jfan

Thanks. I love the business and have a tiny starter position but would love to own it at lower prices.

Tvk.to and ipco starter

I recall you were quite bullish at these prices, did you change your mind?

Definitely don't understand everything around the cryptography lingo, but found some useful tidbits

Q1 Shareholder letters from Wintaai and WinRe TDLR: 1) Wintaai purchases Topa Insurance Group 2) WinRe is raising capital for those interested. From the NAIC website #18031 (Topa) Assets $143M Liabilities $85M Capital Surplus $59M Homeowner's multiple peril $73M 3 year Net Income 2025 - $18M 2024 - ($11M) 2023 - ($32M) 2026 Q1 Wintaai Financial Highlights March 2026.pdf 2026 Q1 Win Re update.pdf

All out of Cameco, closed off my margin, taking a breather to figure out where to reinvest

Mine too. Got super lucky here. Had an intuition about the poor sentiment at the time, gained a superficial understanding of the dynamics, sold to recover the house money too soon, but rode it until it starting trading like a meme stock. The price-implied expectations are high here, and retail friends with no working knowledge (investment or mining) started jumping into this name, so I think it's time to leave the party.

Sold 1/3 of my cameco shares.

My back of the envelope Stonetrust valuation

LRIE and LHC have a bit of an accounting peculiarity that is a bit difficult from the outside to pick apart. His 2023 letter describes that LHC derives its revenue from LRIE's Gross Written Premium (ie before reinsurance ceded). LHC's fee to LRIE is 20% of GWP. His 2024 letter describes an accounting rule change in revenue recognition, where LHC's fee is still based on GWP, but it must be recognized over the policy term. This would create a deferred revenue liability item on the balance sheet. This would suggest to me that if they have multi-year policies, that the cash earnings will be greater than GAAP net earnings (assuming minimal working capital). Without the cash flow statement or balance sheet, I can't interpolate the exact number. In addition, it had assumed some premiums in 2023 that had lower Attorney-in-fact fees. Using a very rough approximation with the disclosures given, the LHC fee % of GWP were: 2023 - 11.2% 2024 - 11.5% 2025 - 18.9% It's not quite their originally stated 20%, but it is close, which I assume might be related to not issuing multi-year policies (my guess). Finally, in 2023's letter, Francis gives us his forecast for LRIE and LHC as shown below: As a comparison to actual results: LRIE GWP 2023 - $45,945 2024 - $163,416 2025 - $204,165 LHC Total Revenue / Operating Income / Pre-tax Income (inclusive of investments and financing/other expenses) 2023 - $5,141 / ($3,798) / ($3,792) 2024 - $18,892 / $4,127 / $4,249 2025 - $38,515 / $17,526 / $18,341 Actual results lag his forecast, primarily driven by less GWP (?more competition). If I adjust their 2025 numbers to "more reflect" their cash earnings, LHC made the following: 1) 20% x $204,615 = $40,923 2) less operating expense of $20,989 = $19,934 (operating margin of 48.7%) 3) add net gains on investment of $8,178 = $28,112 4) less interest expense, bad debt, and other income of $7,362 = $20,750 (pre-tax income) 5) assuming a total Florida tax rate of 26.5% (I might wrong here) = $15,251 (after-tax income) What is this worth? a) 10x cash earnings ~ $150 Million b) 15x cash earnings ~ $228 Million c) 4x revenue ~ $163 Million Wintaai's share (90% ownership) a) $135 Million b) 205 Million c) $147 Million * on a $9 Million Wintaai America's initial investment There are 5,506,360 Wintaai Holding shares at year end 2025. This would peg the per-share value of LHC between $24 to $37/share. *As per Francis, to add to more accounting/auditing peculiarities, LRIE is consolidated at net asset value into Wintaai holdings but this makes no economic sense because the net asset value represents policyholder contributions (not Wintaai equity).

Francis also started a reinsurance business as well. Here is the annual letter. Francis said the returns were only satisfactory for him and that he could do much better. 2025 Win Re Annual Update.pdf

Could always consider eq bank or goeasy.

Thanks. I only own CSU and TOI due to research time constraints and capital. Together they are ~ 11% of my self-managed portion of my portfolio (4% in CSU and 7% in TOI). I weighted it more to TOI because of its smaller size relative to CSU, and been pretty consistent since 2019 on their EBITDA/Average Adjusted Invested Capital of ~ 20-25%. I appreciate your thoughts on Lumine. Carve-outs seem much more complex, and due to their size, may limit PE competitors vs TOI's traditional acquisition strategy. Not sure I can wrap my head around the deal complexity, and integration risk.

How did you divide up your position size among the three entities if you don't mind me asking and why?

Family of 12-year-old Tumbler Ridge shooting victim files civil claim against OpenAI https://share.google/TfCodgtfr0iYewBuF

In my day job, I have a number of paper-based forms associated with quasi-governmental regulated entities, that require photocopying or faxing, coupled with logging into these entity portals, to manual do data entry in order to complete them. I'm not these entities primary client but a necessary cost center whose user experience and time is not considered something valuable. So this automating this with a better user interface is not really on anyone's priority list. So with no coding experience, I signed up for Claude, to build out an interface that solves this minute problem that others also deal with on a day to day basis. It is remarkable what it has built, but there certainly is a learning curve, requirement to download other pieces of software (eg database management, etc), in order to make this feasible. It's been about 5 days, and Claude has create for me a web login with 2-factor authentication, ability to adjust my profile/password, a clickable user interface. Hitting some snags getting it to print onto the designated fillable pdf. Anyways, the point(s) of my story is that AI coding has really unlocked people's ability to address various problems that could have a software solution, where previously it was impossible. I can only imagine the productivity gains software engineers might capture with these tools today. That said, the fear that software engineers will be eliminate is likely overblown. With my experience, it would be hard to imagine that my vibe coding product would be enterprise ready without any software engineering review and ongoing support. Furthermore, for truly mission critical workflows, I would hardly trust a non-technical person to vibe code and replace such legacy software. Finally, despite this simple problem, I'm still hacking away at it for 5 days without a functional product. I would imagine that the average person (non-software engineer) would have little to no interest in spending their time doing this (eg my adult relatives/in-laws, all were yawning at the dinner table as I blabbed on about my little side project). With the problem surface area being pretty much infinite, I'm leaning towards that fact that AI will unlock some productivity (eg more prototyping, faster product releases) and the key skills to develop in this brave new world will be creativity, ability to empathize and understand other people's problems, critical thinking skills and continuous learning habits.

This is so sad. Thank you Guy for such a poignant letter. Our thoughts and prayers are with you and your family.

https://www.darioamodei.com/ People may enjoy reading his thoughts here. One essay on AI benefits and the other on AI risks.

The floor will probably be the cost to run the network. A very simplistic look at it is: 1) network electricity cost Plus 2) cost to purchase the mining rigs to run (ignoring the other infrastructure and labor costs). Below is Cambridge's power estimate. https://ccaf.io/cbnsi/cbeci At 97.32 TWh of annualized usage, at $0.05USD/kwh, this equates to $4.8 billion USD. At 180.77 TWh, this equates to $9B USD. And at 313.86 TWh, $15.7B USD. Below is the tiered cost price/TH of the mining equipment across different efficiency tiers. ASIC Price Ihttps://data.hashrateindex.com/asic-index-data/price-indexndex Coupling this, with the total hashrate in the network, as seen below: https://www.blockchain.com/explorer/charts/hash-rate The cost to replace all the mining rigs with the newest, most efficient version could be $14.3B. Most of these rigs last ~ 3 years, longer if they can be maintained well. So the annual "cost" is ~ $4.8B. With the current transaction fee as a % of the total block reward (reward + fees), the miners should "create/earn" 165K of bitcoins in a year. Bottom line: 1) Bear case ($4.8B + $4.8B)/165K = $58K USD/bitcoin 2) Base case ($9B + $4.8B)/165K = $83K USD/bitcoin 3) Bull case ($15.7B + $4.8B)165K = $95K USD/bitcoin. * many assumptions baked in here, no true insight into true electricity costs, nor distribution of active rigs in the network

Instead of debating the merits, morals, and economic theories of bitcoin, it might be useful to look at what the network is actually doing, since the data is visible to all for analysis (and for those with software skills, verifiable). The following charts are from Block. https://www.theblock.co/data/on-chain-metrics/bitcoin This shows the transaction counts on the network. The counts are not going up exponentially, but meandering around 10 - 12.5M over the last 3 years, up from 7.5 to 10M 3 years prior to that. Block time has been stable at ~ 10 minutes. So likely more capacity to process more transaction volumes. The number of active addresses doing transactions on the network has plateaued at ~ 20M each month. The number of new addresses coming online with a transaction has stepped down from prior years, averaging ~ 10 million each month over past 1.5 years. So in aggregate, this suggests that # of active layer 1 users, has decelerated. The $USD denominated transaction volume is slowly climbing, but far lower than in 2021 and 2022, but this is likely similar to other markets during this time period. There is alot more transaction volume ($) relative to pre 2020. This volume is all on-chain, so it doesn't include off-chain activity such as derivatives or BTC-related ETFs and stock activity. Not on this picture, but the # of addresses with >0.1 BTC (less than 1 BTC) and > 1 BTC (less than 10BTC) have plateaued since 2023 at ~ 4.5M and 1M addresses. Those with > 10 BTC (less than 100 BTC) have plateaued since 2017 at 150K addresses. Those with > 1000 BTC have plateaued since 2019 at ~ 2K addresses. ~ 70% of BTC supply has been active in the past 5 years, 50% in the past 3, and 35% in the past year. This suggest the turnover is pretty slow, at least, relative to stocks, which I believe the average holding period is < 1 year. These 2 graphs show the daily $ volumes of the ETFs and On-Chain activity, with the ETFs dwarfing the daily volume on-chain. So it appears to me that the ETFs have reduced the on-chain activity which might be a partial explanation why the network fees have stagnated. The current average fee to transact on-chain is ~ $0.65 USD/transaction. This is cheaper than my "discount broker". It seems to me there is a choke point at the on and off-ramps from the fiat world to the crypto world.

Csu and toi.v

I think the global electricity rate is too low GlobalPetrolPrices.com

Selling 15% of my cameco shares after a nice run up over the past five years.

Topicus

REQ-Acquisition-driven-Compounders-July-2025.pdfThis is their new 2025 slide deck updating their 2023 presentation