bainsfather

I've been considering reading this book - can you tell me what I might get from it in addition to what I learned in undergraduate maths/physics courses?

I found these courses allowed me to do calculations (e.g. contour integrals, conformal transformations) that were useful but always felt like black magic. I don't have an intuitive feel for the subject. I am wondering if this book will help me with that?

2 years in prison, as I understand it.

I read the Quanta article because I am interested in optimisations for deep neural nets.

Am I right in concluding that your work doesn't apply to that area, as it is looking at a much 'easier' case?

None of our animals' waste is any hazard to you, unless you slip on it!

You are right that it can be done badly, as can many other things. I'd say you are actually arguing against badly regulated animal farming (especially since you mention the US), rather than animal farming per se.

I have learned something new today (re manure from dovecotes).

However, I think this is another example of something that is 'just too small to matter' in terms of feeding the world's population today. Often I see 'nice' stories about e.g. using human waste to fertilise fields, that quite frankly do not (usually) scale at all well. We (UK) use tractors to shovel and spread manure - 10s, 100s of tons of the stuff - a dovecote or 10 really isn't going to help, and certainly won't be economic.

It seems a bit like someone telling me I should use an abacus to increase my compute power - 'they were a good idea in the past, why aren't you using them now?'

A bit after halfway, following: "In this case, .. , and we have", I think you need a "-" sign rather than a "+" in your equation?

Thanks. I see it for sale for £4 second hand, so maybe I should buy it for the novelty value alone. I presume I shouldn't try to refresh my java from it though.

Is it still good? I see it was published in 1997.

I like the style of these books, and might read this one - I'm just worried that Java has changed quite a bit since then. Is it still, say, 90% OK?

Yes, this is true.

If we knew that this pandemic came from a lab, that would have massive implications for how tightly we control this type of lab research - both in terms of whether we do the gof research, and also in how we run and regulate the labs. Surely you can imagine the reaction if, hypothetically, the world knew that this coronavirus was made in a lab and then escaped?

I'm amazed that you think otherwise - it's one of the nice things about hn that I occasionally encounter views that are so different from my own.

Can anyone recommend a 'friendly' overview of the 'calculus approach'?

The paper linked in the article from 2003/2008 has been split up into 3 rather technical articles - I was hoping for something easier that just gave me a flavour of the method.

I bought myself a 2nd hand slide rule on uk ebay last year. Price was something like £5-£10 including postage, for a rather battered and 'cheap' example.

I really liked learning about it, seeing how it worked, etc. Such an ingenious idea!

So if nothing else, it might be worth putting them on ebay - there seems to be a market for them.

Thanks! And that is from the "Construction Project Manager’s Pocket Book (Routledge Pocket Books)"? Presumably it has been added to the 2nd edition, because I cannot find that passage in my (1st edition) copy.

(edit - yes it is from the 2nd edition, thanks for your help)

I did a text search of your reference [0] and it doesn't seem to contain information on bribery. (I searched for "bribe", and also "ethics" - which does exist but not in reference to bribery.)

Could you provide a(nother) reference to e.g. your 40% figure, as it is a topic that interests me? Thankyou.

Consider a different case ...

One can make a good case that Britain's guarantee to Poland in 1939 was done badly. Britain failed to convince Hitler that she was serious, and that invasion of Poland would lead to war.

We can agree that the 'blame' falls on Hitler. But one can also reflect on whether Britain's actions could have been better thought out, and maybe learn some lessons for the future.

In the case of Putin's Russia, NATO is dealing with a major nuclear power. Saying that Putin is at fault is fine. I very much dislike Putin. But we have to co-exist with Russia whether we like it or not. I am interested in reading some 'realpolitik' on this topic. In particular, I'd like to understand Putin's and Russia's motives, fears, ambitions, etc.

I thought this was of interest because:

It illustrates the difficulties/risks faced by the insiders who are actually 'doing the dirty work', know what is going on, and want to stop it and/or blow the whistle. When the state is also part of the scam, being a whistleblower is very tough.

I am still amazed that the libor rate - something that was so 'easy' to manipulate - was the basis for many billions of dollars worth of derivatives. The 'foundations' of so many systems that our society runs on turn out to be shaky and easy to rig - not just IT systems.

I followed your link, but am confused and/or doubtful:

how do you replace 150mm of rockwool with 20mm of anything other than a good vacuum? It says that they are 'very low internal pressure' - but if it contains any kind of gas (which i presume it must do - air if nothing else) then the conductivity does not depend on pressure until the mean-free-path has increased to be > the width of the container.

Is this product real? What is it about the physics that I am not understanding?

And how do you maintain the 'very low internal pressure' over the decades that you will need it for?

Thanks.

I stumbled across this article last week. I've always wanted to figure out "for myself" how to solve a Rubik's cube. I've also always wondered what Group Theory was and what practical uses it might have for me - reading basic introductions to Group Theory never helped me much because I was never able to apply it to anything.

This article helped me with both - it taught me something about Group Theory in a 'practical use case'. And it gave me some pointers that helped me figure out how to solve a rubik's cube - several decades after I first picked one up as a kid.

In particular, I learned:

  - that if you repeat any sequence of moves enough times, you will always return to your starting position - with hindsight this now seems obvious, which I guess means that I've internalised some new knowledge.
  - that there are 'macros' - sequences of moves that shift just a few 'cubies' and leave the rest of the cube unchanged (I vaguely knew this already).
  - that there is a logical way to make such macros using 'almost commuting' pairs of sequences.
  - why I could never find a macro that swapped the positions of only 2 cubies (because such a move would be an odd permutation of the cubies, whereas moving a side (the basic move you can make with a cube) is an even permutation).
  - how to physically disassemble a cube, so that I could recover the solved cube - very useful when I was getting started with finding macros.

Hopefully some others will find this as interesting as I did. If anyone knows anything similar to this for other 'practical' things besides Rubik's cube, I'd love to know about them.

I know what you mean. In my case also, I think it is a combination of me being better informed about the world and the economist declining somewhat over time. BUT I started reading it 25 years ago, and still do read it a bit.

In terms of other news sources - I have no great advice. Here in the uk, I read the Financial Times - it is also rather international - maybe have a look at it. I would say that it is ok rather than great - but it is nice to get your news from more than one place. Private Eye is excellent, but mainly just for a subset of uk news.

I would like to have one "go-to" news source that I could rely on. But that is apparently not possible. So I read a few different things, and try to judge from their different stories what is really going on. Interestingly, I find that hn is a good source of quality information and insight, although it (of course) doesn't cover everything that I want to know about.

Thanks, I will.

I ordered one about 1 year ago, but never received it. Possibly because I didn't have a 'proper' organisation (e.g. a university dept) that I could list in the form.

Stan Ulam's autobiography - Adventures of a Mathematician - talks a bit about this, from his early days in Lwow. It's a good book about interesting times and interesting people - if this wikipedia article interests you, then maybe the book will also.

Could someone explain what is being done here?

I see they are using a GAN, and doing unsupervised training. But then they appear to compare their model to supervised-trained models.

How do they do this? Do they tack a supervised-trained model onto the end of their unsupervised model? I imagine they must do supervised training at some point, else how can they convert sounds to text?

I have a distant relative who works for Springer Nature, and previously worked for Elsevier.

She's a nice person who I rather like. I have tried discussing the ethics of her employer's journals business in a polite way. She told me what I felt were the company's "talking points" - "well, publishing is actually quite expensive, we publish many more papers than we used to, etc" - all of which I think are not valid justifications - e.g. look at the huge profit margins of Elsevier. And yet I think she believes them, or at least "chooses" to believe them.

I find it rather depressing that someone, who is otherwise a fine human being, chooses to work there. I know quite a few other people in the same situation, in various other businesses.

Anyway, I thought it was worth mentioning what some of the people "on the dark side" are like, and that in the mirror they look rather like me and you in most respects. My opinion is that at work they are in an environment of people who all choose to believe the same thing, and so what they do is "normal" and "acceptable".

I follow cricket and (association) football quite a lot, and think I understand a fair bit of what goes on in games. But there are many things that I never notice, and that the commentators (especially in football) do not point out. So I liked this article because it showed me some of those 'hidden' facets of games, that I would not normally hear about.

The article's title is a tad hyperbolic, but it is worth reading nonetheless.

About halfway down , it states the rules:

Every hour, you have to come to the start line and run a 4.16666... mile course (so 100 miles per 24 hours). When you finish, you can rest, eat, sleep etc. But on the hour you have to be back on the startline again. The winner is the one who is still there when everyone else has quit.

Yet another example of the weird and wonderful things that humans can, and do, do.

I just had a look at your comment history. These are (of course) just my own opinions ...

You appear to have mostly commented on 'controversial' topics - you are more likely to be downvoted (and maybe also upvoted) in such discussions compared to more technical topics.

Also, you had multiple typos/errors - this always makes me think that you did not spend so much time and thought on your posts. Why are we spending time reading something that you did not spend time proof-reading?

When I consider posting, I usually ask myself - 'does my comment really add value to the discussion (or, in this case, sub-discussion)?' If not, then probably I ought to just keep reading. Heaven forbid that one writes comments just to get some points!

I just realised that using * twice to represent a power shows up as just a single * - here is the correct version (the numbers are anyway correct):

Prob(0 cases) = (1 - p)^N = 0.9450...

Prob(1 case) ~= p * N * (1 - p)^N = 0.0534...

Prob(2 cases) ~= (p^2) * (N^2 / 2) * (1 - p)^N = 0.0015...

My apologies for that.

If the rate were 1 in 2.3 million, then with 130,000 people vaccinated:

p = 1/2,300,000

N = 130,000

Prob(0 cases) = (1 - p)*N = 0.9450...

Prob(1 case) ~= p * N * (1 - p)*N = 0.0534...

Prob(2 cases) ~= p*2 * (N*2 / 2) * (1 - p)*N = 0.0015...

Prob(3 cases or more) = 1 - P(0) - P(1) - P(2) ~= 2.886e-05

Which is small enough to refute you comment, yes?

I was very struck by this when watching Citizenfour (the documentary on Snowden leaks). The newspaper journalists had little understanding of good procedures (even "don't have a very short password") and thought Snowden hiding his keystrokes from the camera (and anything else) was funny, rather than good practice. And these were the best people Snowden could find. Hopefully, among journalists, there is now more awareness of how important this really is.

As regards Harding/Leigh giving away the password in their book - I find it very hard to believe that this was an honest accident on their part. The defence that they and the Guardian newspaper gave is just not credible ~ 'oh, but we didn't realise'. And beyond that, if you were going to publish the password and somehow thought it was ok, you would at least consult someone who knew more on the topic that they claim that they did.

Well, it is something that I didn't know until a few years ago - ok, I would often disagree with bbc articles, or think them a little biased. But this is something much worse.

And I am still shocked that the bbc does this.

Also, I am interested quite how the process works - who makes the decisions, how many are involved etc, what do the other bbc staff think, etc.