HNHacker News
TopNewBestAskShowJobs

zazen

394 karma · joined November 5, 2014

submissionscomments
zazen··on There Was a Time before Mathematica (2013)
>I was serious that I can separate the person, so that's an existence proof.

Those of us who are mere mortals can't reliably predict ourselves to achieve perfection on complex, subjective tasks like this.

You find your own self-confidence to be highly convincing to the point of mathematical proof. Please have another think about that. And about making "clever" four-word dismissals of other people's comments.

zazen··on There Was a Time before Mathematica (2013)
I've tried to suggest that there are difficulties with separating the person from a personal biographical essay in a way that there aren't in separating the author from an abstract art work.

I don't suppose you feel like engaging with my suggestion with something more than this blunt, unsupported rejection?

If you merely mean that when discussing such a piece we can remain generally calm and give room in the discussion for things other than the author's arrogance, then I agree.

zazen··on There Was a Time before Mathematica (2013)
I would agree if discussion of Wolfram's narcissism were being dragged into a discussion of Mathematica that didn't otherwise feature Wolfram personally; but that's not the case here. The link is to a text written by Wolfram, in substantial part about Wolfram, in which Wolfram's narcissism is on display. We can't separate the person here.
zazen··on There Was a Time before Mathematica (2013)
>nobody is even asking you to do it

It may not have been your conscious intent, but your previous comment was in fact in the imperative:

>Don't implicitly approve Wolfram's narcissism. Just don't.

zazen··on Could 'Oumuamua be an icy fractal aggregate ejected from a protoplanetary disk?
Well, "my" theory is also pretty wrong if it doesn't allow for the human civilisation that built Stonehenge.

A theory that is wrong will sometimes predict correct things by accident. That's unavoidable, but we a) have statistical methods for modelling how often things go right by chance, and b) have a philosophy of science in which theories are forever tentatively held: your aliens theory might beat any theory which says Stonehenge is impossible, but will soon be replaced by a theory which explains Stonehenge more economically.

zazen··on Could 'Oumuamua be an icy fractal aggregate ejected from a protoplanetary disk?
We're going to need to think carefully about what is covered by "this sort of thing".

If your ancient aliens theory didn't actually predict "Stonehenge", but predicted that the aliens would have left "all sorts of cool weird stuff", and then you point at Stonehenge and say "see, that's the sort of cool stuff I mean", then I'm no longer impressed by the predictive power of your ancient aliens theory.

Predictions have to be precise for a theory to be supported by them: they can't be vague generalities. cf. astrologers, who "predict" the future by predicting nothing.

zazen··on Could 'Oumuamua be an icy fractal aggregate ejected from a protoplanetary disk?
If I had confidence that you'd formed that theory without prior knowledge of the existence of Stonehenge, then the fact that you'd predicted Stonehenge would indeed be making your theory look pretty good!
zazen··on What Even Is a Number?
Well, yeah. Axioms are not deduced, they are the starting points for the various theories that are deduced from them. (There is not much you can deduce ex nihilo.)

Did you have some further point?

If you were hoping that mathematics would have - or thinking that it _should_ have - some ontological foundation that is not in this sense "made up", I'm afraid it simply doesn't.

zazen··on What Even Is a Number?
He doesn't necessarily have to. But when he defines "natural zero" as "the empty set", and someone asks "what's the empty set?", and you say "natural zero", you're being circular.
zazen··on What Even Is a Number?
GP understands the notation; he means to cast aspersions about the axiomatic project.
zazen··on What Even Is a Number?
That is a definition of "natural 0", not a definition of "empty set". GP was pointing out that GGP was using the concept of "empty set" to provide a definition for the natural numbers, without having first provided a definition for the empty set.

Replies that consist solely of throwing a quote at someone are kind of rude even if you're in the right.

zazen··on Neural Networks Are Essentially Polynomial Regression (2018)
The first person who made a convnet had an intuition. That is to say, an informal heuristic that gave him reasonable grounds to suppose it might work.

This is not "theory".

You do not understand what "theory" means in an academic context.

zazen··on Neural Networks Are Essentially Polynomial Regression (2018)
That video is an introduction to the concept of CNNs for beginners. It explains in informal terms the structure and mechanics of a CNN, and probably gets around to discussing why it might work in very broad terms.

This is extremely far from what academically inclined people are talking about when they are using the word "theory".

People are hoping to see, ultimately, mathematical proofs that neural networks can e.g. efficiently learn target functions given a certain amount of training data and a certain number of trainable parameters.

To address your chemistry analogy: no, it doesn't discredit synthetic chemistry as an enterprise, it just means there is a gap in the theory. Similarly, the people bemoaning gaps in the deep learning theory aren't trying to discredit deep learning as an enterprise: the practical successes are very real; the theory is just lagging, as it often has in application-led fields.

zazen··on Neural Networks Are Essentially Polynomial Regression (2018)
There is plenty of room to be dissatisfied with the current state of deep learning theory without making absolute claims that there is "no theory" whatsoever, which GP clearly isn't.

If you wish to challenge GP's comment, you could do so constructively by discussing any specific theoretical contributions you feel GP may have overlooked.

zazen··on The Machine Stops
Yes, that Oliver Sacks. Yes, he did die.

Strange of the New Yorker to publish this 3.5 years posthumously with no explanation.

zazen··on How Craig Barton wishes he’d taught maths
You talked about having "a member" and getting to every other member with scalar multiplication only. But even given a 2-member basis for R2, how are you planning on using 2 members, with scalar multiplication only?

I'm afraid you have badly misremembered this stuff.

zazen··on How Craig Barton wishes he’d taught maths
>If you have (2,3,4) and want to navigate to (5,6,7) who is also in your space and you have scalar mult as your tool of choice then mult with 2 gets you to (4,6,8) but then you are stuck. Soon you realize no matter what you do you can’t navigate that space without fractions.

One of us is very confused. It seems to me that I also can't get from (2,3,4) to (5,6,7) by pure scalar multiplication even if fractions are allowed. If I pick a scalar factor of 2.5 to make 2 -> 5 work, then I get (5, 7.5, 10). If I pick anything else, the result won't start with 5.

>>A working definition of a space might be - you have a member in that space, you can get to every other member by just scalar mult.

Really no. You can only access parallel vectors by scalar multiplication. E.g. if your vector space is R2, given a starting vector and scalar multiplication, you can anything in a line with the direction of that vector, but nothing pointing in a different direction. That's more or less why it's called "scalar" multiplication - it scales the original vector, but doesn't change its direction.

zazen··on How Craig Barton wishes he’d taught maths
> It's enough to know that multiplying integers by non-integer scalars can give non-integers

Not quite: you also need to know that multiplying by integer scalars instead isn't an option.

The question as posed asked as to use the "obvious" choice of scalar multiplication, and to a student who hasn't yet taken the "field" part on board, it might seem obvious to achieve closure by using the integers for scalars.

zazen··on How Craig Barton wishes he’d taught maths
Thanks for linking that, it's a great read. I really should read more of Gowers' posts.

The phrase "memory works far better when you learn networks of facts" was a happy find - I've never been able to express that idea so concisely.

I remember discovering they'd moved "differentiation from first principles" away to a further-maths module, as if it's a peripheral, difficult little oddity for the keen kids to hear about. It was the surest, saddest sign that the powers that be had given up on genuinely educating the average A-level maths student.

zazen··on How Craig Barton wishes he’d taught maths
> Agree fully with Barton and OA here. Until recently I taught GCSE Maths re-take students aged 16 and over in a further education college. They were constantly tripping over really quite basic little skill issues and that prevented them from seeing how to tackle the longer and more complex problem solving questions.

I also agree fully. A little while back I did some support tutoring for A-level maths students. The number of students who turned up who mysteriously "had problems with longer questions"... I wish I'd known the example of calculating the perimeter of the rectangle with fractions. That would have really helped explain why the problem wasn't really the length of the question, it was the fact that the student had never properly learned the component skills separately.

Unfortunately, the problem of building impressive-looking edifices on shaky foundations is absolutely endemic in British high-school maths teaching. Thousands of students who never quite understood fractions are "learning" calculus through being taught recipes, and the easier exam questions are formulaic enough that they get through with Cs at least, without any mathematical understanding.

The A-level statistics modules, in particular, have very impressive _sounding_ syllabuses. Students learn T-tests, Chi-squared tests, all this sophisticated statistical machinery. If all these students really understood this stuff, Britain would have a vast army of highly trained statisticians. But nothing of the sort is true, of course: students are just learning a recipe for processing numbers. I can't imagine the carnage if a statistics exam asked the students to write an essay explaining the principle by which a T-test works.

Pardon my rant, this has been on my mind for a while.

zazen··on How Craig Barton wishes he’d taught maths
But if you don't know about fields, you could use the integers instead of the reals for your scalars, and imagine you've made a perfectly good closed vector space that way.
zazen··on Stop Reading News (2013)
This has a name: https://en.wikipedia.org/wiki/Gell-Mann_amnesia_effect

Although I think it bears saying that journalists really might be doing a better job of reporting on the politics that is their own main interest than on the science stories which are far outside their own interest and expertise, and which they may well regard as lower priority for accuracy.

zazen··on Cheap Chinese electric cars are coming to the US and Europe
Are these road-legal in the west? If the off-road farm and construction-site applications are the only use-cases, that would seem to rather limit the impact on the market.
zazen··on Foundations Built for a General Theory of Neural Networks
Brunelleschi was working at the height of the Renaissance, in the early 1400s. They weren't short of book learning in renaissance Florence, but that's why it's called the renaissance. I imagine medieval cathedral building, e.g. Notre Dame starting more than 200 years previously, might have looked very different. They couldn't build a dome like Brunelleschi back then.
zazen··on Foundations Built for a General Theory of Neural Networks
Here are the two papers linked towards the end of the slightly rambling article:

https://arxiv.org/abs/1705.05502 https://arxiv.org/abs/1810.00393

zazen··on Scientists cut the tolerable intake of PFAs by 99.9%
There's better coverage of the lawsuit here: https://www.bloomberg.com/news/articles/2018-11-29/3m-dupont...
zazen··on The China Conundrum – are we headed for a supply chain meltdown?
@evenequator is not using the statistic he quotes correctly. From the article:

>A recent UBS China survey of 200 manufacturing companies with significant export business or supply to exporters revealed the trade war has had a negative impact on 63 percent of those businesses.

>A quarter of those affected have cut jobs, 37 percent have moved production out of China in the past 12 months, while 33 percent plan to move in the next 6-12 months.

So it's 37% of 63% of a sample of 200 particularly strongly affected businesses, not at all 37% of global CEOs/businesses as GP was leading us to believe.

zazen··on Spain will soon overtake Japan in life expectancy rankings
The point is, the parts of traditional medicine that work are adopted by modern medicine. Meaning, although they're still part of traditional medicine, they're no longer distinctive of traditional medicine. Traditional medicine is distinct from modern medicine only by virtue of the parts that don't work.

This is what is meant by the "traditional medicine that works is called medicine" soundbite. The soundbite on its own condenses the argument very imperfectly, and people should be wary of using it.

zazen··on Why American cities are weirdly shaped
I sympathise with @chrisseaton's frustration. When you make a comment highlighting something that looks superficially similar but is actually irrelevant, that's going to tend to derail the conversation. If your comment was unworthy of a parry, maybe that's a hint that it was unworthy of being made in the first place. Please make an effort to be part of the signal not part of the noise, even if your brand of noise is inoffensive.
zazen··on Asterisks in Python
I'm afraid you've misunderstood olavk's point. 'The "quote" you made up' certainly refers to

> "here's this cool feature for doing something you could do a different way that looks like hieroglyphics and nobody understands but you should totally use it because it's awesome!"

not to

> "explicit is better than implicit" and "there should be one and only one good way"

← PreviousPage 2 of 5Next →