zestier's recent activity
-
Comment on Whatever happened to _____? in ~talk
-
Comment on Whatever happened to _____? in ~talk
zestier (edited )Link ParentI wasn't disputing whether or not there is a correlation and being as charitable as I could to the idea that there is to steelman the position. My point with my example was just that even if there...I wasn't disputing whether or not there is a correlation and being as charitable as I could to the idea that there is to steelman the position. My point with my example was just that even if there was a clear correlation that didn't mean the causal link was there.
Regardless of any actual correlation there was at least a reported correlation and what was reported is maybe even more relevant than the truth uncovered in later studies as the starting point when answering a question about what happened to old reporting. Oddly enough the first place I saw anything about the reported link was here on tildes.
-
Comment on Whatever happened to _____? in ~talk
zestier (edited )Link ParentI'm not sure I fully understand what you mean since I don't know what a "true correlation" is, but yes it could have been basically anything when already selecting for a group that is often in...I'm not sure I fully understand what you mean since I don't know what a "true correlation" is, but yes it could have been basically anything when already selecting for a group that is often in pain.
One particularly charitable possible cause that I'm making up could be that autism somehow makes the pregnancy more painful (ie. autism causing Tylenol usage rather than Tylenol usage causing autism). That is an example that I completely made up so I'm not trying to say it's right, but just illustrating one possibility that contains exactly the defined correlation that is so far from having the cause right that it's backward. Even in that example I simplified away the possibility that autism and increased pain are both caused by some other pregnancy complication, meaning that they could share a cause rather than one causing the other. And those examples are still charitably granting that the correlation represents anything.
-
Comment on Whatever happened to _____? in ~talk
zestier (edited )Link ParentA correlation got called a causation. Groups that are wrong so often it could be believed that lying is intentional loudly parroted a false narrative anyway, as they do.A correlation got called a causation. Groups that are wrong so often it could be believed that lying is intentional loudly parroted a false narrative anyway, as they do.
-
Comment on How Europe is gearing up to follow Australia's teen social media ban in ~tech
zestier (edited )Link ParentFood, healthcare, and education (including social and creative skills) are also good for children, yet the big proponents of age verification, at least in the US, are staunchly against ensuring...Food, healthcare, and education (including social and creative skills) are also good for children, yet the big proponents of age verification, at least in the US, are staunchly against ensuring children have those things. With the "think of the children" justification for increasing surveillance, banning art, trying to avoid teaching unflattering parts of US history, and "protecting" kids from the existence of LGBTQ people coming from the same groups that oppose actually helping children I have no reason to read good will into any of their actions. They don't care about kids beyond them being a useful tool.
Yes, it's always possible that a group of liars using the same playbook they always have is differently motivated this time, but seeing as how they only want to "help" when it conveniently aligns with their other motives though I'm beyond a willingness to believe positive intention.
I do concede though that this is a very US-centric view. I don't know much about where it's coming from globally, but I know that in the US it is coming from some very unsavory people.
-
Comment on Whatever happened to _____? in ~talk
zestier Link ParentWhen I looked it up the factors I found were similar. Basically: More likely to be on paved/city roads. Improved paint and coatings make that kind of damage less of a problem. Improved tire...When I looked it up the factors I found were similar. Basically:
- More likely to be on paved/city roads.
- Improved paint and coatings make that kind of damage less of a problem.
- Improved tire compositions apparently help reduce kickup.
- Drag.
-
Comment on Meet Pebble Index 01 - External memory for your brain in ~tech
zestier (edited )Link ParentIt would be nice if the page gave any indication of what the battery life would realistically be for sane usage. Is anyone out there really going to be whispering to their finger for 4 hours per day?It would be nice if the page gave any indication of what the battery life would realistically be for sane usage. Is anyone out there really going to be whispering to their finger for 4 hours per day?
-
Comment on The Walt Disney Company and OpenAI reach landmark agreement to bring over 200 characters from across Disney’s brands to Sora in ~tech
zestier Link ParentOne thing I was wondering is if it could be a hedge against inevitable IP theft. It seems like AI companies have been more-or-less getting away with taking and generating whatever content they...One thing I was wondering is if it could be a hedge against inevitable IP theft. It seems like AI companies have been more-or-less getting away with taking and generating whatever content they want under the justification that it's "transformative". If I thought my stuff was going to be stolen anyway I could see wanting to make a deal that gets me something out of it.
-
Comment on How Europe is gearing up to follow Australia's teen social media ban in ~tech
zestier (edited )Link ParentOn a personal level I'm way beyond giving any emotionally driven "think of the children" legislation even the faintest benefit of the doubt. It's been used for literally decades as an effective...On a personal level I'm way beyond giving any emotionally driven "think of the children" legislation even the faintest benefit of the doubt. It's been used for literally decades as an effective way to push through weak positions on emotional grounds.
https://en.wikipedia.org/wiki/Think_of_the_children has much more info about why I trust nothing using this rhetoric. Here's just a snippet from the very start of the History section:
Art, Argument, and Advocacy (2002) argued that the appeal substitutes emotion for reason in debate. Ethicist Jack Marshall wrote in 2005 that the phrase's popularity stems from its capacity to stunt rationality, particularly discourse on morals. "Think of the children" has been invoked by censorship proponents to shield children from perceived danger.
We still see those things today with the right openly claiming things like not supporting their LGBTQ hate makes you a groomer.
-
Comment on Day 10: Factory in ~comp.advent_of_code
zestier (edited )Link ParentIt looks like there's a minor mistake in your example that converts to a mask. Specifically that 10001 would be 0,4 rather than 1,4. Doesn't really matter, but for a moment there it threw me as I...It looks like there's a minor mistake in your example that converts to a mask. Specifically that 10001 would be 0,4 rather than 1,4. Doesn't really matter, but for a moment there it threw me as I thought I was missing something.
-
Comment on Crunchyroll to end ad-supported streaming at the end of 2025 in ~anime
zestier (edited )Link ParentI had also got curious about such a definitive statement because I remembered the unsanctioned contractor situation, but from what I could find the situation is more complicated than that. For...I had also got curious about such a definitive statement because I remembered the unsanctioned contractor situation, but from what I could find the situation is more complicated than that. For example, https://www.cbr.com/crunchyroll-israel-ai-subtitle-controversy/ says that the Crunchyroll President has stated interest in using AI and when directly asked if they're using AI the company spokesperson neglected to answer that and corporate spoke around it.
The rumor seems to be that they switched from using Aegisub to OOONA, that OOONA is an AI tool, and that Crunchyroll seemingly refuses to deny this. My understanding is that this is also directly related to the worse subtitle formatting because the new tool isn't capable of dealing with cases like translating signs in screen space, leaving just the audio subtitles. This would line up with a new tool that uses AI on audio without consideration for what else is visible in frame, but on the other hand it also lines up with a new tool that simply lacks the dynamic overlaying features regardless of the author.
For what it's worth I don't know anything about the validity of this source.
-
Comment on Four proposals to improve the design of fuel economy standards in ~transport
zestier (edited )Link ParentTo be fair I do also think it's worse. To me it not only doesn't overcome the burden of being better, it is actually worse. There are a few reasons I think this: The bigger number being worse...To be fair I do also think it's worse. To me it not only doesn't overcome the burden of being better, it is actually worse.
There are a few reasons I think this:
- The bigger number being worse thing. We'd introduce the problem that 2 is better than 3. Not the end of the world, but we live on the same planet that https://eathealthy365.com/why-a-w-s-third-pound-burger-failed-a-deep-dive/ happened. People will deliberately choose the less efficient car because they don't understand the number.
- Lines like the "would only save one more gallon" come from a trap in that representing it like that makes it look like it says things it doesn't. 100 miles was chosen arbitrarily and is a kind of nothing unit, so why is it okay to treat specific deltas to that any stronger than if the base was 50 miles? It isn't, it is just subtly an unintentionally deceptive unit that can even trip up the well educated among us.
- This is one I didn't previously mention because it doesn't really matter that much, but since I don't think about driving in groups of 100 miles it simplifies the math I don't do while it makes the simple mental math I actually do do harder. To give an example of math I actually am interested in, let's say I want to estimate how much it'll cost me in gas to drive a 15 miles out of my way in a 40mpg car. I can look at that and easily go "around 1/3 gallon". Represent that same car as 2.5gp100m and the quick estimation math becomes that little bit harder. It's admittedly easier to get to the exact number with gpm, but since fuel usage isn't exact I'd rather the trivial approximation. And that isn't a contrived case, I do those exact problems with things like estimating my price delta of driving vs public transit or when it's worth it to drive further to get a better deal on something.
- I think that mpg representation makes the reason for diminishing returns more obvious. Why 2->1 is harder than 3->2, even though they're both just a reduction of 1gp100m, could take some explaining depending on the audience. Same numbers though are +50mpg vs +17mpg. +50 is harder than +17 just seems like it would be immediately intuitive to more people.
Though I wouldn't call my feeling "hate". I've spent years with one of the most important parts of my job being reviewing designs and poking holes in them. I don't hate a design for not holding up to being poked at, but that doesn't mean I won't challenge it anyway.
-
Comment on Four proposals to improve the design of fuel economy standards in ~transport
zestier (edited )Link ParentI get the math, but how is any of that helpful? How is arbitrarily messing with the bases because the units are inherently confusing good? Is 2.36gp100m really a sane unit that will be meaningful...I get the math, but how is any of that helpful?
How is arbitrarily messing with the bases because the units are inherently confusing good? Is 2.36gp100m really a sane unit that will be meaningful to people? Cars aren't rockets, we don't think about their usage in fixed increments because we don't use them like that. Sure we could even further change the base, but that'll only serve to confuse people even more. 0.0236gpm is both more reasonable and completely insane to look at. And that's still ignoring that there will be a group of people will think that 2.5 is better than 2.4 because higher is "better" on just about any other numbers they're going to be interested in comparing (ex. crash safety ratings, number of seats, towing capacity, total driving range, top speed, etc.).
For example, 2 gallons per hundred miles is twice as good as 4 gallons per hundred (going from 25 mpg to 50). Going from 2 gallons to 1 gallon (from 50 to 100 mpg) is both very hard and would only save one more gallon. At that point, going electric is easier.
This logic is very flawed. It's still halving it. Again, cars aren't rockets. That's halving continuous usage, not an independent event. That's because it's not saving one more gallon in general, it's saving one gallon per usage of an arbitrary unit that doesn't align with usage. Change the unit to some other arbitrarily selected thing, like gallons per 10 miles, and 2 -> 1 still "would only save one more gallon" but it's obviously very misleading to say that. Mess with the bases in the opposite direction, like gallons per million miles, and small changes can read like saving hundreds of thousands of gallons. What makes it sane to say that at 2->1 when using gp100m but not 2->1 when using gp50m? The fact that it even reads like this could be meaningful is just more evidence to me that this presentation creates confusion rather than solves it due to the real danger with introducing extra arbitrary units that interpreting them requires extra care taken.
And gas guzzlers already stand out in the current model: they have very low numbers. Using a reciprocal pulls most realistic numbers closer together. I guess it'll draw a larger line between like a hummer and a tank, but I can't say I think that's a range worth optimizing consumer messaging for.
As a random thought that's completely impractical but at least doesn't introduce something as arbitrary as 100 miles as a unit: cups or ounces per mile. Not necessity the ideal sizes, but cups and miles are at least normal units small enough that I can actually reason about them with thoughts like "1 cup is a lot of fuel to only go one mile". I realize that a similar thing is supposed to be the idea behind gallons per hundred miles, but 100 miles is so big that it's difficult to feel anything about it.
-
Comment on Four proposals to improve the design of fuel economy standards in ~transport
zestier (edited )Link ParentI'm thoroughly confused as to how this would be helpful. People are already aware that 20mpg is better than 6mpg, so I'm not clear on how inverting the fraction and multiplying it by 100 really...I'm thoroughly confused as to how this would be helpful. People are already aware that 20mpg is better than 6mpg, so I'm not clear on how inverting the fraction and multiplying it by 100 really does anything. If anything, as we saw with the 1/3 pound burger, lots of people don't really get fractions and just default to "bigger number = better" which is accurate in the current model but not in your suggestion. I'm sure there would be a non-zero amount of people that end up thinking that 5gp100m is better than 4gp100m.
It seems like it's based on the idea that companies would then give numbers based on the average over 100 miles, but I can't connect why that would be true. They'd just follow the current model that sounds better by saying "5 gallons per 100 miles highway". If you want the numbers stated to need to be averaged over some amount of mixed use that seems like a completely tangential requirement. Maybe a good one, but I don't follow how inverting the fraction makes that happen on it's own.
I'll also throw in that it reduces whole number granularity to a point that it completely breaks down looking nice at high fuel efficiency. Above 50mpg is what? 1? 2? How are we rounding it? Or we throwing decimals in to the fractions so we can actually see the difference between 52 and 72mpg. And above 100mpg we can't even use whole numbers to represent it all. People have enough trouble with fractions and decimals without mixing them like 0.6/100 gallons per mile. Whole numbers are already less granular even in your given example (is 5gp100m 19mpg? 22mpg?).
-
Comment on I fixed my lactose intolerance -- by chugging all the lactose in ~health
zestier (edited )Link ParentI don't believe so, but she also doesn't have any other gluten reactions. It is only the combination of specifically white flour and dairy. She can have as much other gluten as she wants and still...I don't believe so, but she also doesn't have any other gluten reactions. It is only the combination of specifically white flour and dairy. She can have as much other gluten as she wants and still have dairy. Or she can can have as much white flour as she wants, but then cannot have dairy.
-
Comment on I fixed my lactose intolerance -- by chugging all the lactose in ~health
zestier LinkMy wife kind of fixed her lactose intolerance, but in a completely different way. We did a low carb diet for a while and while on it her lactose intolerance magically disappeared. She has since...My wife kind of fixed her lactose intolerance, but in a completely different way. We did a low carb diet for a while and while on it her lactose intolerance magically disappeared. She has since learned that she's only lactose intolerant when also consuming white flour. We still don't know exactly why it's the case, but any time she eats too much white flour her lactose intolerance returns for a few days. Gut biomes are weird.
-
Comment on What programming/technical projects have you been working on? in ~comp
zestier Link ParentWhile I suspect you aren't concerned with bugs in your snow effect I'll report one for you anyway. Maybe it'll one day be relevant in something you are more concerned about. When using a mobile...While I suspect you aren't concerned with bugs in your snow effect I'll report one for you anyway. Maybe it'll one day be relevant in something you are more concerned about.
When using a mobile browser with bottom anchored address bar the effect stays at the location that is the address bar's top when the bar it is fully extended. The problem with this is that the address bar goes away when scrolling and then the effect is no longer anchored to any visible element and leaving a gap under the snow. I observed this is multiple browsers.
-
Comment on Fizzy, a new source-available Kanban tool by 37signals in ~tech
zestier Link ParentI got curious myself and found https://jakelazaroff.com/words/dhh-is-way-worse-than-i-thought/. I guess here's the corresponding Reddit thread...I got curious myself and found https://jakelazaroff.com/words/dhh-is-way-worse-than-i-thought/. I guess here's the corresponding Reddit thread https://www.reddit.com/r/ruby/comments/1nwcwad/dhh_is_way_worse_than_i_thought/. The reason I added the Reddit thread was the potential for the "well actually" counters, although they don't really seem present.
The tl;dr is racism.
-
Comment on IKEA finally arrives in New Zealand. Even the country's leader came out to celebrate. in ~finance
zestier (edited )Link ParentThe pdf implies that its more about room conversions rather than putting carpet in a traditional garage. Anecdotally I've seen this once in my life, although with nothing to do with NZ. In that...The pdf implies that its more about room conversions rather than putting carpet in a traditional garage. Anecdotally I've seen this once in my life, although with nothing to do with NZ. In that case the owners went a bit further than I'm understanding from the pdf, but they basically shrunk what used to be their garage by moving the wall almost right up to the garage door (I think it remained like 2 feet deep or something), removed part of the wall that was connecting it to the main house, and then finished it as like a semi-open rec room. The pdf is pretty sparse on details, but to me it implies the lite version of that where you don't add or remove any walls but just never open the external garage door anymore.
Here are the relevant bits of the pdf:
Our home visits reveal that New Zealanders make the most out of their spaces at home. For instance, many rooms are multifunctional and serve more than one purpose. 93% of garages are carpeted and repurposed for activities such as hobbies, laundry and exercise. Living rooms are a place to relax, watch TV and do yoga, while spare rooms provide a space for guests to stay over, work or play games. These multifunctional rooms have created a desire to introduce even more functionality and efficient solutions.
In New Zealand, garages aren’t often used for parking cars. They’re carpeted and repurposed as an extension of the home for almost anything – storage, laundry, work, exercise, hobbies or children’s play and this is very different from the rest of the world. It is the most multipurpose room of the home due to its potential to accommodate so many different activities.
I'm also not trying to make any claims about the truth of the claim. While outside NZ I have observed a garage being converted to another room before the idea that anywhere is doing it at a rate of 93% seems very very high. With rising housing prices I can definitely see the appeal of recovering the square footage of the garage as a living space, so it does seem reasonable it could be rising in popularity I suppose.
-
Comment on The turf war for Seattle in ~transport
zestier LinkI watched this before it was posted to Tildes, but something I recall seeing posted a few times in the comments is that this video is much higher quality than expectations for a channel first. My...I watched this before it was posted to Tildes, but something I recall seeing posted a few times in the comments is that this video is much higher quality than expectations for a channel first. My money would be on that this person has experience as a paid editor and decided to branch out to independent content.
I'm personally glad to see that no airline is taking complete dominance at this airport. It keeps things competitive, which is great. I do wish that Alaska would improve their offerings a bit though. Between the two I would probably say I'd prefer to support Alaska, but Delta feels like a just slightly better flying experience. Admittedly its little things, like that Delta has the built-in TVs versus Alaska trying to upsell renting tablets, but those add up when trapped in a seat for hours on end.
SeaTac was also a relevant player in another recent video, I think https://www.youtube.com/watch?v=3_V6HJNDjDs about American Airlines from Wendover Productions. The tl;dr of its relevance on this specific topic, assuming I recall it correctly, is that where Delta managed to get a proper foothold into using SeaTac as an international hub American failed at that same strategy.
Ah, yeah. That came from me misinterpreting what was meant by "true correlation".
I was initially taking it to mean that a "true correlation" is one where something, anything, of significance is being said. Basically a correlation that, no matter how indirect, means something (ex. sharks and ice cream are both connected by summer).
I see now that the actual intent was existence. Like a "true correlation" is one that has enough statistical backing to say there is one at all. This is probably partially because I was still in the steelmanning mindset of not wanting to fully dismiss it entirely.