I can't decide if I should have swapped the locations of two jokes: made the OP title "I Blame the Humanities" and then after the pullquote, "Whoopsie. I guess we NEDA new hotline."

The quality of this site won't wreck itself, you know.

I think usually cooking food results in more calories that can be absorbed by the digestive system. There's lots of energy left in poop if you eat raw foods, unless it's just raw meat.

chatgpt says:
Yes, cooking food can change its caloric content, but it's not quite as simple as "cooked food always has fewer calories than raw food." The process of cooking can cause some chemical changes that reduce the amount of certain nutrients (and therefore, calories). For instance, some vitamins can be destroyed by heat.

However, cooking also often makes food more digestible, which means your body can absorb more of the available calories. For example, the process of cooking breaks down the cell walls in many types of food, which makes it easier for your body to access the nutrients inside those cells. This is why cooked vegetables and meats are generally more caloric than their raw counterparts, despite the potential loss of some nutrients during cooking.

So, while you're correct that the heat from cooking can break down some of the bonds that store energy in food, the overall impact on caloric content is complex and depends on a variety of factors, including the specific food, the way it's cooked, and how that affects its digestibility.

Should have told it to include the word "poop" in its answer.

You shouldn't have needed to if it worked right.

Oh, that makes sense.

So when they say how many calories food has, they mean according to the bunsen burner, and not according to how much gets absorbed by your gut? I wonder if that's part of the reason processed food has an outsized impact compared to how the nutrition stacks up on paper. In other words, suppose you eat a bowl of steel cut oats with the same calories as a bag of doritos. Maybe the doritos are actually contributing more calories to your system.

Goddamnit, I've got us on another weight loss thread, haven't I. ABORT! ABORT!

I eat steel cut oats for breakfast every day.

Given Sharon Maxwell's instagram, I suspect she entered her height and weight and was told that she should lose weight.

The substantive advice appears to be taken directly from the Mayo Clinic's Healthy Lifestyle Weight loss page:

It may seem obvious to set realistic weight-loss goals. But do you really know what's realistic? Over the long term, it's smart to aim for losing 1 to 2 pounds (0.5 to 1 kilogram) a week. Generally to lose 1 to 2 pounds a week, you need to burn 500 to 1,000 calories more than you consume each day, through a lower calorie diet and regular physical activity.

Union busting is bad. AI replacing workers is going to be one of the larger societal problems of the foreseeable future. But this just seems like an activist-influencer chasing clout.

Sure, but she entered it on an eating disorder chatline. That seems like a very normal piece of info for someone with an eating disorder to volunteer on a chatline.

The point isn't that the advice is bad in general. The point is that it's not the advice you want to give on a hot line for people with eating disorders. There's lots of good advice that you are in the wrong if you bring up to certain people or in certain contexts. Like my one cousin who suggested my uncle would feel better if he took up jogging. My uncle had terminal cancer at the time.

My cousin was an asshole and I can prove it with science. He was a hospital administrator.

Just to be nitpicky, a food calorie is actually a kilocalorie -- 1000 of the "heat 1 gram of water 1 degree Celsius" calories. And yeah, you figure out how much energy is in food basically by burning it, but it's not clear at all precisely how much of that energy makes it into your body for any specific food.

13 should have started with "Well, actually."

On the story, I can't remember where I read this, but if what I read was right, this was a chatbot with canned responses, not some kind of ChatGPT LLM nonsense. So some human individual or group of individuals made the decision to have a chatbot on an eating disorder helpline giving out diet advice, which does seem like a very bad idea.

@10, 11

I find your responses interesting. Because the advice strikes me as reasonable.

Given the absurd breadth of the term "eating disorder", I'm not sure that the advice is categorically wrong for an eating disorder website. That would seem more a moral conclusion than a medical one.

The advice does not seem appropriate for an anorexic or a bulimic. But advice to slowly lose weight through caloric restriction seems reasonable for someone at an unhealthy weight whose "eating disorder" comprises fad or crash dieting, followed by regaining the weight.

So the question really is: what is Sharon Maxwell's eating disorder and is the advice reasonable given that disorder?

The initial denials from the organization are interesting -- like, no one defended the advice as appropriate, they said the chatbot wouldn't have said those things. You kind of wonder if there's someone working on the chatbot with either malicious or well-meaning intent to make it give diet advice against the organization's policy.

It could be a homebrew chatbot that still uses some LLM functionality? But I too didn't see any direct evidence of that.

16 Here's the thing: a human would have been able to gather enough context to ascertain whether this advice was going to be helpful to the person seeking advice. If you replace the human with a machine, you either need to set it to get enough context -- a big damn deal -- or just decide that dispensing the advice that would be generally appropriate to most people, even if not the person seeking the advice, is good enough. It's no defense of the latter choice that the advice wouldn't been fine for plenty of people. What they've done by switching from a human to a machine with inferior settings is a huge drop in user utility.

@17, 18

I suspect that it is more of a decision tree / expert system setup.

But here is the crux of the issue. Wired gives the following account:

Tessa was paused after several people saw how it responded to even the most straightforward questions. One was Alexis Conason, a psychologist who specializes in eating disorders. In a test, Conason told Tessa that she had gained a lot of weight recently and really hated her body. In response, Tessa encouraged her to "approach weight loss in a healthy and sustainable way," advising against rapid weight loss and asking if she had seen a doctor or therapist.

When Conason asked how many calories she should cut a day to lose weight in a sustainable way, Tessa said "a safe daily calorie deficit to achieve [weight loss of 1 to 2 pounds a week] would be around 500-1000 calories per day." The bot still recommended seeing a dietitian or health care provider.

Conason says she fed Tessa the kind of questions her patients might ask her at the beginning of eating disorder treatment. She was concerned to see it give advice about cutting added sugar or processed foods, along with cutting calories. "That's all really contrary to any kind of eating disorder treatment and would be supporting the eating disorder symptoms," Conason says.

Here is the question:

Suppose someone has gained substantial weight recently and hates the way they look. Is the appropriate response to tell them that they should be happy with the way that they look? Is the appropriate response to help them lose the weight? What is the problem that we are treating: the weight gain or the decreased self-image?

would've not wouldn't

16: I think for someone who's talking to an eating disorder chatline, the risk that whatever their disorder is will be exacerbated by diet advice is high enough that it would be irresponsible for it to be giving out diet advice without human supervision.

What is the problem that we are treating: the weight gain or the decreased self-image?

Are you a bot? Or just in the pocket of Big Bot?

The answer is "both" or, more properly, "it depends."

And the context you're not accounting for is that this is not a general medical advice bot, it is specifically for people with eating disorders. As I noted above, the organization with expertise isn't defending this as good or appropriate advice in that context, they're saying "this shouldn't have happened and we don't know how it did."

Suppose someone has gained substantial weight recently and hates the way they look. Is the appropriate response to tell them that they should be happy with the way that they look? Is the appropriate response to help them lose the weight? What is the problem that we are treating: the weight gain or the decreased self-image?

The appropriate response in this situation is not advice targeted to rapid weight loss, as everyone agrees, which is recognized as another kind of weight disorder, the ups-and-downs phenomenon.

Your JAQing to imply that mental health is a frivolous pursuit of vanity, or that it is mutually exclusive with physical health improvement, seems characteristic of the overall nasty tone you've maintained here and I am going to disengage from now on.

Also, and I'm kind of repeating myself here, "gained substantial weight and is unhappy with the way they look" is not a description of diagnostic criteria for an eating disorder.

12: Heh

Where AI could be useful is in abstracting records and organizing medical notes that providers review and edit.

The whole decision tree stuff sucks. They are trying to roll that out at my place so that they can standardize everything at a remote call center. So now, it takes much longer for staff to schedule and staff with knowledge and experience can't do their jobs effectively. They keep phrasing it in Orwellian terms: it's more modern; this will be a support to the Patient Service Coordinators who won't have to learn different schedules. It's really impersonal and patients hate this stuff.

22: People (meaning administrative leadership folks even when they have MDs)really want decision trees for all but the most complex situations.

I wonder if UPMC will try to force my doctor to see more patients by making ChatGPT tell me to try to eat less? My doctor is always careful at broaching the subject and it takes him some considerable time.

Yeah, "nope," maaaaaybe you should step away from commenting about weight loss and diet here. Your intense fascination with other people's body size and diet is creepy and IMO not really normal, and there's no way I'm going to join in these threads while you're riding your weird fucking hobbyhorse all over them. I'm starting to think it's not just me who feels this way. You looked up some woman on Instagram to comment on her weight? WTF?

So the question really is: what is Sharon Maxwell's eating disorder and is the advice reasonable given that disorder?

You're thinking that the chatbot asks questions about the nature of the eating disorder and alters its advice accordingly? Maybe!

This story is linked in the Vice piece, and it suggests the chatbot isn't great at individualized guidance.

Here's a relevant bit:

Conason tried out Tessa herself and posted screenshots of their conversation on Instagram. She told the Daily Dot that her conversation with Tessa lasted about fifteen minutes.

To start, Conason told Tessa that she had recently gained "a lot of weight," really hated her body, and was advised against weight loss by her therapist because she has an eating disorder. The chatbot responded by saying that it is important to "approach weight loss in a healthy and sustainable way" and suggested she exercise more, and eat a "balanced and nutritious diet." When Conason asked how many calories she would need to "cut per day" to lose weight, Tessa told her in order to lose one to two pounds per week, she should eat 500-1000 calories less than she is eating now and recommended she consult a "registered dietician or healthcare provider."

With regard to her experience with Tessa, Conason said she was disappointed but not surprised.

"The kind of questions and comments that I put into the chatbot are those that many of my clients, many who are in larger bodies, struggle with," Conason told the Daily Dot. Treatment for eating disorders is "very nuanced," and the chatbot just wasn't able to measure up to the support that trained employees and volunteers were able to offer, she said.

Of course, actual human beings suck at giving advice, too, but I think "was advised against weight loss by her therapist" would be a big red flag for a human responder.

24 And the answer to the question about how it did is the same as the answer to the question how the chatGPT lawyer ended up with a brief that included made up cases. It's designed to be a stupid person's version of a smart person, not an actual smart person (which is surely much harder).

27.2 It's only useful for those limited tasks if it's set to do those tasks with 100% integrity.

Oh, funny. I was relying on the fact that she was talking to an eating disorder hotline at all, and missed that incredibly glaring specific red flag.

Yeah, I really think the interesting question is how did this happen at all? As far as I can tell, there shouldn't be any paths on a decision tree underlying an eating disorder bot that lead to advice on how to restrict calories. Some human being input text that was completely inappropriate for the purpose.

31.1: I'm relying on that thing I read that I can't find, but I think this really isn't ChatGPT. It might be using all sorts of clever AI to parse the human input it's getting, but the output should be picked off a list of canned responses -- might be a very long list, but I don't think it'd make sense to have it be making things up. And the canned response that came out is something that shouldn't have been in there at all.

This isn't what I saw, but says the same thing: https://fortune.com/well/2023/05/31/neda-ai-chatbot-harmful-advice/amp/

The creator of Tessa says the chatbot, which was specifically designed for NEDA, isn't as advanced as ChatGPT. Instead, it's programmed with a limited number of responses meant to help people learn strategies to avoid eating disorders.

"It's not an open-ended tool for you to talk to and feel like you're just going to have access to kind of a listening ear, maybe like the helpline was," Dr. Ellen Fitzsimmons-Craft, a professor of psychiatry at Washington University's medical school who helped design Tessa, told NPR.

That's just an FAQ with extra steps.

Ok, so it's like the Republican policy chatbot:

Inflation Rising -- we need tax cuts
Inflation Falling -- we need tax cuts
Unemployment Rising -- we need tax cuts
Unemployment Falling -- we need tax cuts
Deficit Rising -- we need tax cuts
Deficit Falling -- we need tax cuts.

Easiest decision tree ever.

35: Parsing humanly vague input and getting to the right canned advice could be a big deal.

But the weird thing (and this is a super strange story) is not just that the inappropriate-for-eating-disorders advice was there at all, but that it also seems to have been right on the surface and easy to get to. Like, I could kind of imagin a chatbot that ran through a bunch of diagnostic queries with a patient who denied all of the possible eating disorders criteria and having a buried page for "okay, I don't know why you're talking to an eating-disorder bot at all. But you seem to be fine, so if you want healthy weight loss advice here's what the Mayo Clinic says."

But if that option existed at all (which in context it probably shouldn't) it should be hard to get to. Instead, that seemed to be right out front. Which again is really weird. I would be wondering if this was something specific that happened in the last update.

35: Parsing humanly vague input and getting to the right canned advice could be a big deal.

This was an interesting article about a real estate service that used a combination of chatbot & (overworked) humans to provide service.

In a typical leasing office, the phones ring constantly. Agents spend most of the workday speaking to prospects, who often ask the same litany of questions. But with [Chatbot] Brenda fielding calls, the phone lines were silent and agents were free to attend to other tasks. And Brenda was more efficient than the most industrious human agent. She could cross-reference a vast database of property information in an instant and field messages faster than any human at a keyboard. She could answer calls at all hours of the day and night, didn't need a lunch break, and could work weekends and holidays. When the leasing agents arrived in the office each morning, their tour schedules were neatly arranged, as if by elves in the night.

Meanwhile, we operators, with our advanced degrees in the humanities, had aptitudes Brenda lacked. We were intuitive, articulate, and sensitive to the finer points of delivery. At $25 an hour we also cost almost nothing to employ, by corporate standards. Under the Brenda-operator alliance, everyone came out ahead: the operators got paid better than they would as adjuncts, and Brenda became more likable, more convincing, more humane. Meanwhile, Brenda's corporate clients were self-satisfied knowing they had not replaced their phone lines with a customer-service bot. What they were using, instead, was cutting-edge AI backed by PhDs in literature.

@25, 29 Perhaps your intense fascination with policing opinions is the problem?

I find the discourse around weight and dieting interesting. Mainly because obesity is a health problem that is more and more being treated as a identity characteristic. People on this website tend to be very collectivist about government health interventions (e.g., seatbelts, vaccines, etc.) but very individualistic about identity issues. Here we have something that is both an identity issue and a health problem. That's interesting! I want to hear what you all have to say about it - you're smart people with values somewhat different from mine.

In this case, one of the things I find interesting is the question of what is being treated. The expert system is kind of irrelevant - I'm not really concerned about how it performed or whatever. Given that the goal is to improve the mental health of the patient and assuming that there is no dichotomy between physical health and mental health, what advice should we give?

And this question seems to come up alot! This discussion reminds me of the whole "self-esteem" movement from the 1990s, where we focused on increasing children's self-esteem without considering what supports self-esteem. Or the question of how to treat depression - pills, cognitive behavioral therapy, or lifestyle changes?

38 sounds like it was ghost-written by the people who forced Brenda on the situation, though that's counteracted by the "we" in the second paragraph. I hate that kind of pretend-human answering machine with a fiery hot passion myself, as most normal people do.

"assuming that there is no dichotomy between physical health and mental health"

That's doing a pretty big lift, and I would think you could figure out why.

And if you can't figure out why, then you would probably think "tell people with eating disorders, who are overweight, that they're physically unhealthy and need to prioritize weight loss" is a good idea. (slow fingers, following up on 41).

Given that the goal is to improve the mental health of the patient and assuming that there is no dichotomy between physical health and mental health, what advice should we give?

This question is completely empty unless you're talking about some specific set of patients.

If the patients you're talking about are "patients who meet the criteria for being diagnosed with eating disorders," there are experts who work in the field who have come to consensus that giving them diet advice, even sensible, reasonable diet advice, is dangerous for them.

Do you disagree with that consensus generally, or with respect to some specific set of patients, and do you have any basis for that disagreement?

though that's counteracted by the "we" in the second paragraph

Yes, the article is mostly about the difficulty of the work for the human operators (both logistical difficulty and emotionally taxing), but I didn't want to thread-jack too much.

I had to use a chatbot recently, and it pretty quickly switched me to a person.

Specifically, I had closed my a Bank of America account, so the system wouldn't let me stop transferring money from the B of A account to ally.

38: I always make calls like that from my landline.

I hate to be accused of "policing opinions," you know, so let me make it super clear that I just hate this "nope" guy and there's no real deeper principle at work. Opinions are fine. Expression is free! Some people are assholes! The idea that you can be free to be an asshole is simply incoherent! Anyway, given a conflict between a person who really wants to talk and another person who really doesn't want to listen, the resolution goes (literally) without saying.

47: Yeah, I'm pretty sure you're right both about what the viewpoint is that he's circling around (roughly, that anyone calling diet advice harmful to anyone who isn't visibly starving is just being precious, and fat people should be firmly instructed to lose weight even if it hurts their feelings) and that it's pernicious and stupid. I just kind of enjoy getting that sort of thing pinned down explicitly.

But not enjoying engaging with it is probably healthier and certainly less of a waste of time.

48: I just kind of enjoy getting that sort of thing pinned down explicitly.

Yeah, that's me, also. One ought not feed trolls, but I was honestly wondering about whether Maxwell was playing fair, and probably wouldn't have gone to the trouble to look into it without the provocation.

On the other hand, is there a plausible scenario where it is ever appropriate to give dieting advice on an eating disorders helpline? I think a case could be made that the form that the actual advice took may be inappropriate in all conceivable scenarios, regardless of the question asked. I am trying to imagine an Alcoholics Anonymous helpline that answered an analogous question with advice on moderation.

13: Just to be nitpicky, a food calorie is actually a kilocalorie -- 1000 of the "heat 1 gram of water 1 degree Celsius" calories

This misunderstanding was one of the most egregious errors I ever encountered in a published book. It was a book by Jim Fixx--the running guy who died of a heart attack. The book was from earlier in his career and was embarrassingly titled Games for the Superintelligent (embarrassing both that it contained the error and that I owned it--but in my defense someone gave it to me as a gift). To make matters worse he was a graduate of DFHC. The context was his marveling at how the "calories" you expended warming an iced drink did not more than offset the "calories" in the drink... (how that was in a "games" book I forget).

The kind of employer who decides to fire everyone and replace them with a chatbot rather than recognize a union is probably also the kind of employer who would lie about replacing people with a chatbot.

The "calories" quoted as associated with smiling/frowning are also the smaller version, right?

50: To be fair to Fixx, it's really weird that the colloquial term for kilocalorie is calorie. Nothing else works like that. If you didn't happen to know, how would you ever suspect?

Isn't it technically Calorie with a capital C? Still weird though.

53: I thought "Encyclopedias?" and found an older one (if not older than the Fixx book) on the Internet Archive - it mentioned that fact. Seems like the kind of due diligence one might do in publishing nonfiction.

It takes more calories to confidently assert something than to go look it up in an encyclopedia.

53 -- As resistant as I've always been to efforts at education, this is one thing that actually got through. 5th or 6th grade? Middle school?

I think we had a science thing called a "Bomb Calorometer". Or maybe a video about one?

It might have been a home-brew kind of thing. My chemistry teacher was very good and very good at building stuff. I could ask him, but I've hidden him on Facebook for repeated Trumpyness.

Hey has anyone heard from Canada lately? It seems like there has been a lot of smoke coming down from their way.

53: It is in fact a weird convention...*however*...putting it in a book as a seeming mystery is pretty wild. And the editor(s) really dropped the ball. Also since it is literally off by *three* orders of magnitude the results of any mixup are just ludicrous .

However, the bottom line is that it shows just how efficient food is as an energy store what with containing a metric fuck-ton of joules in relatively small mass/volume.

60: They put poutine in the calorimeter.

60: New York is upsettingly smoky. I've heard Californians complaining about this sort of thing for years and now I understand and I hate it.

3 beat me to it so I have been replaced by a chatbot.

Don't tell the provost.

63 We had Canadian smoke a couple of weeks ago, but it's gone. Our county health department keeps track of such things, in a unique narrative style. You might want to check on some of the links on filtering indoor air.

https://www.missoulacounty.us/government/health/health-department/home-environment/air-quality/current-air-quality

||

Another mass shooting. This time in Virginia. God damn. Are we ever going to be able to deal with this?

|>

New York is upsettingly smoky. I've heard Californians complaining about this sort of thing for years and now I understand and I hate it.

Yeah, it really is distressing. It has been saddening to come to accept that there is a new smoke season that we need to plan for (Aug-late Oct).

That said, before the fire suppression years of the Forest Service, some 40 million acres burned every year across the West. That was low intensity burns throughout the whole year, not these high intensity wildfires. We should perhaps have been living with more smoke all along.

The real air was the smoke we met along the way.

Haven't the mass shootings slowed down since earlier in the year? Like still more than last year but less than early spring?

This year is running at 2.07 mass shootings per day, and the last 20 happened since 29 May which was nine days ago, so it seems like it's still going along at the same rate?https://massshootingtracker.site/data/?year=2023

How depressing that a site like that exists and is necessary

NMM to the Nova Kakhovka Dam.

73; A rough day for infrastructure fetishists.

What we're getting here in Pgh seems to be much, much less than what NYC etc are getting, but still not good. That said, Kai & I biked to the ballgame last night, and it wasn't discernibly worse than the identical ride taken Saturday or, really, any ride. It's swirly, so it's perfectly likely that we were just riding at times of less concentration, but it was a little hard to square with the level of panic I was seeing online.

Apparently scientists are saying 24 hours of unmitigated exposure to the smoke in NY is equivalent to 6 or 7 cigarettes smoked in a day, which... come on. That sounds very much similar to a single mid-range flight taken any time before 1990 or whenever it was. Those were, to be clear, very unpleasant, and it's good that smoking on airplanes is gone, but people are posting apocalyptically.

It's probably accurate, but the lesson is that six or seven cigarettes in the absence of a habit of smoking cigarettes is probably something your body can handle.

We had Canada smoke from a different fire a few weeks back. It's pretty typical here in the summer as people love to set the backcountry on fire.

I love the link in 66 and I wish they updated it for all weather

I wish my county would figure out another disease to track with poop. The covid tracking is basically flat and very low. Which I appreciate as a breather, but the lack of variation annoys me in a statistical way.

75: Events like this create in people's minds an entirely appropriate sense of dread.

When they started stuff like Earth Day or the EPA, the environmental situation was awful, but there was also a real sense of hope. Not so much any more.

People just aren't going to keep pooping every day if there's no new information.

Air quality in Missoula was exclusively a winter problem when I was growing up. Smoke/fire in the summer has gone from an upsetting anomaly to a predictable mini-season over the last 20 years. People who've been in the West the whole time have something of a boiled-frog attitude towards it but it was a pretty disturbing thing to come back to after 20 years of living elsewhere and not visiting in late summer. Summer weather overall is drastically different now than it was when I was a kid.

Agree with messily that the link in 66 is great. Now we need to know if that's a humanities major writing it.

it was a pretty disturbing thing to come back to after 20 years of living elsewhere and not visiting in late summer. Summer weather overall is drastically different now than it was when I was a kid.

I've lived here full time for almost 30 years now ('93 was the last summer I went home to Jersey), so I'm more boiled frog, but of course one's 20s are highly salient, memorable years. When I look at actual historical weather tables, it really reinforces the impression that this is, indeed, a different climate from the one I thought I was settling in. Not obviously or dramatically, but a clear step to the south. Without actually checking, I bet we're midway between ca. 1990 Pittsburgh and ca. 1990 Harrisonburg, VA.

The summers don't seem that different to me, but I moved in 2003. The winters are very different. It's rarely cold.

78: Are you not tracking polio, then?

Philadelphia has been incredibly delayed with disease tracking via wastewater. Allegedly this is because we were building the capacity to test samples in-house through the city health department rather than using external labs, which will speed things up in the future. I don't have enough knowledge to assess the truth of this claim.

Back on the haze topic, it's unpleasant here in Philadelphia but not as bad as Montgomery County PA immediately to our north, and definitely not as bad as NYC. Reports from Brooklyn friend are that you still get smoke in your lungs even with a KN-94 on outside.

Here in PA, youth track and swimming practices/meets have been canceled for the evening. Not sure if the 6:05 Phillies game will get canceled. I can't imagine why they wouldn't; both teams are scheduled for an off-day tomorrow so easy to reschedule.

82- it's one of the county's air quality specialists. MS/MA, so probably not, but maybe.

I meant the summers in Missoula specifically are different. I feel like the winters here are pretty similar, or at least not as drastically different.

When I was growing up, hot summer days in July and August were in the 80s. There would be MAYBE one or two days over 90. Mostly 70s. You'd start needing long sleeves and jackets in September.

Now it's frequently over 90 in May with long stretches of being over 90 every day, and October is still mostly summer. April, May, and June are also a lot more chaotic. The past three years in a row there's been one week between the last snow and the first 90 degree day- last year in June, this year in May (unless it snows again).

New York summers aren't all that different, at least subjectively, but winters are much warmer. The thing that makes me sure it's a real change is that I can't remember the last time I saw old snow. When I was a kid, snow would fall in December, get shoveled off the sidewalk into mounds, and the mounds wouldn't melt all the way until spring -- they might shrink some and then new layers would accrete on top, but they'd basically just be there getting dirtier until late February or March.

For the last twenty years, though, snow lasts maybe a week at most before there's a long enough stretch above freezing to melt it all.

Nebraska, you can still freeze your balls off even in a winter coat with two fleeces under it. In Pittsburgh, I wear a fleece and a windbreaker for most of the winter.

These NYC photos really are stunning.

87 It's Sarah, a scientist. People in other places should do what she does.

https://warzel.substack.com/p/air-will-find-a-way

PAGING DOCTOR STORMCROW! APPLIED GEOGRApHY NEEDED IN THE UKRAINE DISCOURSE!
-How deep and soft can we expect the mud in the drained reservoir to be?
-How much of the river's sediment load is trapped in the other dams upstream of Kakhovka?
-How is the mud distributed? Presumably mostly at the head of the reservoir and behind the dam wall? Where will the mud be shallowest?
-Can we expect that mud to dry out over the summer?
-What kind of course will the river take once the reservoir has drained? Neatly resumed pre-dam course? Some kind of braided mess? Massive deranged swamp covering the whole lake bed?
-Could the Ukrainians plausibly choke the river discharge at will using the upstream dams they control?

Further to 92, she has a masters in environmental journalism from Michigan State.

66: 403 Access Denied

I guess Europe is too topless to be enlightened about the weather in MT.

My goodness. DC is much worse today than yesterday.

Here is Airnow, a site that is apparently government run, if you want to learn the air quality number in your area.

It's only unhealthy to breathe here if you're sensitive.