neilkk

Didn't our whole software industry come to a consensus to stop doing things like this because the results were fundamentally inaccurate and the invariable outcomes were stakeholders feeling hard done by and/or ICs being victimised by middle managers?

Same. I read it cover to cover multiple times. The only part I didn't understand was the bit about Quicksort. I was a C programmer by the time I finally grokked it.

It's not a generalisation, it's just the definition of the word 'competent'.

You're claiming that police chiefs deliberately overreact to these calls in order to put pressure on lawmakers via the FBI and the media?

All communication is communication with people in the future.

But Github and similar tools actually solved this problem where Git failed to do so. Nowadays people have a setup with Github or bitbucket where they can navigate from a piece of code right to the pull request, where they can read the code review discussion, see the build log, reach linked resources like the Jira, etc.

Right. You've given an example of exactly what I said was the only reasonable use case for detailed info in the commit message: someone in the future will need to know the history of that particular piece of code. It seems like the point of your specific example is to say 'in some cases you might want to know the history of a gap'. Fine. That seems like a nitpick to me.

No one in the future will need to know the history of a particular ascii encoded blank space (among a whole file of ASCII encoded blank spaces). Anyone who needs the general info that the file needs to be ascii will be helped by it being somewhere else, as opposed to in a random commit message.

The thing is that writing a good commit message for future people doing `git blame` is only worth it if it's a line of code which someone in the future will look at and need to know why it was changed from its previous form to the current form.

If you simply want to comment the current state of the code, you should add a comment in the code.

No one will ever need to know in the future why that particular space character is an ascii space, so the whole commit message is just a blog entry in the wrong place.

It would have made sense to just put a comment at the top of the file saying "make sure encoding is whatever".

There are 100 cents in a dollar fyi.

The youngest person who was able to drive when gas was a quarter dollar a gallon is now in their early nineties.

In other words they have attempted to rebrand themselves after justified criticism that they were promoting the mindset that sees climate change as fictional and academic enquiry as a waste of time.

> In October 2005, John Mainstone and the late Thomas Parnell were awarded the Ig Nobel Prize in physics, a parody of the Nobel Prize, for the pitch drop experiment.

Embarrassing that the Ig Nobel prize has so little understanding of what valid science is. The awarders seem to be motivated by chasing the approval of very ignorant and biased people in the general public and the media.

The implied probability of Brexit a few days before was around 25%. The implied probability of Trump beating Clinton a few days before was around 20%.

Neither the time value of money nor the cost of trading explain these dislocations.

No, that's not right.

Do those first four steps. You wouldn't (necessarily) cover every black point in your rectangle. Choose a remaining black point and flood fill from that, say green. Keep on doing this with different colours until you've covered every black point in your rectangle. You have a bunch of regions of different colours.

Now, if the different coloured regions are all nicely separate, then your set is locally connected. Because each point is either cleanly in one component or cleanly in the other.

If on the other hand your drawing looks like https://commons.m.wikimedia.org/wiki/File:Julia_set_for_the_... with mixed up boundaries where some points are infinitesimally close to more than one colour, then it's not locally connected.

The difficulty with intuition is that in our intuition, coloured regions always have reasonable boundaries (think countries in a map: the border can be wiggly but there's never infinitely many tiny bits of one country mixed up in the boundary of two others). In fractal geometry, things like the Newton fractal picture above are quite usual.

Suppose you choose such a rectangle, let's say it incorporates the 'fringe' of the main cardioid and the fringe of the biggest circle. Within that rectangle, color green all the pieces which connect to the main cardioid and blue all the pieces which connect to the circle. Local connectedness means that there won't be any points in that rectangle which have both blue and green points arbitrarily close. So there are places where 'locally non-connected parts' of the set can be close together, but there must be a border between them, rather than them being hopelessly entangled.

Strictly speaking, you would do this coloring with all connected components of the intersection of your rectangle and M. (And the rectangle could be any region.)

The example messes this up, although it is a correct example, the square on the diagram showing the local piece containing non-connected parts is wrong. The comb has more and more teeth, infinitely many in a bounded space, on the left side. Only a rectangle which includes the left edge properly shows why the set isn't locally connected. The rectangle pictured includes finitely many teeth which have a separation between them. A rectangle overlapping the left edge of the comb would include separate components which get arbitrarily close to that left edge and so can't be separated by a border.

Right, so it was just as much in their interest as it was in Clifford Stoll's to pretend that this was an international espionage incident?

Past a certain point optimism simply becomes wilful naivety.

Don't have exact figures but it's a lot less than zero.

Surely this (unlike many other environmental and social issues) is effectively solved by price?

If helium balloons cost $100 each vastly fewer of them would be sold. If they cost $100 but you could get $90 back by bringing them back to the store for the helium to be recycled, much less helium would be consumed.

If there is a limited supply of helium for essential medical use, the price should rise until either demand is affected or less of it gets discarded by hydrocarbon extractors. If this doesn't happen we are probably ok.

"Will you ban Finsta?" was not an example of such a question. It was a profoundly stupid question from someone who had the resources to educate themselves before asking it.

The package in transit belongs to Amazon not to the addressee.

Why do you think that the survey wouldn't include drug dealers in the respondents?

Crime survey data often removes many of the noise in recorded crimes data. Basically surveying a cross section of people and asking them if they've been a victim of crime in the last X months, and what type of crime.

The Great Wall of China is military technology, would you consider that its effectiveness is proven?

Right so by analogy, it's possible that medieval castles were built a) for a reason that we don't understand or b) for a reason we do understand, but where the justification for why castles fulfil that requirement is dubious.

Egyptians built pyramids in order that their kings would live forever. The pyramids were not effective at allowing their kings to live forever so why did they build so many?

Stone Age people painted the animals they were going to hunt on rocks so that the pictures would entrap the animals' spirit and they would be easier to catch. This method of ensuring a good hunt was not effective so why did they do so many paintings?

Your reference definitely doesn't show a calculation of the type you are trying to do. Likelihood ratios are not the same as odds ratios; they do not have the problem I described.

I would be very surprised if you can find any reference at all to the number you describe as 'evidence bits', or anything equivalent, made by anyone who can show an understanding of basic probability, statistics, or information theory.

I understand how you get 20,000,000 as the answer to the calculation you carry out. My point is that that number is not meaningful in any way.

I'm having a hard time understanding why this is better than CLOCK. It seems like a hack to make updates easier which however makes both practical performance and conceptual justification much worse.

The elements which are between 'hand' and 'tail' are going to stay there for a long time, until the hand pointer cycles. If there are lots of elements which get put in the cache but don't really deserve it, that could take a long time. But that's exactly the case where you wouldn't want to keep the elements behind the hand around for a long time. Remember, they got 'promoted' for appearing twice in around the cache size records. Not a great indication that they should be kept for a long time. On the other hand, when the pointer cycles, they could get thrown away quite easily.

On the other hand CLOCK converges to a situation where elements which have appeared a lot are at the front. They are much less likely to be thrown out even if they have a temporary run of scarcity. The elements which don't appear much quickly bubble to the back and become much more likely to be thrown out in the next few evictions. This is accentuated if there are lots of elements which are getting hit relatively more frequently.

After a long time of running SIEVE, there's not really anything about the state which will improve future performance compared to when it starts cold. In many cases it seems it will be locked in to a suboptimal state by overadapting to past observations.

Yes, you are doing math with odds as though they are fractions or ratios, which is deeply incorrect. 20:1 is not the reciprocal of 1:20 but the complement. Odds ratios and similar calculations do not work like this. You can do this type of calculation using X:1 odds or 1:X odds but not both in the same calculation.

Or perhaps you can provide a reference to a justification of this type of calculation?

This isn't how the mathematics of odds works, as the GP correctly pointed out. An event which is 20:1 on is not 20 times more likely than certainty.

Going from 1:4 to 1:2 means that the event has become twice as likely. But going from 2:1 to 4:1 does not: it means that the complementary event has become half as likely.

Based on this, we can't do math with odds treating them identically to ratios.

If you do the math correctly, the two types of information measure are basically the same thing.