The infallibility of official, open-source and audited software

How to run multiple Spacemesh nodes in MacOS, using a cloud provider to create your postdata

After two decades, I recently migrated my blog from WordPress to Jekyll, a static website generator. I&#39;ll write another article soon on why and the process, but as I continue to tweak the site, I wanted to quickly post how to do something that was surprisingly difficult to find.


How to display category and tag counts in Jekyll

I own a Specialized Roubaix bike that came with a 4iiii power meter installed on the Dura-Ace crankset. I also have a Garmin Edge 830 bicycle computer, which alerted me the other day that my power meter batteries were almost empty. This article explains how to access and change the batteries.


How to change the battery of a 4iiii power meter

One of the use cases of smart contract blockchain networks that has always excited me most is programmable, composable finance. In that domain, an interesting project I’ve been paying attention to recently is @pendle_fi — Pendle Finance.


Pendle Finance

International attitudes

As a bag addict in continual pursuit of the optimal setup, I&#39;ve recently updated my daily carry to include a Waterfield Essential Cross-body bag.


Review of the Waterfield Essential Cross-body bag for the iPad Pro

How Americans spend their money


THREAD — On understanding the fundamentals. When I was at university, I took a three-semester course on radio-frequency (RF) engineering, which taught me the profound importance of understanding things from first principles.
When two electrical systems are interfaced, each one, looking towards the other, sees a certain “impedance”, or resistance. Only when the two impedances are “matched” can the combined system be stable.
An important activity in RF engineering is, therefore, the design of a third electrical system that sits in the middle, and matches the impedance seen by the two interfacing systems, thereby providing overall system stability.
So the first semester of the course was spent learning how to design matching circuits, using a procedure involving the drawing of “stability regions” on intricate circular graphs, called a Smith Charts, printed on paper.
Once you had properly drawn a stability region, you could then be sure that as long as your matching circuit’s characteristic parameters fell inside that region, the resulting electrical system is guaranteed to be inherently stable.
The notion of bounded inherent stability struck me as elegant, similar to the Shannon Sampling Theorem, which guarantees the reconstructability of an analog system, as long as you’ve digitally sampled at twice the highest frequency.
(As an aside: Only those who understand the Shannon Sampling Theorem will understand the incredible scam that selling “oversampled” compact disk players was, back in the day.)
Back to the story, the second semester opened with a surprise, when Dr. Hertling told us that having spent three months learning to design circuits with Smith Charts, we would never use them again! In the real world, we’ll use software!
So why was one third of the three-semester course spent learning a technique we’d never use in practice?
Because it taught us the first principles. It taught us that there’s boundaries to design values, if we want a stable system. It gave us the intuitive understanding of RF design to recognize when software output is likely to be wrong.
The importance of understanding things from first principles was one of the most valuable lessons I took away from university, and that course in particular, and which has proven transferrable to so many areas of life, including investing.



First principles


Over the past few months, I’ve been going down the DeFi (decentralized finance) rabbit hole—concluding in participation in the recent @mStable_ auction of the $MTA token. This thread summarizes the notes I’ve taken along the way.
For context, I’ve been an investor all my life, learned about Bitcoin many years ago, but only recently have started taking a more detailed look at how projects work in the Ethereum network. I became interested in DeFi seeing some investor friends I trust making seed investments.
For example, as an early participant in CeFi products like Lending Club, @AaveAave ($LEND) looked promising. And seeing the SEC problems Abra had offering synthetic assets, a decentralized approach like @synthetix_io ($SNX) looked promising.
DeFi applications (dapps) are like centralized finance (CeFi) platforms in function, i.e. exchanges, lending markets, savings accounts, etc. Unlike CeFi, however, DeFi platforms are permissionless, like the Bitcoin network.
For example, whereas my son isn’t old enough to open a bank account, he can deposit USD stablecoins in a DeFi app and earn interest, just like a savings account. And I can exchange tokens on Uniswap without having to upload my passport, and justify where I got my funds.
Using DeFi apps requires “connecting” an Ethereum wallet. There’s a growing number of standards for this, including the well-designed Coinbase Wallet for iOS/Android. But the universal standard is the browser-based Metamask wallet, available for Chrome on OS X.
Every interaction you have with a DeFi platform requires a confirmation in your wallet, which maps to a transaction on the Ethereum network, and therefore has a fee associated with it. The fee is dynamically determined, and can get expensive when the Ethereum network is busy.
As someone with a background in product, I was quickly attracted to the @mStable_ platform, for its focus on UX (compared to something like Curve.fi, which feels like the cockpit of an old airplane.)
Shortly after I’d finished exploring their product, mStable announced plans to auction off 2.6M $MTA governance tokens, of the eventual total supply of 100M coins. So I decided to participate.
The mStable team decided to conduct the $MTA auction on the “Mesa” platform, a decentralized exchange built on the Gnosis protocol.
In the days leading up to the auction start, people would enter publicly-visible bids. You would bid on $MTA with the mStable stablecoin, $mUSD, which you could “mint” in their dapp, using another stablecoin like USDC.
So the process was: (1) Mint some $mUSD using $USDC at mStable, (2) Deposit that $mUSD on Mesa, (3) Make a bid on $MTA using $mUSD, (4) Possibly change your bid as the order book changes, (5) Hope your bid gets executed during the auction.
Few people in the mStable community Discord chat seemed to understand the mechanics of how the auction would be executed, since the Mesa “price solver” is complicated, “blocks” of bids are processed each 5 minutes, and priority is given to large-volume bids.
Speaking of the Discord,  the quality of the conversations was very high a week prior to the auction, but devolved dramatically in the final days, as large numbers of speculators appeared who knew nothing about the project, but smelled a profit opportunity.
The notional price declared by mStable was $0.15/$MTA, but two days prior to the sale, the order book already showed a clearing price of $0.50, i.e. there were bids for 2.6M coins at a price of $0.50 or higher.
The clearing price increased dramatically when mStable annnouced an investment from @SBF_Alameda’s FTX research arm. In the interest of token distribution “fairness”, I’m not sure whether the timing of that announcement helped or hurt the cause.
I decided to wait until just prior to the July 18, 14:00h UTC start to place my own bid, and placed a bid of $1.50 about 30 minutes prior to start, given the at-the-time clearing price of $1.20.
The price action around the launch was enough to cause the gas price on the Ethereum network to spike. While it cost me about $0.70 in gas to place my order, it would have cost me $180 to cancel the order once the auction kicked off.
Mesa/Gnosis had some technical hiccups, but the auction eventually executed, and the clearing price ended up at $1.80, meaning my bid didn’t get executed. Shortly thereafter, $MTA started trading on the Balance and Uniswap exchanges (both are DeFi apps), and the price approached $3.00.
Many community members expressed dissatisfication with the fairness of the auction, when it was seen on the network that a very large percentage of the tokens were purchased by a small number of actors (around 10 whales!) My view is indifferent on that.
Post-auction, an alternative to buying $MTA outright on an exchange, is to earn it by providing liquidity to a “pool” at @balancerlabs. You deposit equal parts $mUSD and $USCD, and each week receive your pro-rata share of distributed $MTA tokens (50K per week, I believe).
I still don’t understand the exact purpose of these pools, but do understand that there’s a relationship between the level of interest in staking coins in a pool (and earning $MTA), and the yield offered in the mStable SAVE product (paid in $mUSD).
A word on risk. In addition to all the normal investment risks one has to consider, investment and participation in DeFi platforms exposes you to technical risks, as these platforms operate on Ethereum “Smart Contracts”.
The code in these contracts is public, and as the value of a DeFi platform increases, so does the potential value in attacking it. The funds from many contracts have been drained in the past, due to discovered software bugs and expoits.
So when thinking about things like position sizing, remember that this is still experimental technology, subject to exploits. (Over time, the security of smart contacts has increased, and should continue to increase.)
It’s been a lot of fund learning about this stuff, and I look forward to learning more. I’m bullish on this whole DeFi space, in large part due to the innovation of what’s being built, but also from a user experience perspective.
As the KYC/AML onboarding process at CeFi institutions ever grows in friction and intrusiveness, the ease of permissionlessly interacting with DeFi platforms is refreshing, and is a force towards global financial inclusiveness.

