Comment by rdiddly
3 years ago
Last thing I saw was "The son of two high school English teachers, Pontzer grew up on 40 hectares of woods in the Appalachian bla bla bla..."
You know that sentence. So many articles have that sentence. The one where it veers off course to talk about the (usually uninteresting) person who found the thing the article is ostensibly and nominally about. Am I alone in being perfectly happy reading about science for the duration of an article, and not particularly caring about the scientist and what breed of dog he has etc.?
Whenever there's insignificant science making "controversial" claims like "you can't lose weight with sport" journalist will provide storytelling instead of science.
The whole premise of that "You can’t exercise your way out of obesity" is that you burn more calories when you start running than later on as you continue. What's so controversial about it? It's normal that body optimizes for energy expenditure that's why we develop strength, endurance etc. And that's why when you're truly doing sports you're running more, faster, lifting more, etc. you increase the challenge to give your body greater burden to carry.
If I start lifting 20lbs and 3 years later I still lift that of course I won't lose weight.
You have that same thing in other disciplines - "historians" who can't make any meaningful contribution yet want to make name for themselves go on claiming "that or that king was gay" or "vikings were trans" or whatever fits the trending topics of the day and they get media exposure.
> What's so controversial about it?
Your body burns incredibly more energy just "being there" than you burn moving about.
Exercise certainly increases your calorific expenditure, but it's much easier to ingest less fuel than to run 5 more miles because you had a portion of fries.
There are benefits in exercise, apart from the obvious ones the raising of the baseline metabolic expenditure, because muscles are expensive to maintain, but again, if you want to lose weight eating one less portion of fries is easier and takes less time than going for a 5 mile run.
> body optimizes for energy expenditure
No that's wrong, it optimises for energy maintenance. Genetically it's better to maintain fat and survive the next famine, than burning all the energy today. Which is why the body is so efficient at _not_ losing weight unless it has no other choice to maintain homeostasis. And thus to lose fat, we need to preferably eat less, not run more. Do both, and you'll do great.
While I agree with that excerice isn't a very efficient way to lose weight I think that you understate how many calories running burns.
When I run consistently I need to eat what feels like a lot more food to maintain my weight. Running burns about 700-800 kcal per hour for me (I have a quite small build) which ends up quite a bit over a week (I run 5-8 hours per week).
24 replies →
> Your body burns incredibly more energy just "being there" than you burn moving about.
Bullshit. My body goes through 1600-ish kcal on an idle day.
If I go out on a 4h steady bicycle ride (= 100km), I go through 2000 kcal on ride alone. That’s much more than my body burns just being there.
21 replies →
> If I start lifting 20lbs and 3 years later I still lift that of course I won't lose weight.
This example is kind of bad :-)
You don't lose weight by lifting weights, in general.
And amusingly, if you did want to lose weight by lifting weights, your example is precisely how you <<would>> lose weight. You lift 10kgs over and over and over again, or even better, incorporate the extra weight into some sort of cardio routine.
Instead of increasing the weight, you'd increase the reps or the motion you use for lifting weights.
One does not only lose weight because of the exercise itself, but by retaining muscles since they are a major calorific expenditure factor. Large mass muscles burn calories e.g. glutes
1 reply →
This is not accurate, weightlifting with a reasonably difficult weight for your strength level is one of the most effective ways to burn calories.
15 replies →
I'm pretty sure you have that backwards. What you're describing is effectively trying to turn weigh-lifting into cardio, while most research shows that interval training / weigh training are more effective at burning calories than steady-state cardio.
> If I start lifting 20lbs and 3 years later I still lift that of course I won't lose weight.
I don't understand, your body optimizes, and uses 20% less energy. However, you are still using the 80% which is more than not doing anything at all don't you?
Plus at least for running you can just run faster and make up for that 20% effioency gain. I personally do not think excercise is very good for losing weight but that particular argument is BS.
> Whenever there's insignificant science making "controversial" claims like "you can't lose weight with sport" journalist will provide storytelling instead of science.
I think this is quite the claim, maybe you should provide some evidence. I think it sounds reasonable to assume that those that lack the scientific basis will use storytelling to convince, because they have nothing else. But I'm not sure about the reverse.
Here we have storytelling, but the story is peppered with some data from experiments that support the claim. So how does storytelling imply lack of evidence?
No you’re not alone in it. But yes there are plenty of us who find it a really effective way to get drawn into an article. It works for me through infectious enthusiasm. If I have a sense of a curious mind trying to figure something out, I get interested in the thing they’re trying to figure out too. Bill Bryson’s Short History of Nearly Everything made me interested in tons of scientific areas I never considered interesting before, because it is all about fascinating people repeatedly failing then eventually succeeding in understanding some natural phenomenon. The human angle is the perfect gateway for me. Maybe if I had more discipline I could just force myself to start reading a dry, factual scientific article in a field of little immediate interest to me until I start to notice data points that pique my own curiosity. But it’s so much easier for me to get infected with the curiosity of some compelling character who is overflowing with it. And it sticks better in the memory.
If it’s a field I already have a strong personal interest/background in, then I generally just want to get straight to the findings, so in that case I wouldn’t read a long form popsci article like this one. But calories/diet/exercise stuff? For me that’s not very exciting, yet I do want to understand it better (for practical benefit), so anything that helps pique my curiosity in it is good.
> Bill Bryson’s Short History of Nearly Everything
I have read that book multiple times, first when I was 10. It left a huge impression on me. I heartily recommend it.
Bill Bryson in general is a great author. One of my favorites.
There are different kinds of articles out there. Some are about getting the information across, some give the context and the story behind it. Some aren't even about the science at all. I wouldn't compare John McPhee to a Geology textbook. And I wouldn't say it's correct that this article was "ostensibly and nominally" about just calories. The headline is "Scientists bust myths about calories". It's about the scientists as much as it is about the calories.
Anyways, that's all to say, I do enjoy articles like this that give context and explain not just the research but the people behind it. Let people enjoy things.
Am I alone in being perfectly happy reading about science for the duration of an article, and not particularly caring about the scientist and what breed of dog he has etc.?
science.org is like a good quality popsci magazine. People read it for the science, but also for entertainment. The aim is to make science engaging to non-scientists. If you just want the science then you should be heading to Nature or Arxiv.
Actually, both Nature and Science call themselves 'magazines', and they have plenty of pop-science-compatible content, mixed with primary literature.
Arxiv needs to be taken with a huge grain of salt because much of the content is not peer-reviewed.
Am I alone on HN in being perfectly happy to get a bit of insight in the human nature of scientists doing impressive work?
Seriously, yeah, some of this stuff may be fluff, but it does matter. Two scientists can be doing equally good work on a similar topic, but one's work makes headlines and the other's gets buried in some obscure journal and forgotten. Why? What can I do to be more like the former rather than the later? One popular magazine article may not give me that insight, but after a bunch of them patterns emerge.
HN is both tech oriented, full of nerds and people with probably lower than average EQ and also an echo chamber in the sense that nerds at some point start geeking out about this and turn it into virtue signaling.
"Look how bad I am at reading people."
"I'm worse than you at..."
"I'm so bad that..."
https://www.youtube.com/watch?v=ue7wM0QC5LE
> Well, when I say 'house' it was only a hole in the ground covered by a sheet of tarpaulin, but it was a house to us
Seems a bit harsh to jump right into an accusation of the dreaded virtue signaling!
Would you agree at least that there’s a balance to be made, and articles could lean either on the too-dry side (all facts and no color) or too-fluffy (facts buried in an avalanche of color)?
I don’t want no color, but I agree with others here that many pop science articles seem to lean very heavily towards fluff these days. Online recipes are a more extreme example -- it’s increasingly hard to find the actual recipe these days.
Titles matter. A title like “A Short History of Nearly Everything” pretty clearly tells you that there’s going to be a lot of storytelling. But if you have a clickbait title like “Scientist busts myths about how humans burn calories”, you ought to deliver on those promised myth-busting new facts!
I think it’s interesting to consider a) is this a real change; b) if so, is it driven more by culture or by technology? (i.e. by ads)
1 reply →
You are not alone. A subset of HN users have an identity marker around having a preference for "explicit material claims and nothing else for color or emotion". They then enjoy signaling this preference to their tribe in the comments. I've found it best to ignore them.
I would agree in principle, but puffed up articles rarely give you the kind of insight you are referring to.
I doubt what colour dog they had and their preference for cornflakes instead of oatmeal at buffets is going to let you enter their scientific mind as such.
I was thinking the exact same thing. I stopped reading after three sentences and couldn't find the scientific paragraph right away so I just when to the HN comment section to see if some could give me the gist of the story.
Many of these articles are written to increase volume, but its my take that no one really cares about who ate what this morning. Scientific papers often do the same thing
I think they're written this way for seo optimisation.
That's funny. SEO optimization for keywords "woods" and "Appalachians"? :-)
No, it's written that way to turn the person into a) a character, b) a relatable character.
Regular people like stories, not dry facts piled up.
2 replies →
You’re not alone. This reads like one of Grandpa Simpson’s stories about onions.
I also like the way they’ve broken my browser’s reader mode too.
I was once advised to read the opening and the last 2 paragraphs of articles like this. If there’s no actual content there then move on without reading.
Thats generally what I do, I skip to the last paragraph, read it, and if I've missed any important context (that I care about of course), I go up one, rinse and repeat until I get the point they're making, or until I get bored and decide I can quite happily live my life without knowing the busted myths
It is a genre. You may prefer drier style. On occassion, I do as well. But sometimes I enjoy fluffy writing like this, especially in the human-contact-starved covid days.
I would say "live and let live". There is enough written content on the Web for everyone to find their niche.
Right? Can someone just list the myths and their refutations so I can move on with my life?
It would be nice if articles actually made use of the possibilities of the Web: show me just the brief summary of what it's about, and let me click links to choose whether to read more about the science, or more about the 40 hectares of woods where the guy grew up.
You could go to the primary literature and read the abstracts.
This is a reason we shouldn't feel ashamed about developing a habit of skimming. In many cases it's simply a proper adaption to an incredibly low signal to noise ratio in most articles you're exposed to on the internet.
Props to this.
I have developed a habit of skipping to the end of the whole page, and reading towards the beginning, paragraph-by-paragraph. Works..sometimes :D
Paste bin copy with most of the fluff paragraphs removed: https://pastebin.com/LXtLkdgN
It’s maybe three paragraphs total of his life, then it’s back to the findings. I skipped those paragraphs too.
The article is essentially a summation of Pontzer’s past decade or so of work (and I do find the work to be really interesting), so if you want just the meaty bits you can probably try reading the original papers.
Skipped it totally. That's the part of scientific American intended for bored readers more interested in gossip. The part after it is quite cool. The conclusions are basically drawn from how much CO2 is exhaled. So the key contribution seems to be the data collection methods.
It's apparently the canonical way all long story/articles have to be written. It's as if those people wanted to be authors instead of journalists...
I didn't know this about myself, but apparently I feel the same because I exited there after finding out very little about the claim in the title of the article.
Could machine learning could help with this? Could some algorithm be created to strip the padding from such articles just leaving the useful stuff?
Yes. See https://smmry.com/https://www.science.org/content/article/sc... for example.
Thank you! Works very well.
I would expect machine learning to be more useful in generating this kind of fluff.
Seems like a lot of work (and a lot of time to get working reliably) for a relatively minor issue that can be mostly fixed by skim reading
No.
If you want this to change, change the source, don't patch cargo cult on top.
You're wrong, it is very well possible with algorithms, and it has been a thing for many years already. For example check the Auto TLDR bot for Reddit. Here's an example of what it could look like for the article posted here: https://smmry.com/https://www.science.org/content/article/sc...
It's not even hard to do, you mostly have to count token frequencies which is hardly machine learning. Although you could also do it with BERT.
Yeah, this is a really strange article. It reads more like a human interest story about this one scientist's bio, with his work sprinkled in.
I'm not a fan of this style of reporting. It can be interesting to learn about a person's life if the subject is especially notable, but given the headline it feels a little like a bait and switch.
Indeed I find a lot of NYtimes articles are filled with fluff like this and I find myself skipping to the core info.
Yes there is a name i think for that rambling on with insignificant details about the person
At best it should be seen as a warning for possible biases in the science due to the person's background or upbringing, or random life anecdote.
I'm really having a hard time reading this. They lost me even before, when they said he likes hiking and he's not fat. Good science can come from the worst people. Show me the data, no need to sell me the authors.
I hate that, and I usually just skip that part. No point fighting it, I guess.
I had a very similar outburst recently :p
https://news.ycombinator.com/item?id=30344510
You could read his papers if you're only interested in the science?
I just skipped over that part...
Verbiage is proof of work, and journalists need to make a living, just the same as planning-to-plan product managers.
You can train yourself to skim for maximal extraction per minute. I burned through the article in about 3-4 minutes. I tend to get bogged down by my software developer's close-reading instincts.
I hate this type of journalism. Article writers are driven by metrics given to them by publishers. In this case, the metric they might be trying to optimise is time spent on the website.
This type of journalism works well if the backstories are tied well together to the main story, and is done by a good writer. But when average writers do it, the stories are tedious, and hard to read.
>Article writers are driven by metrics given to them by publishers. In this case, the metric they might be trying to optimise is time spent on the website.
It's not that. It also happens in written media. It's not just as filler either. They have the dellusion that this makes the story better...