (I did both.) Before you dive in too deep, take some time to consider any concerns about Radix. (This seems like a good place to put a disclosure to counter any FUD rebuttal: I own a disclosable amount of XRD/EXRD.)
Timeline
Radix has a long timeline till all of the beautiful things it is promising operate together (sometime in 2023 assuming no delays). That’s somewhat of a positive given its lack of exposure, but largely a negative. The last thing an investor wants is a niche investment failing to deliver its ambitious products on time, or worse, it launches a dysfunctional product — hence, why it’s not the worst thing that it’s not over-ambitious.
However, what would hurt its humble roadmap would be if, by the time the project’s kick-ass features are fully functional, it’s too late to capture sufficient market share. Etheruem does not suffer from this. I personally don’t think it will launch a sharded Ethereum 2.0 when it says it will, but I also don’t think that’s going to affect anything. Lots of deep-pocketed investors, institutions included, will not pull out from a hiccup in what will be Ethereum’s catch-up to PoS peers threatening a flippening (passing Ethereum in market cap).
Timing
Radix needs to be lucky with great timing. Its 1.4 million TPS is a beautiful marketing asset, along with replaying 10 years of Bitcoin history in under 30 minutes. This gift, however, may also be a curse. Its current Olympia mainnet only processes 50 TPS. Imagine going on a Tinder date and finding out the photos your date posted were actually AI-generated photos of what their plastic surgeon projects they will look like after a full-body surgery. Needless to say, I felt a bit deceived, disappointed and underwhelmed that Radix is currently not what I thought it was.
I almost forgot how I started the previous paragraph: Timing, Radix needs great timing. If the masses hear about Radix via its over-enthusiastic, cult-like community who constantly promotes its marketing assets, and then see 50 TPS is its current capacity, they might also feel deceived, if you will. Investors would ideally want the masses to come after the great features are launched; unfortunately for investors, Radix enthusiasts tend not to do their own research and often post about the great things yet to come as if they are already here.
(Although the Cassandra testnet demonstrated that Radix DLT could support a decentralized Twitter, that functionality still has to be implemented onto the mainnet. Once on the mainnet, by all means, Radix enthusiasts are right to shill post the devil out of the project, until then, Radix mainnet cannot support that kind of throughput.)
Fees
Why leave Ethereum for a blockchain like Solana, Polkadot, Terra, or maybe even Radix at some point? The speed and the fees are the typical reasons. There is no question about Radix’s speed (unless, of course, it utterly fails at launching new versions of its mainnet). Its fees, however, might be a future turn-off. Right now, in an underdeveloped mainnet that settles basic transactions pretty quickly, fees are low. As development kicks in, complex products are created, and developers get a taste of setting their own royalty fees, I’m afraid fees could potentially balloon for network users.
Yearn.finance is an incredibly intricate, automated financial application that, built on Radix, would no doubt use a ton of components or blueprints to complete transactions. All those components or blueprints mean more fees for users, should the developers define a royalty. In one instance, those fees could help prevent spam transactions on the network. In another, users may be less likely to play around on Radix dApps if fees become prohibitive.
Radix allows developers to adjust their fees; however, this is completely at the will of the developer and is not a sure-fire way to solve an issue of rising fees. (Radix Defi White paper, pg. 24.) Why can’t the protocol just add a reasonable flat fee increase for component and blueprint usage types that get split evenly among components and blueprints used, and have a DAO or governance structure dictate future fee rates? This would be a simple, community-driven solution that still allows for an innovative developer royalty system, also protecting users from high fees.
My solution is not perfect, however, and comes with trade-offs. Radix made a trade-off that makes royalties more lucrative for developers and fees potentially higher for users. Whereas my solution would put a cap on developer royalties while guaranteeing fee protection for the user. Under Radix’s current developer royalty structure, developers have more economic incentive to innovate and build better blueprints and components. Under my proposed royalty structure, users are simply protected from a future that is not guaranteed to be out of their favor.
Conclusion
I’m invested in Radix. I stake Radix. But I clearly have my issues with Radix. The timing has to be right. People cannot flock to this ecosystem expecting to build dApps and participate in a thriving DeFi environment, it simply doesn’t exist yet. People also can’t come expecting the 1 million TPS throughput when it will at least take another year to implement the fully sharded Cerberus consensus layer. I get the feeling that Ethereum users are ready to jump ship if Ethereum 2.0 underwhelms with delays; however, they won’t be jumping to Radix, because if they do, they will have nowhere to go, nothing to do. They would probably seek greener pastures in Solana, Polkadot, Cosmos, Algorand, or Kusama (also Polkadot) ecosystems, places that are more developed.
Radix does a lot well… in theory. Test environments and rhetoric are one thing, but supporting the global financial and communications systems is another. The R&D journey that founder Dan Hughes went down to arrive at Cerberus is incredible; however, we still have yet to see a viable platform for developers to launch smart contract applications for mainnet users. The incentives are an incredible selling point that will hopefully drive development. They may also bring the unintended consequence of high transaction costs for executing complex applications. High fees could potentially push users to other established ecosystems like Polkadot or Cosmos, which are currently hosting many DeFi projects that users can interact with.
With competition actively supporting applications for mainnet use and Radix planning to have a full release of its network by sometime in 2023, it may be too late to capture a reasonable share of the market (a big loss of first-mover advantage — for outside-of-Ethereum options). It also would ideally benefit from people finding out about its incredible network stats when they come out. Learning Radix can process 1 million TPS is almost too good to be true, till you do some more research and realize it is too good to be true because it can only support 50 TPS on its mainnet while many competitors can support thousands. Regardless, I still invest in Radix and stake XRD. I believe in good leadership, clear goals, innovative approaches, a quality website (my biggest investment turn-off is a poorly maintained website), and potential.
Thus far, very few projects offer — or plan to offer — what Radix is planning. It essentially seeks to automate services like PayPal, Visa, Mastercard, and backend banking systems. It’s about time the white-collar elite who invest in automation technology for automotive and retail get threatened by automation themselves. (I’m not a populist, though.) If Radix succeeds, there may be no bigger selling point in all of crypto for a completely decentralized, secure, and infinitely scalable network to support virtually any industry’s ledger-reliant operations.
About The Author
Micah Casella
is a digital asset investor with a background in traditional banking. Working on the inside, he has seen the benefit that a little digital disruption can have on the archaic finance industry. He sees blockchain and digital assets similarly modernizing other industries for the better.
