Viewing profile — hnmullany

I was a private equity analyst in the 90's and we owned Berol - one of the classic American pencil companies. Their classic wood pencils were made the exact same way - batch orient…

I worked in private equity as an analyst in the 90's - we looked at an ambulance service that was up for sale - the underlying operating margins (before you load on debt) were pret…

I worked in private equity in the 90's and looked at one of the big regional ambulance service companies that was up for auction. The underlying operating margins were fairly reaso…

I think the strength of the "vouch" is understood to be much weaker in Silicon Valley - it's more like - I've talked to this guy for 30mins to an hour and I can vouch that he (or s…

On every tech that would have made HTML5 an effective application platform (databases, advanced AV, animation) - Apple has been 5-10 years behind Google and Firefox, and in many ca…

Flash didn't disappear when internet technologies got better. Flash disappeared because Apple refused to distribute it on iPhone - partly because Adobe couldn't effectively shrink …

You might want to look up the value of those "fat pensions" - a lot of them are substantially less than US Social Security. e.g. in Ireland, the state contributory pension (aka wha…

I don't see any on data on nutrition in that reference. Organic grown food tends to be higher in anti-oxidants and secondary metabolites, possibly because the plant immune system i…

Organic grown food is more nutritious and I'm glad we have a significant percentage of farms who have gone organic, but the yield gap is very well documented. Organic fertilizers r…

I think he peaked in the aughts - his latest stuff is ho-hum.

I came across one of these in 2018 with a "hot" open source company raising a Series B. An impressive star ramp (about 300% YoY growth) before the (high-priced/competitive) raise a…

Battery chemistries for grid storage are moving toward commodity elements and rapidly reducing reliance on rare earths. Sodium and Iron/Air batteries can take over fairly easily.

Just a note that 46,000 is a 2050 target - which covers current demand AND all the growth in electricity and heat demand over the next 25 years.

Solar still makes sense economically in the US without explicit subsidies - that's why it is still getting built. But the Trump admin is also with-holding permits and cancelling lo…

Heroku got a lot of attention and funding within Salesforce at least for the first few years - they grew from about $1M in ARR when they got acquired, and I think they peaked at ar…

Nuclear capacity additions in 2025 were about 1% of solar additions - there is no comparison. Nuclear capacity: +2GW in 2025 ( https://world-nuclear.org/information-library/country…

The cheapest solar auction to date was $13 per MWh (middle east) - so utility solar in the best regions is already very very cheap. When you add 4hr batteries, it's still competiti…

That wikipedia article needs to be updated for the last few years. 2025 was the first year where coal generation declined YoY. Nuclear capacity additions in 2025 were about 1% of s…

Mono-crystalline silicon - which is now the dominant technology - is a pretty clean, but thin film PV - which is on the wane - had high heavy metal content. Good news.

Coal generation production in China did decline in 2025 vs 2024 - but that was the first year for it to happen.

It's two orders of magnitude difference between renewables and nuclear though. China commissioned about 3GW of nuclear and almost 300GW of solar last year.

I think one of the phenomenon that people haven't mentioned is that the question space was heavily colonized by 2016. I was one of the top 30 or 50 answerers for the SVG tag on SO,…

What are the bets that the NSA has been encouraging Microsoft to keep shipping this?

Everything from Stephenson after Anathem is an unremitting slog. He needs an editor who won't back down from telling him he needs to cut a third of his pages.

I did a meta-analysis for my masters thesis and I was convinced that 2 of the 39 papers in my dataset had fraudulent data.