rob_lh

Not the LLM kind :)

Adding some personal analysis to your point, the last time I looked at plastic market research in detail, the two biggest markets by volume and revenue were packaging and insulation, which I believe were dominated by thermoplastics. Those are the big ones if you want to make an impact on plastic usage.

Also, pulling from memory that may not be quite right, but I recall roads taking a substantial amount of polymer additives and that tire degradation is a major source of microplastic exposure for humans. The tire problem is poised to get worse with EVs being so heavy and accelerating so hard that the tires are bigger and wear faster, but I'm not aware of even any promising research there besides more bio-based feedstocks to improve the sustainability.

Having worked in polymers and seen the development of some recyclable, 3D-printed thermosets like what's proposed here, I think this is a fair take. On the whole, I'm glad to see other researchers continuing to research the space.

There are a few big challenges to be managed. One is material diversity—it's cool they got it to work with one monomer (most thermosets are two that you mix together), but it's a long road to showing the process works well in just one application let alone. To make a substantial impact, the process would have to be suitable to a bunch of different applications, likely requiring different material properties that would require many different monomers.

Then we can talk about value. While the value of fully recycling plastic can be managed as an externality by government taxes/fees, the big question is how much it actually costs in energy, time, money, and material waste that somebody has to pay for—there's no free lunch here, and it's rolling all those costs back up into the product. I have been secretly hoping that automated waste sorting and excess solar capacity could be used to run such waste management processes and close the business case, but energy demand and prices seem to only be going up due to AI. Still, it's good there's another example of it being possible but a long road to really reducing single-use plastics.

An excellent question. The long-term consequences are the realm of analysis, not politics, which is why I wrote this article. Those ideas shape how politicians make decisions and which issues to prioritize. To understand how your vote may impact foreign policy, study a candidate's knowledge, attitudes, and experiences of our relations with other countries. Specifically study their attitudes towards security guarantees, their language towards allied nations (especially the ones failing to meet their commitments to NATO), and how they respond to Russia's nuclear threats. If they are an incumbent, consider signing up for https://www.govtrack.us/ to get personalized updates about how your representatives vote on these matters.

>How is this problem anything resembling actionable, for normal people?

Vote

I'm the author, and I'm always deeply flattered when people submit my stuff. Thank you! Happy to chat more, as there was a ton of research that went into it, most of which didn't make the final cut.

The US government is dedicating something like $39B to the silicon business—is it using it wisely? Part of a series on the modern silicon foundry business, this chapter sets up the problems in the US supply chain circa 2022 that the CHIPs Act could address. The result is a set of strategic questions that will help grade the program's success.

People seemed to enjoy my deep-dive on 155mm, so I'm sharing another on leading-edge semiconductor manufacturing, an industry I actually have some experience in. Please let me know if you enjoy it and want more.

Done, forgot I hadn't included it here. For simplicity, I can be reached at rob.lheureux.3@gmail.com.

I'm a huge Dan Carlin fan. I started my career as a mechanical engineer, and I have a deep appreciation for chemistry and materials science, as everything rests on that. Most of the early transistor legends were chemists after all. Hence, I wanted to go a little deeper on the technical side with this blog, focused on the here and now.

FWIW I have spent a few years leading GTM strategy for startups, am consulting on marketing for startups, and thinking about getting back into a startup, so I'd be open to a conversation about what you're trying to do.

Fundamentally, marketing is about 2 things: awareness and engagement. Awareness is generally about socializing the problem, how current solutions fail, and what the gaps are, in a way that directly maps to your product. Each market is different enough that you have to appreciate the context: who has this problem? Do they know they have this problem? Have they budgeted for a solution (i.e. are they in-market or are you trying to convince them to spend money on it)? The classic b2b trap is everyone telling you it's a problem, but it not being enough of a priority that they actually spend time fixing it. That's often more indicative of a product problem than a marketing problem, but it shows up most in weak GTM. Validating that in your early GTM is key.

Similarly, engagement is context dependent, but I like to think of success as moving them to the next step of the funnel. Do they believe they have a problem? Are they looking for solutions? How are they going to make a decision about which to bring on board? What information can you offer to frame your offering in the best possible light vs. their other options (including no solution)? Build your collateral to that.

I'm delighted you enjoyed the piece so much. Honestly, I expected to get 20 views on the whole thing, as I felt like a crazy person researching this area for like 6 months. That collapsed in the belief I probably can't do much to help, at which point I just wanted to get it out there for anyone else in the wilderness.

I did speak to a few missile manufacturers facing the same thing, including DOD's lack of urgency, which I interpret as Congress' lack of urgency. In the nightmare situation of a hot war in the Pacific, that attitude would change on a dime, and I still think Operation Warp Speed is the primary example of how we would respond to direct threats. Similar to COVID, I would much rather deter such a situation with bold action now than wait until we're certain we'll need them, which will cost many more lives.

And I've heard dubious things on Replicator too, but thankfully, smart defense private money is going to truly disruptive dual-use tech that is autonomous, resilient, and ultimately attritable. That combination is the best answer for democratic countries that care about soldiers' lives vs. authoritarians sending human waves.

This was an idea I considered, but I'm not enough of a metallurgist to know how feasible it is. In a crisis, I imagine we would try lots of things, but it's easiest to start by just getting out of our own way with proven tools and processes. After all, they're co-designed with weapons systems that are well understood and highly effective, and we'd be replacing all that with entirely new systems in a conflict. It's messy.

Arghhhh super embarrassing. Corrected now, I stared at this for way too long and was eager to just get it published.

Author here. In short, no, I wasn't surprised, I just wanted to cover it. I worked in 3D printing, albeit on the polymer side. On the metal side, there's been a lot of progress and 3D printed thruster nozzle assemblies are one of the key advancements SpaceX has made. It would never be a high volume process, but I was curious if you could get the kind of steel you needed for "surge" demand when the price matters less than total throughput.

I am skeptical the US would really need it in a war, but our allies certainly will. The US has air superiority and integrated combat arms. Ukraine shows how essential artillery is without at least air superiority, and they have just a few years of integrating tanks and other vehicles into actions with their infantry.

Just waking up to find this on HN. The points you make are completely correct. I tried to keep the focus on just US production, as I was specifically interested if there was anything I could do to help. Splitting up the process was something I considered, but didn't have the bandwidth to put into knowing when/how you split it. In a crunch, I'm still not sure if it would be best to split it or to just do what's usually best and co-locate all the machines (or some combination of both). Similarly, I couldn't get good answers about how many shifts the existing facility runs. I expect it's 2, and in a war, it's easier to staff the 3rd shift (or automate it in a new facility).

The contract length part I covered, and I'm convinced that's the main issue.

This is a project I've been grinding on since my time with Intel. It's in an admittedly weird space - productivity software for a leisure activity - but I got so frustrated with how hard it was to give a personalized recommendation for what a gamer should buy. I welcome HN's feedback on what they think of this tool.

We have a really fine-grained view in their for CPU selection that we just launched too (when building a rig, hit the FPS tables while selecting a CPU). I'm curious for feedback on that and if it makes sense for understanding how CPUs perform relative to one another. If people like it, we'll likely do the same for GPUs.

This is a concept I have been stewing on for a while and hadn't seen it discussed much as a potential solution or application for drone technology. It's very light on engineering, as I wanted to go for breadth and put as many options on the table before doing heavy analysis of the tech required and system constraints.

I welcome the feedback of this community - I'm new to avionics and much of the image recognition technologies I was extrapolating from existing applications.

And the Russians dreamed of going to the moon but that's the rub with consumer products - the market has to accept them for it to be an accomplishment. Google thought they'd found the magic recipe and pitched it as something for everyone that was a huge breakthrough. Being first with features or flows doesn't matter at all. The only thing that matters is making it relevant to users' lives, which all parties failed at. Circles is a bad model for a mass market.

Circles for different messages to different groups, as well as the universal profile for logging into all your favorite Google services. It's not my fault these were harebrained ideas, but they were presented as revolutionary. The ideas are still arguably revolutionary (Slack delivers where Wave failed) but Google's execution was lacking. Same fundamental issue facing so many of their other programs.

They were absolutely pitched as huge ideas that would change everything we know about email and social networks. Bearing the scars of their public failures, Google started calling new projects moonshots to shape perception of the programs as chasing ideas, while preemptively deflecting any criticism of their readiness or desirability to large markets. They are not materially different from moonshots except in presentation.

I was reading the interview with Jimmy Iovine today and everything he said about curation rang true to me, but when I listen to Radio One, it's too much Top 40 and talking about the music. I love Spotify because it's just the music - just get to it, let me find whatever the curator saw in the music instead of explaining it to me. I get so mad listening to people in a studio talk about music they love or just filling time that could be another song with what is most likely an ad. It's my single biggest gripe with SiriusXM too, but at least I've been really happy with their curation.

I don't know if you're quoting or referencing something, but that's fantastic.

Indeed, that is the most salient point. I like your suggestion of looking at many attribution models to assess true performance. This seems especially relevant in larger organizations where there is an incentive to game the budgeting process - presenting just the best possible case is what marketing is all about but can lead to skewed perspectives of success.

Isn't that an argument for assessment on a campaign basis? I'm still wrapping my head around his calculations. It looked like he included the cost of the ads prior to the reporting window in his multi-touch model. That's a fair way of measuring the true cost of user acquisition, but if the ads you are paying for today won't pay dividends until you're out of the reporting window, it would skew the ROI.

I'm not sure if this is the same thing as what you cited, as my argument is about the cost association during the reporting window as opposed to the future revenue association. I can't figure out a way to back those out in real-time (i.e. in enough time to make adjustments to the campaign) though.

For your first part, immigrants displacing jobs is a news story pretty much anytime it's not western European. These stories are common place with Hispanics, eastern Euros, Middle Easterners, Africans, etc. The difference here is the perverse corporate (read: institutionalized) incentives to benefit from displacing those jobs.

Your second parts about the reality of employment in the US is dead on. There are some companies (or at least some organizations) that benefit from grooming employees and keeping them engaged, but most pretenses of commitment are embarrassing lies. Reid Hoffman's spiel about "tours of duty" strikes me as the most suitable framework for the current corporate job market.

That's some awesome data, thanks for sharing! I have to imagine geo matters here too. China, Latin America, and Africa have huge Android growth (though even then a huge chunk must be AOSP) but I would guess it's mostly older OS versions coming from the smaller OEMs that are spawning so much of the growth. Who knows if those would ever be updated, but it's less relevant unless you're a pervasive, global app (like FB or Gmail).

I would hope this to be correct, but I didn't see Google address it at all. A lack of transparency is a great way to foster FUD and a bad perception of what could be a great piece of tech. I get that they're building the marketing hype first, so hopefully it will be forthcoming.

I am curious if this will be a source of friction with app developers. If Google is scraping app data usage for both content and context ostensibly to extend the benefit of that app's data and utility, wouldn't that make it considerably easier for them to clone the app and improve upon it?