mlyle

I think I'm not speaking of the technology itself but the actual ways it's integrated into society. E.g. what is considered socially acceptable, what is given legal rights of passage on a sidewalk, or how we choose to build society/culture around technical implements.

And I agree it's not immediately obvious, but thinking a little more about second order effects before making a long term durable choice makes sense.

Tech's recent overall track record on impacts to quality of life is not great. And there's a lot of possible ways we can use advances that have different benefits and costs. Falling into them for the long term purely by accident isn't always avoidable, but it doesn't seem desirable.

i.e. it makes sense for society to make a conscious choice about how to trade off different interests for road users and sidewalk users, and to be willing to revise that choice based on experience rather than being stuck with inertia (while avoiding excessive churn and unpredictability).

I think you're not really hearing my point. I'm not speaking about evaluating things just on the aspect of eliminating jobs.

I am saying that I would hate to automate away most joy or most things people find purposeful, even if it increases purpose or joy for some people.

We've already given up a lot of our social fabric recently with tech. I think it makes sense to be conscious and careful to not make it much worse.

It's my guess that delivery bots are a net add to well being.

I think we should automate more of deliveries.

But I think we should also make deliberate choices to keep humans and human meaning in many things, even if they are less "efficient" by some chosen metric.

Automation and mechanization is desirable to the point that it serves core human wishes and desires-- and through some kind of global/social metric, not that it serve individual humans at the expense of everyone else.

Just because something must yield to you doesn't mean you have an infinite right to obstruct it.

It's quite possible (and likely!) for you simultaneously to have an obligation to yield and the other party to have an obligation to not block.

The benchmark announced in the article seems to be around weird edge cases (which are often important) and whether it can prompt the person with relevant information rather than just listening to what they want to do.

Top tier tax professionals can do pretty well on this benchmark (better than agents) but won't hit 100%.

The kind of people spending time preparing oodles of returns would not do very well on this metric, though (especially midway through that pile of returns, as opposed to "think carefully about this trick question").

This is a quote of the article he linked:

https://www.npr.org/sections/planet-money/2026/09/14/g-s1-14...

In turn it links the source: https://www.journals.uchicago.edu/doi/10.1086/685594

Just make it so that the companies that employ the most per market cap are filled, so we can see how empty the rest are.

This is false / misleading. Ukraine's use of Starshield runs on normal Starlink infrastructure and is administered by SpaceX. From the WSJ: https://archive.is/1pzLc

> What is Starshield?

> Starshield is SpaceX’s more secure satellite service for governments. It has two variants: a custom-built satellite network SpaceX launched for the Pentagon; and a lower-tier, military-grade service that connects to commercial Starlink satellites. The more basic service offers encryption features, and allows the equipment to travel at faster speeds than those on a commercial Starlink account.

> What service does Ukraine use?

> Ukraine is using the more basic Starshield service, which uses Starlink’s equipment. Ukraine’s weapons—attack drones, interceptors and uncrewed boats—need to travel at high speeds to intercept Russian Shahed-type drones, bomb the Russian army’s logistics and stall Russia’s invasion force. Many of those drone missions are now failing because of the speed limit imposed by SpaceX, according to the Ukrainian military.

Yes-- my takes on it in the case of a delivery robot are elsewhere in thread. This is specifically addressing a person who said:

> 1A says she is free to protest human workers by getting in their way in public.

And the answer is, no, the first amendment does not let you block people and prevent them from proceeding in their work or other activities.

So that's existed for a long time in the form of OpenFlexure. The microscope is a pretty easy / fun build.

> Do you think picket lines are illegal?

Picket lines that actually block other people from crossing instead of trying to persuade them not to? Yes. Use of threats, physically blocking, or just general impediment? Yes.

Not only does it violate state law (e.g. obstruction of free movement, interference with a business open to the public, false imprisonment, and assault are all offenses here in California depending on the facts), but it also violates NLRA rules.

(e.g. a recent labor case found repeated 1 to 4 minute "inadvertent" blockages unlawful -- https://www.laborrelationslawinsider.com/2018/02/the-nlrb-si...)

The Moscone Act that recognizes picketing federally also makes clear that this is not OK.

> It is not the intent of this section to permit conduct that is unlawful including breach of the peace, disorderly conduct, the unlawful blocking of access or egress to premises where a labor dispute exists, or other similar unlawful activity.

Standing in the sidewalk, and refusing to get out of the way of another sidewalk user.

I mean, I could block a wheelchair user by standing in a cut-out, too. Is that OK?

Is it not better for takeout food to be traveling this way than by car?

Openflexure is really cool-- I have a couple openflexure microscopes. https://openflexure.org/

Building optically precise platforms with bearings and sliding surfaces is annoying and difficult. But it turns out you can do pretty dang well with applying force to bendy metal or plastic.

The article is talking about tumor sequencing:

> You may need to push for comprehensive genetic testing. Some doctors were weird about this and didn’t order it promptly, even after telling me they would. My surgeon, who told me he’d get it done, eventually referred me out to an oncologist. This caused a multi-month delay to getting genetic results, since the first results were done on the surgery tumor, not the biopsy tumor.

The quote further upthread lost the last bit, which is pretty clearly talking about tumor sequencing.

> Strictly speaking, BRCA is a genetic test for cancer (even if it's only a "risk factor"), and it's done on the regular genome, not on the cancer.

Yes... that's why I clarified that I believed we/the article were talking about tumor testing and not BRCA (which is mentioned in the second link I gave).

BRCA is one risk factor for certain cancers. You can get genetic counseling to understand your cancer risk looking at this and other markers.

In this context, though, we're talking about measuring the actual cancer (including the tumor's mutations) to determine what therapies it is likely to be responsive to.

e.g. this: https://www.cancer.gov/about-cancer/treatment/types/biomarke... not this: https://www.cancer.gov/about-cancer/causes-prevention/geneti...

Yes. And the use of money and undue influence of money are the problem-- not the idea of lobbying per se.

An 8 hour interviewing day plus an hour on the front end (screen) or back end (confirm/double check some things/really meet manager) can make sense.

3x 3 hours sounds -exhausting-.

But I really care about "there's a 95% chance the benefit is somewhere between 1.9x and 5x, so it's worth it."

So that's why CIs are interesting and p-values are less so.

Nah because we care about the size of the effect too.

We can get a p<0.0001 finding of an otherwise terrible, annoying change that saves 1%.

> You can easily review 10-100x that in an hour, even if you're being super pedantic about it.

I think spending 9-18 seconds per line of code is a good baseline for review. There are diffs where you mostly moved a lot that can go quicker. There are also diffs where I spend 20 minutes thinking about 5 lines of code.

About 40-45% of US petroleum use is burning it in passenger cars.

We're already a net exporter of oil; if we pushed a bigger fraction of cars and other transport uses to electrification we'd be a massive net exporter.

That wouldn't completely insulate us from global energy markets but it'd do a lot.

I think the scary bit is down the road. We can all use AI well because we have the gist of how things work and a well-honed set of technical empathy for "how they probably wrote it."

I'm not sure how you build that sense as well if you're starting out today, though.

It's just memory mapping the peripheral.

For historic reasons (e.g. 32 bit address spaces, plus the need to reserve the space for multiple pci peripherals) it has been a narrow, movable aperture.

Resizable BAR lets the size of the aperture be chosen (which is usually chosen to allow all of VRAM to fit in and be directly accessible).

Moving 1000W of heat generation takes way less than 1000W.

Hyperscaler PUE (which includes cooling, power conversion, etc) is typically 1.10-1.15, so multiply 1.5% by 1.1.

The person you replied to already mentioned cooling and multiplied by 1.67 to cover ancillary uses.

I think mechanical / electronic intuition are important, and they come from tinkering (especially at younger ages).

Sure, being able to do math and detailed analysis are important. But having a good guess on step 1 is important, too, to narrow the search of huge design spaces.

I think we've all experienced different kinds of engineers. The one you trust to do a month-long trade study with deep mathematical reasoning is usually not the one you trust to just know a plausible path forward. And we need both.

I operate a small fleet at my school:

* 1x Prusa XL Gen 1 with 5 toolheads * 1x Prusa Mk4IS * 1x Bambu Labs X1C with multi-material * 2x Voron 0.2 * 1x Sovol Zero

My experience: the typical higher-end printer runs out of the box and runs hundreds of hours (or more) between maintenance beyond wiping the plate down, occasionally greasing the rails, and blowing little plastic bits out of the enclosure. Every year I end up doing a few hour significant maintenance job on about one of the six printers.

But then there's lemons. That Sovol Zero is terrible and has been down more than it's been up, because of something in the extruder assembly. It also came with a defective camera and needed it replaced. I need to decide whether to rebuild all that with many hours of labor or just get rid of it.

I mean, on the one hand, manufacturing has grown near-continually for those 70 years, and its demand for labor has reduced. So, it's not like you're wrong.

At the same time, our production of lower value, commodity-ish things has indeed shrunk rather than grown with population of demand. That manufacturing expertise is more concentrated more on higher value (aerospace, information technology, etc) things and less on fasteners, bulk materials, or extrusions.

We are not vertically integrated in the way that China is.

Is this really true? Or are we counting like the games kids made in a computer lab somewhere.

Or there's this 87% number, but it's based on "no longer a legal way to buy it," not that it's not archived/preserved: https://gamehistory.org/87percent/

That doesn't really have anything to do with what I said.