Posts by Brewster's Angle Grinder

Re: Can't neutron stars grow bigger from accretion?

When neutron stars merge, as when anything merges, they tend to shed mass in spectacular fashion. (Astrophysics: the science for people who like big explosions seen from a safe distance.)

Accreting material gets burnt in a jet of X-rays. I'm not sure you could get enough mass from the ash to bulk up the star. But if it got too heavy, it would suffer explosive collapse which might produce a black hole or might produce another neutron star.

The real problem is we don't have a solid value for the upper limit for neutron stars. If Wikipedia is to be believed, this object is slightly lighter than the heaviest neutron star known (2.74M_☉) And that's in pretty good agreement with the lightest stellar black hole (ibid - scroll down). But there's no theoretical minimum mass for a black hole and merging objects could, after shedding some mass, produce one that was, ahem. lighter than the upper limit for neutron stars. So it's probably a black hole

Re: Not surprised

To be honest, you'll have to explain to me why a situation demanded Skype. Was it a video call where they were showing images? Did the recipient only have a laptop? What were the specifics that meant carrier pigeons couldn't be substitute for Skype?

I initially downvoted that but then changed my mind. Even the country which is Covid-19 "free" still has border restrictions affecting the economy.

The entropic decrease of entropy

We're confusing two different things - statistical randomness and predictability. We really want passwords that are unpredictable but we use statistical randomness as a proxy since it's all but impossible to know whether a string of bits is predictable.

This is deep philosophical water. But if our dictionary of predictable strings constantly expands then the likelihood of password being predicted increases with time. So "entropy" (randomness) of passwords decreases over time until we hit a tipping point when the dictionary becomes unmanageably big. At which point we have to remove the least likely passwords - for predictable values of "least likely".

"...harmless like the coronaviruses that cause common colds."

I'm not sure we have a clear picture of that. People who die with or because of "flu" aren't routinely tested to see what viruses are actually present. I've read papers suggesting the endemic coronaviruses are often diagnosed as flu but might be orders of magnitude more dangerous that typical "seasonal" strains of influenza. I've lost the link.

The rest is fine. It's been suggested each of us have 8-12 chronic, "asymptomatic" viral passengers. It's certainly reasonable that some of us might be immune. It could also be that most of the people who are going to spread covid-19 far and wide have already contracted it. The spread from here on in could be much slower with flare ups as it reaches superspreaders who haven't been in contact with other superspreaders and that ignites an isolated chain.

Re: Artificial intelligence

The human brain is massively parallel, though. It also has the advantage of a lot more data - it's looking at the shapes, not a set of bytes.

There's probably a whole bunch of neurons representing portions of letter-shapes which trigger. The neuron which recognises "explainable" sees most of its triggers and says "could be me - not 100%". But no other neuron triggers so the next level goes with that.

And of course, if that subconscious process goes wrong, there's the conscious process which can recognise the mistake and retask the subconscious process to examine it more closely.

Re: optimize / optimise

Self modifying code was a lot of fun. But an absolute bastard to maintain. And lets face it: memory, even cache, isn't in short supply and writing to the code segment is a security nightmare.

Re: Doomed to eternal limbo

"...many protocols have been redesigned to work with nat and often losing features or performance in the process..."

It's telling that people would go to those lengths and put up with those inconvenienced in order to avoid IPv6.

There is a whole field devoted to understanding the sound of stars - astroseismology. There's even a wiki page covering it for neutron stars.

If you're being pedantic, and depending on the field, you might restrict sound waves to longitudinal modes of pressure waves. But they'll be in any medium. And the speed of sound is going to be important to analysing these and/or allow you to deduce important information about the medium and it's internal structure.

In the good ol' days, we'd have written all over the stack and it'd've worked fine - mostly...

Well there's this new-fangled thing called automatic bounds checking. If you try to access an out of range index it throws an exception and your code catches... Oh.

Well, the worst that can happen is it safely takes out the process without corrupting memory. It can then be restarted and... Oh.

Maybe we do need to think about this a bit more carefully. This error checking stuff is quite hard, isn't it?

Re: AMD Dreams

Itanium was a mixture of second system syndrome and a genuine attempt to overcome the limitations of x86. But in the end, it failed even on technical grounds - it just didn't perform. It relies heavily on static analysis which is very inflexible. Run-time analysis enables you to have completely different cores optimising the code for the resources they have without recompilation.

Re: "the politically fraught task of policing home-grown implementations"

"Googapple"

I propose we take the "App" from Apple and the "le" from Google and make the portmanteau for the the combined Apple-Google collective: LeApp.

Re: Chocolate teapot

"It's the functionality they want..."

But the functionality they want is to join Whatsapp or Zoom conversations and to (apparently) access Google Drive natively and all the other functionality apps provide...

For all its faults, the law stops us turning into an elected dictatorship. The law can and does constrain the government. Notable examples include the government's attempt at prorogation and the time I beat the government in court myself.

Re: Worst case scenario

Out of interest, did you ever raise the potential for a global pandemic to shut down the world economy?

Re: How to stop people from having "fun"

"they enter a few of the persons details, and it generates a code, perhaps using the persons details as a seed, to make it unique, so no one else could use it?"

I'm hoping there'll be private keys involved. But the keys and the algorithm used to generate the code will leak - far too many people will need it.

It worked in South Korea - the ususal source:

In South Korea, a non-app-based system was used to perform contact tracing. Instead of using a dedicated app, the system gathered tracking information from a variety of sources including mobile device tracking data and card transaction data, and combined these to generate notices via text messages to potentially-infected individuals.

Re: So what's their position on Oxford commas?

So the AI that's adjusting the spacing will grok the content and determine whether it's the end of a sentence or an abbreviation?

I'm playing devil's advocate here - not least because these dots are rarely used. But the software could treat two spaces after a full stop as an end of sentence marker and adjust the kerning accordingly. Contrast this with an algorithm that expands the spacing following a full stop unless it's a single letter (J. R. R. Tolkien) or on a list of (hopefully) user-definable abbreviations. And the user has now go to realise this is happening, locate the list of abbreviations and add every abbreviation they want to use (and never use the abbreviation at the end of a sentence.) All because...?

Re: Another silly question...

Black holes a couple of orders of magnitude bigger than these are really rare (I'm not sure there are confirmed example) so finding two merging into each other would set scientists abuzz.

But in black hole terms, they're only medium sized "intermediate mass". Supermassive black holes are a lot heftier; for example, the one at the centre of the Milky Way is 4 million M_☉.

Re: Maybe I have missed the point

You've confusing antigen and antibody testing. "Antigen test" is a synonym for the PCR tests used to find parts of the virus RNA (they're testing for the presence of the antigen itself); antibody tests are serological tests for the presence of, well, antibodies.

In the case of SARS and, it appears, SARS CoV2, the memory B cells lapse quickly and it's only the T cells that linger. That's making antibody testing harder. It's also possibly that contributing to instances of "reinfection". (Although, I'm still inclined to believe false negatives - PCR is pretty accurate but getting swabs with a sample can be hard. And T cells were sufficient to fight off SARS 17 years are initial infection.)

As always, neutrinos are left out the mix.

Point 1 is the only point that's fine - at least as I parse it. Photons and other particles experience identical "mechanisms of motion" (geometry) which is why light bends round stars and black holes. But, obviously, without a quantum theory of gravity we can't be certain.

Re: Only in Astronomy

In ballpark terms, a star can range 0.1 - 100M_☉. Wikipedia has a list of some inferred superweights - with it topping out at 315 M_☉ for R136a1.

However, big stars tend to be shed a lot of mass duriing their short lifetimes, and turning into a blackhole typically involves shedding even more. So those aren't good guides to black hole masses. OTOH, a lot of star systems are binaries so a black hole can easily double it mass by slurping up its partner.

Re: Security Services

I'm hoping it's PR and that they played with Zoom so they could say, "See, we're all in it together - we're using Zoom like you plebs" before switching to something more secure.

Re: Interesting project

Yeah, stock interferometry. Their press release goes into a bit more detail.

The signals in question are observed down to 30kHz (10km). Although, according to the press release, they're only planning to look from 100kHz (3km) to 25 MHz (12m).

They're using off the shelf "GPS" which they think will be able to pin down the sats to under a metre (a fifteenth of the wavelength at 25 MHz) and claim the satellites' positions "do not need to be controlled to better than 1 km".

Good luck to them!

Re: They emailed...

"Oh well, just change all of them!"

That rule only applies for people called "Perky Pat".

Re: Pharmacologists on el Reg?

"I just put some baby soap in a pan of boiling water and breathed in the fumes under a towel and fuck me if my breathing isn't a lot clearer."

When you get bunged up again, try it without the soap. I think you'll find the effect is about the same - and probably slightly less than if you add something like Olbas Oil to the water.

Re: Interesting process

I reckon we put in a new moon. We could "borrow" one of the dwarf planets. If we got it in close enough it would stir up Mars' interior and, given a few millions years, get the magnetic field humming again.

And, much like the Earth/Moon system, our new Mars/moon system would orbit its barycentre. So, if we plum for the obvious choice, we'd have Ceres in parallel with Mars.

_{Sorry. That was a helluva lot of set up for a really lame joke.}

Re: Here's the detail

If Apple had any sense they'd use some of their money to lobby for patent reform. But one suspects that, under the current system, they gain more than they lose.

Re: 2012 is the corpus of words

>Try "chocolate" and you'll similarly see the decay down to 2012.

Or try "google". (I did worry it might break the internet - but it seems okay.) Apparently Google peaked in 2005 and dropped off so that by 2012 we weren't talking about then at all. I wonder who they got bought out by?