Before I clicked the link I thought "I bet they got facebook warnings for some junk posts, but they'll play leaving facebook off as a moral decision." But I was surprised to be so wrong. With phrases like "corrupt academic organizations" they aren't even even trying to hide their culture-wide embrace of pseudosciences.
There are corrupt academic organisations. The idea that 'fat makes you fat' was exposed recently as having started with Harvard researchers being paid by sugar companies. Most nutritionists know that 'fat makes you fat' is simplistic (it ignores digestion) and accept that sugar contributes more towards obesity than fat does now.
Weirdly, FB banned a group advocating against 'fat makes you fat' in 2019.
> Recently, Facebook deleted without warning or explanation the Banting7DayMealPlan user group. The group has 1.65 million users who post testimonials and other information regarding the efficacy of a low-carbohydrate, high-fat diet. While the site has subsequently been reinstated (also without warning or explanation), Facebook’s action should give any serious person reason to pause, especially those of us engaged in activities contrary to prevailing opinion.
I'd be careful generalizing this beyond Germany; DE is a weird place to do studies about AC because it's so much less common there than it is in e.g. the USA.
However, anecdotally, it's not just women. It's basically everyone except for obese men.
Tough issue, especially since we don't allow partial nudity in work environments and it's much easier to put on a sweater than it is to lose weight (or, in some cases, muscle).
Yeah it's to do with heat retention and surface to volume ratio. Bigger things have a lower surface to volume ratio so lose heat more slowly. I wonder if the particular study controlled for weight/height.
> Also, just my own opinion based on observation, why do cemeteries get all the best real estate?? Dead people can't enjoy that view.
Old cemeteries are de facto parks in cities that grow up around them. They're quite nice and a decent use of land in areas that don't have the sort of pressing housing needs that SFBA has.
In Munich, I lived near the "Old Northern Cemetery", where the last person was buried in 1944, but I only recall seeing graves there that are far older than that. It is now officially considered a park, and people use it for jogging or just lying there to relax:
https://upload.wikimedia.org/wikipedia/commons/4/48/Alter_No...
It's quite a weird feeling, looking at those gravestones (sometimes quite beautiful and impressive, to be honest), and knowing that, for almost all of them, no one is alive anymore that could have actually known them.
Hmm, why? As I said, the cemetery has not been in use for a very long time now. Long enough that the corpses have probably rotten away completely, and nobody knows the people who laid there anymore. So I don't think anybody is going there for grieving.
The Wikipedia article itself mentions jogging as one of the uses, so I'm not making it up (at the same time it mentions "with due respect", and personally I also don't think jogging is "disrespectful"): https://en.wikipedia.org/wiki/Alter_Nordfriedhof_(Munich)#Pr...
Not everywhere, and real estate is precious. Near by we have a Chinese cemetery, complete with coloured neon lights, karaoke machines, outdoor fitness centre, and jogging track going through the tombstones. You don't want to get bored in the afterlife.
Warning: it's ridiculously expensive. You need to be charging your users 4+ figures per license for your software in order to be able to pay Wolfram their cut.
The open/libre stuff is a huge red herring -- cost is the real bottleneck. It doesn't make any sense to use Wolfram/Mathematica unless there's a built-in function that you NEED, for which there is no alternative, and which requires a team of PhDs to re-implement.
E.g., even a non-production single user corporate desktop license is $3K.
But the "its" that the article is referring to is the Wolfram Engine. It doesn't say that is all the algorithms in Mathematica. And the Mathematica website lists Algorithm Base and Wolfram Engine as two separate line items, which is at least surprising if one is contained in the other.
I actually believe the Algorithm Base is probably included, but it's definitely anything but clear from the website or any of the press release. The Mathematica website makes them seem like two completely separate unconnected components.
It's really unclear what is included in Wolfram Engine and what is included only in Mathematica... the above was just one obvious example. It would be very nice to have a concise list.
If one of those companies decided to drop decent money on turning SciPy+Jupyter (or Julia) into a proper high-performance CAS, Wolfram would be in huge -- maybe existential -- trouble.
mathematica is already in existential trouble. yes some theoretical physicists supposedly use it (i used it undergrad to do all of the annoying integrals in e&m) but that's such a small market it's basically irrelevant. i can almost guarantee there is no production code running that uses mathematica in some way.
The KKK and the nazis managed to form and grow without the help of the internet. Also, finding other people who enjoy talking about cars or writing fan fic doesn't have this sort of harmful in-group/out-group dynamic.
So I'm a bit skeptical of the narrative of the piece, especially because there's no actual evidence provided.
My skepticism extends to the broader narrative of this newsletter.
Illich's alternatives -- especially the conviviality stuff -- always struck me as dangerously Utopian: if only we were all the same, then everything would be great.
He's like that well-meaning stoner who asks "why can't we all just get along" and sort of shakes his head and tells you that you don't get it if you ask how, concretely, we're supposed to "just get along" in Gaza or Darfur or Kashmir or any other place where there's a lot of zero-sum resource/power allocation underlying centuries of conflict. The dismissal of real and concrete harms on both sides of conflict is at least unhelpful and possibly harmful.
Conviviality is a nice sentiment, and the world would perhaps be a better place if everyone shared that sentiment. But sentiment is a starting point, not an actual solution. The world's problems are usually too complex to be solved with pure sentiment, and things will go wrong in unexpected ways if you try.
One concrete example: the modern commercial internet's ad-driven information economy elucidates a major flaw with Illich's "Learning Webs" from Deschooling Society: the company that owns the platform just happens to be an ad company. It's a flaw that even the strongest critics of Illich could never have anticipated in the 1970s.
The point is more general: convivial societies only work if everyone is convivial, and there will always be insanely inventive non-convivial people. Even people who are more-or-less decent folks and even people who adopt slogans like "don't be evil" will end up throwing wrenches in your plan.
>Also, finding other people who enjoy talking about cars or writing fan fic doesn't have this sort of harmful in-group/out-group dynamic.
Purely anecdotally, but I beg to differ. There is more than a little bit of tribal hostility on e.g. Tumblr around various fandoms.
To me, the real issue is that surrounding yourself with like-minded people only teaches you to interact with people you primarily agree with and are comfortable with, rather than the more valuable skill of interacting (civilly) with people you disagree with.
> that well-meaning stoner who asks "why can't we all just get along" and sort of shakes his head and tells you that you don't get it if you ask how, concretely, we're supposed to "just get along" in Gaza or Darfur or Kashmir or any other place where there's a lot of zero-sum resource/power allocation underlying centuries of conflict
It just hit me while reading this that stoners are (often) slackers, and it really does make less sense to fight over resources instead of sharing if you start with this mindset.
The Nazis in particular utilised the mass media of the times, most especially audio, public address, and radio, though also video newsreels and cheap paperback publishing, to spread their message.
During and prior to WWII, german advances especially in audio capture (mic), recording (mag tape), playback (speakers), and broadcast & receiver (radio) were decades ahead of the Allies' own technology.
You don't get the Nueremberg rallies without high-quality mics, massive public-address ampifiers and speakers and cinematgraphy (Leni Riefenstahl). Hitler's ability to broadcast live-quality radio addresses across Germany stumped Allied intelligence -- their best recording technologies were wire recordings and low-fidelity vinyl, both with very obvious artifacts (wire recorder demo here, at beginning of video: https://youtube.com/watch?v=90ihiTwJPCc). The only way to achieve this quality otherwise was to be in the studio, and this clearly wasn't possible.
("Wearing a wire" refers to wire recorders.)
After WWI, Bing Crosby, with support through military and government intelligence, was instrumental in developing US magnetic audio and data tape technology, through AMPEX and 3M.
It was German use of mass media -- though in WWI-- that turned the meaning of 'propaganda" from literally a holy undertaking (the propagation of faith by the Roman Catholic Church) to its present pejorative sense. WWII Nazis capitalised heavily on and greatly extended earlier practices.
Tactical use of radio communications also playe a decisive role in war -- the key differentiator between Grman and French armour in the Battle of France was that German tanks all had radios, and could respond to developing circumstances. French tankers could only play out prescribed batle plans, or act independently and uncoordinated with all other units.
There is actually a long history of the disruptive (and often highly harmful) effects of new and especially mass media, and numerous historical inflection points can be traced to revolutions in information and communications technologies: moveable type and the Thirty Years War, vast advances in printing technology and literacy and the revolutions of 1789-1914, and later ("the long 19th century" https://en.wikipedia.org/wiki/Long_nineteenth_century), yellow journalism and the Spanish-American War, WWI, tje Russian Revolution, WWII, the Chinese Revolution, Father Coughlan, Jim Crow, Civil Rights, the Vietnam War ant ant-war movements, the rise of hard-right talk radio and cable television, and lately, social and mobile Internet.
Today's Nazis and KKK are using the Internet. It's the cheap, high-fidelity, visceral-imact mass medium of the age.
> 1. It's not clear that AP Stat disadvantages students versus AP Calc.
This is what stuck out to me.
Isn't AP Calculus the obvious choice if you have to choose between AP Stats and AP Calculus?
1. Calculus is really, really useful. Maybe AP Stats is also useful, but singling out Calculus as an example of useless signalling sets off really loud alarm bells.
2. If you want a STEM/Engineering degree, at least one Calculus course is required. More importantly, often a long sequence of 2-4 courses (Calculus I, II, III and ODEs) are required. Because of that long list of sequentially dependent required courses, getting one or two calculus courses out of the way in high school is enormously useful (like, "graduate a semester earlier for each course" useful). AP Stats is not at the head of this sort of long sequential course dependency.
3. The AP Stats course has a major disadvantage: lots of colleges/majors that require a stats course don't accept AP Stats as credit because they require a calculus-based statistics course.
> 1. Calculus is really, really useful. Maybe AP Stats is also useful, but singling out Calculus as an example of useless signalling sets off really loud alarm bells.
34 years old. Been writing software since I was 14. Used to know some calculus and sometimes poke at picking it back up because I feel like I "ought to". Have usually been the one to tackle tough or odd problems where I've worked.
Haven't once managed to find a reason to use calculus for anything whatsoever. Not a damn thing to differentiate, not a damn thing to integrate. I think the need for it is in a very, very narrow slice of all jobs, even in "STEM".
Statistics is 100% for sure more useful to me, in everyday life and at work. And I've not even worked on anything especially stats-ish, it just happens to come up a lot. That's the thing I really ought to work to get better at.
While people often don't have to do the explicit differentiation and integration taught it calculs that doesn't mean you aren't using knowledge learned in a calculus class.
I often think about rates flows and how they relate to each other. For example to do monitoring we count the total number of messages received in each task at a given time, then use the diffrence to get a rate. That is basically differentiation and understanding calculus makes it easier.
The other thing you acquire from doing Calculus is problem solving experience, figuring out how to apply various approaches to a problem to get to a desired solution. In this way calculus strengthens your brain similar to how a sports player might do weight lifting even if their sport doesn't involve lifting heavy objects.
I think lots of people understand rates without doing calculus. For instance distance vs velocity vs acceleration is a pretty simple and intuitive idea to understand.
There is an enormous amount of evidence that weight lifting transfers to sports. But I don't think there is much if any evidence that calculus transfers to general problem solving.
> 34 years old. Been writing software since I was 14.
We have very similar profiles.
I use calculus every day.
> Statistics is 100% for sure more useful to me, in everyday life and at work.
I use statistics every day too, working on ML systems.
> That's the thing I really ought to work to get better at.
The calculus comes in pretty damn quick. See also the third point in my original post; concretely, every ML course I've ever seen requires Calculus as a pre-requisite.
It really depends on the field. In many branches of science, statistics is really important and calculus not really at all. I'm a biomedical scientist and I use statistics every day to interpret the results of experiments. Calculus, not so much. And yet I was required to take three semesters of calculus as an undergrad, and while I took a statistics course then (and more in grad school), that was optional. It really should be the other way around.
What a weird and unnecessary rant.