curiouscavalier

but i also “align my tires” or might check the “alignment of two vectors.” if anything, thinking of alignment of anthropomorphizing seems to only inherit it from all the other anthropomorphisms used. rogue originates from a term for vagrant, right?

Yeah, I probably should have reported it at the time. They didn’t cite anything specific even after we requested what in the on-roof (not drone) inspection indicated an issue or was missing from the inspection.

Though to be precise, this was over a year ago so I’m not remembering their exact wording. But it likely wasn’t literally “drone footage alone,” but probably something more like “our drone footage and associated analysis” — in any case the effective interpretation being “we don’t care about installation date or your inspection. Replace it or lose coverage.”

Not that it makes it right, but there’s enough providers in my area it was less effort to get new coverage than continue to fight them.

Yes, that ended up being the solution. Cost aside, I’d find it pretty wasteful to do the replacement.

The report was done at the time of the dispute. The roof was 5 years old.

The issue is that often the data gathered and used with no consideration of other evidence. For example, I had a similar situation of home insurance sending a letter demanding I replace my roof based on drone footage. No issue was immediately obvious from the footage and we had just replaced the roof less than 5 years earlier. We had an inspection done and then sent both proof of replacement and the inspection report. We were told the drone footage alone would be used in their decision, and we still had to replace the roof to keep coverage.

So it is not necessarily objective, fairly considered observations (and I think pretty different from speeding tickets). At least in our case it was maximally for the benefit of the insurer. I think that is relevant to their claimed premise.

having developed multiple apps on it and tried every which way to use it (as an XR enthusiast in general), I have never been so happy to put a headset up on the shelf and never pull it out again.

using as a spatial monitor was cool. for about 10min until my neck got tired of the added weight. but I’ll give credit that those 10min were pretty cool.

yeah this is really a part of it. Both founders and investors get spooked by the rapid entries into a market but persistent not just when it’s hard but when it’s boring goes so much farther.

Or that there is a sufficiently generalizable objective function for all “spatial intelligence.”

The adoption of it in various industries for training is larger than most people might suspect. First responders and retailers have some of the largest internal deployments out there, but they aren’t massively publicized (most people would never guess who’s fielding the largest fleet of headsets right now). That said, it’s still not mass adoption.

At the end of the day, you are asking someone to put something on their face that is still very different ergonomically than glasses (and I’m not sure even glasses would overcome enough friction). The ROI has to overcome the business (or personal) friction of buying the hardware, the friction of the form factor plus any friction from changed workflows.

Now put that in an operational workflow instead of training and the risks go up. Most are still skeptical of device reliability (not to say there aren’t suitable devices for operational roles but the perception is still a hurdle, and the applicability is often device-specific). Now add on to that limited experience with devices (many decision makers have never put one on), added security complications, specialized software development skills, limited content libraries and very real accessibility concerns and a lot of enterprises can never get past an “innovation center demo.”

For many industries the value proposition just isn’t there yet. But that said, I’d recommend digging a little deeper as there’s a lot of existing use-cases and deployments, both failed and successful, outside of IVAS.

Thanks for posting this. Water rights and who gets priority was the first thing that came to mind. The thought of “wind augmentation plans” is fun to think about.

Yep, the procurement process (and related) requires a lot be baked into pricing. I’d also be curious what the fully burdened rate of in-house staff is compared to consultants. I’ve seen situations in the gov (not DoD) where despite high consulting rates, the full cost of hiring was even higher per hour.

But I’m loath to defend the big firms. Generally, quality plus the ever push for expanding scope leaves a sensation of waste. The solution is just going to need more than simply tossing them out.

Appraisals also act as a gate for prices due to lender requirements. A bit round about, but I can see how looser appraisals can enable inflated prices. Imagining the opposite extreme is interesting: What if appraisals never returned with higher prices than the last sale of that house? Some markets would see increases from people paying the difference out of pocket, but I’d guess the rate of price increase would be much lower. (plus other effects, of course).

I’m not following the “unconnected” part. There are definitely problems with our patent system (referring to US), especially around software, but in my experience examiners are indifferent to your background and lineage (though not indifferent to their own status at USPTO). There is absolutely a monetary barrier to entry on using a lawyer to draft your patent application, but I feel like that’s more an issue of private law firms than patents in general. Though I’m sure others might have comments on how those intertwine.

But filing fees, etc (ie those things set by the USPTO) are really quite reasonable imo. Strictly speaking you don’t have to use a lawyer to file (I know that can be a minor concession in the landscape of practical success). Maybe you can clarify what you mean by “connected” vs “unconnected”in this case? I’m missing how patent law directly related to connections/lineage beyond what sister comments have said re: ability to litigate or be patent trolls. But I think that’s the point of the sister comment on it (at least ideally) cutting both ways.

One of my favourite texts. One of those that I found influential early in academics as well as when re-reading later in my career. Even for younger students I think it can be great introduction to more formal approaches, as well as a taste for the austere.

Agreed. There’s a lot of variables (and I think price is a big one). But, while slow, adoption in enterprise is showing signs that the basic concept has some legs. Even if the tech today needs some time to marinate.

That said I also don’t think we’re are a time-local maxima of users either.

And it penalizes in many ways. Focusing too much on grades can be detrimental in graduate studies, despite graduate admissions focusing on GPA and test scores. I remember seeing 4.0 undergrads really struggle with research in grad school, sometimes to the point of dropping out. Certainly not always the case, but for the ones that did I think it speaks to your point about different skillsets.

Maybe worse was seeing the undergrads who passed on research opportunities out of fear it would distract them from keeping a high GPA.

Who said what we want has anything to do with our vision. Please don’t make assumptions about what drives our product development.

What I want in a product comes from customer interviews. It’s not “my opinion” other than perhaps our team’s interpretation of customer requests. A customer can want certain pain points addressed and have friction to move to a particular solution at the same time.

Or does wanting a product that meets customer needs too opinionated?

Solid point. We’re building in a space adjacent to this and it definitely feels hard to balance removing friction for adoption and delivering everything we want.

That’s a reasonable point. And I’ll admit I’m expressing more of an opinion on privacy in general. I don’t know what the real risk is, immediately or over time and it would vary to an extent by individual. I suspect, again uninformed, it’s fairly small for most people.

That’s true, but those are also different agencies, so at a minimum you’re spreading the data (and risk surface). My concerns are generally (though not entirely) around data breaches rather than the collector. And sadly we have ample evidence that federal agencies aren’t always great at securing PII.

But more to the point the reason for doing it is there, for better or worse — want a passport? Have a picture taken. In this case I am (at least for now) comparing opting out or not. You can still take your flight without them taking that additional picture. That of course may not stay true.

Are they basic RGB cameras or do they collect depth as well? Would even know if they decided to change the system to start collecting depth data, even if comparisons are to flat images? Honest questions, as that’s where my concerns are.

Boil the frog. If they start to collect more, it’s not as simple as “they already have my picture.” Should there be a breach, I can’t easily change my face.

I’m generally of the opinion that when it comes to personal data collection we should always say “no, convince me you should” rather than “sure, why not?” Other than an ever-so-slightly faster processing time is there evidence of any real security gain? My initial reaction, uninformed on the specifics but fairly versed in this type of security, is that’s unlikely.

Many do go out of business, and many substantially undercut others. This is particularly true for the diesel market, where OTR truck operators are more price sensitive. Most stations are buying gas around similar prices at the terminal, but their overhead to provide the station can vary wildly between large chains and small, family owned stations.

“Skimmers” (drastically lower-priced diesel stations) are a real threat to larger chain stations. They don’t put them out of business entirely for a number of reasons, but one is because gas stations ultimately sell more than gas/diesel (e.g. a clean bathroom is often worth a few extra cents per gallon for me, but there’s a lot more than that too). Also any given station can’t sell an infinite amount of fuel. Beyond tank capacity, wait times are a real impact on sales volume, and so if a station offers a price too low (even if still profitable), it is just leaving money on the table.

So not to say collusion doesn’t happen, but one can also arrive at similar pricing with a commodity like gas without it. Especially since fuel margins are typically low.

This reminds me of working with a company that provided market-wide pricing data for a particular commodity across the US. They were the de facto aggregator of pricing for the sector and at the time I worked for one of their larger customers. We requested they add another vendor’s pricing in a particular region and received a response along the lines of “Sure, as soon as we can figure out how to add another entry. We’re at the maximum row count on the system.”

Needless to say it gave my team a few days of flabbergast and speculation on what system they must have built on to hit a row limit at only 5 digits. And yet for a non-tech industry it was mostly working.

Dating is done with more than just material analysis. Evidence of tools used to make the toy, techniques for things like joins and stitching, etc. can all be indicative of methods that can give at least a lower bound. How applicable method differentiation is to this specific case obviously depends on a number of things.

Been a while since I was in this area, so I may be off base. But my knee jerk thought on Ubn is getting another valuable data point on nuclear structure. We’ve got a magic number at Z=82 and one at N=126. Do we get it one at higher Z? This doesn’t answer it directly, but it’s a step along the path. And at the very least is a great data point for confirming/constraining various structure models.

I had a similar experience with an annoyed agent, who also made sure to point out how I was holding everyone else up instead of just letting my picture be taken. The agent likely didn’t realize that showing that pettiness really reaffirmed my choice as to why the data collection is dangerous to begin with.

I appreciate this perspective on prompt engineering. I’d love to think that one of the great outcomes of LLMs are people returning to more decent and precise forms of communicating. Imagine the progress if we could get that to transfer to human-human communication as well.

Statistical Mechanics by Pathria. Every read teaches me something new and reminds me how little I know.

That's an interesting point. You potentially also have alignment issues with the window being placed spatially, i.e., rendering text that is not perpendicular to the plane of the screens. When moving my head in the AVP I'm moving the screens, unlike when I move my head to look at a different part of a monitor.

We've sent spiders to space[0] (honestly, it's like we _want_ a horror movie to play out). But who knows - maybe one day we'll get DNA from extraterrestrials.

[0] https://www.si.edu/newsdesk/releases/world-s-first-spidernau...