* Posts by Paul Kinsler

602 posts • joined 9 Aug 2007

Page:

Journal that published Facebook emoto-furtle study: Proper boffins get CONSENT

Paul Kinsler

Re: Required:

Possibly also required, reading the study itself...

http://www.pnas.org/content/111/24/8788.full

E.g. "On Facebook, people frequently express emotions, which are later seen by their friends via Facebook's "News Feed" product (8). Because people's friends frequently produce much more content than one person can view, the News Feed filters posts, stories, and activities undertaken by friends. News Feed is the primary manner by which people see content that friends share. Which content is shown or omitted in the News Feed is determined via a ranking algorithm that Facebook continually develops and tests in the interest of showing viewers the content they will find most relevant and engaging. One such test is reported in this study: A test of whether posts with emotional content are more engaging."

The following paragraph is also informative as to what process was followed.

Speed of light slower than we thought? Probably not

Paul Kinsler

Re: does heat bend light?

heat is energy, and energy bends spacetime, and light beams follow curved paths in curved spacetime. So yes, heat does bend light.

Paul Kinsler

Re: I guess Franson must have assumed ...

If it's on the arXiv, you dont have to guess. You can go check it yourself. Or indeed you can check the open access article itself, using the links in the Reg article.

Titan sprouts 'Magic Island', say astroboffins

Paul Kinsler
Joke

Re: how can an island just appear?

How? By magic, apparently. :-)

And before all you nay-sayers start off, it is well known that any sufficiently primitive magic is indistinguishable from technology (or, possibly, just indistinguishable from some not especially obvious physical process or other).

Indian climate boffins: Himalayan glaciers are OK, thanks

Paul Kinsler

Here's at cut&paste of the conclusion of the article...

... because how something is said is also useful:

The results of the present study indicate that most of the glaciers were in a steady state compared to the results of other studies carried out for the period prior to 2001. This period of monitoring almost corresponds to hiatus in global warming in the last decade[32] . It may happen that an interval of one decade could be smaller than the response time of glaciers to be reflected in terms of any significant change with 23.5 m spatial resolution of data. This point requires further studies using high-resolution data for a longer interval of time.

~

So, should we trust these results now, even though the authors themselves worry about the data resolution on which they are based? Or should we wait for a more solid confirmation before using them to inform our actions?

We're from the same dust cloud, bro: Boffins find Sun's long-lost sibling

Paul Kinsler

"can be viewed even with low-power binoculars "

does this, for most of us, really mean "might be seen with binoculars, if only it wasn't for all the light pollution"?

(can be found in kstars, for those with KDE stuff on their system)

anyone planning on a big network rewiring job?

Paul Kinsler

anyone planning on a big network rewiring job?

... maybe you should read this:)

http://arxiv.org/abs/1405.2185

Easily repairable networks

Robert S. Farr, John L. Harer, Thomas M. A. Fink

We introduce a simple class of distribution networks which withstand damage by being repairable instead of redundant. We prove a lower bound for the expected cost of repair, and show that for networks on the square and triangular lattice, this bound is achievable and results in a network with exactly three levels of structural hierarchy. We extend our results to networks subject to repeated attacks, in which the repairs themselves must be repairable. We find that, in exchange for a modest increase in repair cost, such networks are able to withstand any number of attacks.

Truck-sized asteroid slips silently between Moon and Earth

Paul Kinsler

Re: LOHAN could be coming from 70.000 ft through a window very close to you ...

and will she be bringing some Milk Tray?

Boffins tag Android app privacy fails

Paul Kinsler

Re: I used to have a small PDA device ...

I've got an old Zaurus I would sell you ... except I'm still using it. Otherwise couldn't agree more ... if only I could get Opie on my phone instead of android :-)

London Tube has new stop at Azure Station

Paul Kinsler

Re: So what are the 'things'

Since this is the London Underground we're talking about, I think by IoT they mean either "Internet of Trains", or "Internet of Tubes"

Microsoft: The MORE Surfaces it sells, the MORE money it loses

Paul Kinsler

Re: "If he were to sum up the quarter in two words it would be “execution and transition.”"

hah! - but to sum up in words of just one _sound_ is the best thing to do

DeSENSORtised: Why the 'Internet of Things' will FAIL without IPv6

Paul Kinsler
Happy

the lightbulb moment...

pawlik: "all the very disappointed customers who say 'oh, I bought my lightbulb but now I can't switch it on and off anymore because I can't reach it on the internet.'"

Well, back to candles it is, then.

Nothing's as SCARY as an overly aggressive SOFTWARE PIMP

Paul Kinsler
Happy

Re: planned obsolescence [...] Planet Linux does it too

Oh no! Are they going to retire the fvwm window manager now? Dammit, I'll have to switch back to twm!

CERN team uses GPUs to discover if antimatter falls up, not down

Paul Kinsler

Re: Aha!

you might want to be careful: one newspaper claimed today that scientists had discovered a new (invisible) eighth colour of the rainbow. But they really had - over two centuries ago: see W. Herschel, "Experiments on the Refrangibility of the Invisible Rays of the Sun", circa 1800. http://www.jstor.org/stable/107057

Paul Kinsler

Re: mining Bitcoins...

... and so some are already working on ow to maximise return, see e.g.:

Do Bitcoins make the world go round? On the dynamics of competing crypto-currencies

http://arxiv.org/abs/1403.6378

Stefan Bornholdt, Kim Sneppen

Bitcoins have emerged as a possible competitor to usual currencies, but other crypto-currencies have likewise appeared as competitors to the Bitcoin currency. The expanding market of crypto-currencies now involves capital equivalent to US Dollars, providing academia with an unusual opportunity to study the emergence of value. Here we show that the Bitcoin currency in itself is not special, but may rather be understood as the contemporary dominating crypto-currency that may well be replaced by other currencies. We suggest that perception of value in a social system is generated by a voter-like dynamics, where fashions form and disperse even in the case where information is only exchanged on a pairwise basis between agents.

New IPCC report: 8 ways climate change will throw world into peril

Paul Kinsler

Re: thermometer reading at a rural weather station

... but how does that reading map onto the average temperature of the area in which it inhabits? Is it intended to represent a ground temperature or an air temperature? What if I want to use it as a measure of air temperature in a surrounding box of area A and height h? Do the thermometers have a changing response as they age? Can I trust what the manufacturer said? What about when they are replaced for new ones with different properties? How do we correct for the changing nature of the landscape around it? Thinks like this need to be checked and ironed out.

Paul Kinsler

Re: much data manipulation

As it happens, here's a recent discussion of solar irradiance validation

http://arxiv.org/abs/1403.7194

(but published formally in Astrophysics and Space Science 350(2), 421-442, 2014)

Would you like the raw satellite data as well? I'm sure there are all sorts of ways the existing analyses could be challenged (apart from those covered in the paper already, of course).

Paul Kinsler

Re: much data manipulation

It's important, however, to distinguish between types of "manipulation"

(a) that done with intent to deceive

(b) that done by mistake or carelessness

(c) that done as part of a calibration or compensation process to ensure the numbers output by the sensor correspond (to the greatest extent achievable) to what the sensor data is understood to mean (or is intended for use as).

This raw data you are so keen on may or may not be what you think it is, or may or may not be misleading (wrt your purpose) in subtle ways. For example, a few years ago, I happened to have a conversation with a physicist whose specialism was (IIRC) the calibration of satellite data related and needed for an estimation of the Earth's energy budget. Such things are not trivial.

Paul Kinsler

Re: how to communicate

Those interested in the dynamics of social discourse may find the following article amusing:

Mental ability and common sense in an artificial society

Krzysztof Malarz, Krzysztof Kułakowski

http://arxiv.org/abs/1403.5970

We read newspapers and watch TV every day. There are many issues and many controversies. Since media are free, we can hear arguments from every possible side. How do we decide what is wrong or right? The first condition to accept a message is to understand it; messages that are too sophisticated are ignored. So it seems reasonable to assume that our understanding depends on our ability and our current knowledge. Here we show that the consequences of this statement are surprising and funny.

Selfies are over: Welcome to the age of 'Sleeveface'

Paul Kinsler

Re: Years ago

Personally, I could quite imagine citizens of ancient Rome standing round so that they matched up with nearby statues or other representational art in some amusing way, let alone the reasonable likelihood boxes of old Edwardian b&w photos doing a similar thing with paintings or whatever; or someone or other doing this - with actual album sleeves - in the 70's.

No doubt someone might have some sort of priority claim over this kind of amusement, but I not only strongly suspect they have been dead a while, but they might even have been dead for millenia.

First pics: Comet-chaser Rosetta hurtles towards icy prey, camera in hand

Paul Kinsler

Re: bring the rock into orbit around Terra.

... isn't NASA planning such a mission? Although the aim is to put it in orbit around the moon instead....

Star Wars movie to start shooting in UK this summer

Paul Kinsler

Re: I can't see ... what much coherent story can be told after I to VI ...

hmm .. something that goes off at a tangent, and does something new with the SW storyverse, whilst also cleverly including hints to set the scene for the next sequel trilogy after this one? :-)

What did you see, Elder Galaxies? What made you age so quickly?

Paul Kinsler

Re: "the actual paper is perfectly correct, it's the press release that is wrong."

... so a lesson for scientists - check that press release drafted by the institutional PR department, and make sure it's right (or at least not-wrong ... they're doing their best but they aren't experts). In my case, we rewrote it substantially. That said, you have very little control over what gets bootstrapped by journalists off the back of the release - or worse, of a press-agency remangled precis of the press release.

X marks the... They SAID there was a mystery planet there – NASA

Paul Kinsler

Re: 1.49597871 × 10^15m

Hmm ... Register Units?

About 8 mReg?

(Being 8/1000 of the distance from here to Regulus. Although mysteriously, the Register-star is in the constellation of Leo (probably eating its entrails...), and will be obscured by 163 Erigone on the morning of March 20th - expect articles to go missing or be delayed around that time.)

Paul Kinsler

1.49597871 × 10^15m

... is about 1/6 of a light year, if it helps any.

Global Warming is real, argues sceptic mathematician - it just isn't Thermageddon

Paul Kinsler

Presumably this is the peer-reviewed version?

An Objective Bayesian Improved Approach for Applying Optimal Fingerprint Techniques to Estimate Climate Sensitivity

Nicholas Lewis

JOURNAL OF CLIMATE 26, n19, 7414-7429 (2013)

DOI: 10.1175/JCLI-D-12-00473.1

http://journals.ametsoc.org/doi/abs/10.1175/JCLI-D-12-00473.1

A detailed reanalysis is presented of a Bayesian climate parameter study (as exemplified by Forest et al.) that estimates climate sensitivity (ECS) jointly with effective ocean diffusivity and aerosol forcing, using optimal fingerprints to compare multidecadal observations with simulations by the Massachusetts Institute of Technology 2D climate model at varying settings of the three climate parameters. Use of improved methodology primarily accounts for the 90% confidence bounds for ECS reducing from 2.1-8.9 K to 2.0-3.6 K. The revised methodology uses Bayes's theorem to derive a probability density function (PDF) for the whitened (made independent using an optimal fingerprint transformation) observations, for which a uniform prior is known to be noninformative. A dimensionally reducing change of variables onto the parameter surface is then made, deriving an objective joint PDF for the climate parameters. The PDF conversion factor from the whitened variables space to the parameter surface represents a noninformative joint parameter prior, which is far from uniform. The noninformative prior prevents more probability than data uncertainty distributions warrant being assigned to regions where data respond little to parameter changes, producing better-constrained PDFs. Incorporating 6 years of unused model simulation data and revising the experimental design to improve diagnostic power reduces the best-fit climate sensitivity. Employing the improved methodology, preferred 90% bounds of 1.2-2.2 K for ECS are then derived (mode and median 1.6 K). The mode is identical to those from Aldrin et al. and [using the same Met Office Hadley Centre Climate Research Unit temperature, version 4 (HadCRUT4), observational dataset] from Ring et al. Incorporating nonaerosol forcing and observational surface temperature uncertainties, unlike in the original study, widens the 90% range to 1.0-3.0 K.

Dark matter killed the dinosaurs, boffins suggest

Paul Kinsler

Re: What an amazing ...

Indeed, it can be rather interesting to treat ancient (or even Medieval) science seriously ... e.g. this (nothing to do with dark matter or pyramids, mind)

http://arxiv.org/abs/1403.0769

A Medieval Multiverse: Mathematical Modelling of the 13th Century Universe of Robert Grosseteste

Bower et al.

Abstract: In his treatise on light, written in about 1225, Robert Grosseteste describes a cosmological model in which the Universe is created in a big-bang like explosion and subsequent condensation. He postulates that the fundamental coupling of light and matter gives rises to the material body of the entire cosmos. Expansion is arrested when matter reaches a minimum density and subsequent emission of light from the outer region leads to compression and rarefaction of the inner bodily mass so as to create nine celestial spheres, with an imperfect residual core. In this paper we reformulate the Latin description in terms of a modern mathematical model. The equations which describe the coupling of light and matter are solved numerically, subject to initial conditions and critical criteria consistent with the text. Formation of a universe with a non-infinite number of perfected spheres is extremely sensitive to the initial conditions, the intensity of the light and the transparency of these spheres. In this "medieval multiverse", only a small range of opacity and initial density profiles lead to a stable universe with nine perfected spheres. As in current cosmological thinking, the existence of Grosseteste's universe relies on a very special combination of fundamental parameters.

Massive new AIRSHIP to enter commercial service at British dirigible base

Paul Kinsler
Headmaster

Re: Vectored thrust...

Hmm ... surely all thrust is vectored? I mean, how could one make sense of something caller "scalar thrust" (or "bi-vector thrust", for that matter)

/takes of geometric algebra hat and returns to coffee

UK libraries trial free access to scientific research

Paul Kinsler

Re: Lets start ...

I agree. But before you cackle with glee at the thought of how you might cleverly dismantle the data or assumptions behind some research finding or other, you might want to consider that you'll actually have to engage with the technical content, rather than the simanglified version of it reported by your favourite news provider. You will almost certainly find the experience a chastening one - even as a scientist, peer reviewing papers in your own field can be extremely hard work - so how might it be for the non-specialist?

Paul Kinsler

Re: "The research is ours "

Tom 7, that's not necessarily true - even on a directly taxpayer funded grant, you just paid for /some/ of it. I do not want to deny the very significant claim the taxpayer/donor might have as a result of their funding, but if you really wanted to claim the lot - lock, stock, and barrel - then you really should have been paying more than you did.

E.g. EPSRC grants are now calculated on a "full economic costing" basis, which is indeed an attempt to charge for everything associated with a grant (even overheads like fractional office space, etc), whereas before that was not the case. As a result, the taxpayer now pays more, and, I suppose, also gets more claim. I do hope you are enjoying that sense of ownership over your personal fraction of the 1/4 of a room in a 50's office building that I inhabit while at work :-) Some charities (as I understand it) refuse to pay (some types of) overheads, and so, I suppose, have less claim.

In this specific context, note that taxpayer funded research has not always included money for publication costs. If the taxpayer didn't pay publication costs, then they didn't pay to get free access to read the results. Maybe things should have been different, and, as I said above, things are changing now, ... although the result will not be some kind of utopia. Notably, taxpayers/donors will pay for the extra access for everyone that you & I want, but the vast majority of those granted access to the literature will either not bother or find it beyond them.

Paul Kinsler

Every tax payer should have access to *any and all* research ...

Well, I agree. But the funding doesn't work like that yet, although here in the UK we are making the transition. Personally, I try to avoid publishers with illiberal redistribution policies, and put a version of all my output on arXiv irrespective of its appearance in a journal.

There are costs in the new model though - notably authors (mostly taxpayer or charity funded) will (now) have to pay publishing costs themselves, typically in the range £1000-£2000 per article, so the tax payer/charity donor will end up paying this. Further, anyone without funding - say an independent researcher, one whose funding/employment has just terminated, a new PhD trying to write up work independently, someone trying to publish work their institution doesn't care about (or perhaps even doesn't like) - now also has to (will have to) pay these costs, which are not negligible. The "author-pays" model on which your proposal rests will narrow the range of people with the ability to publish their science.

Further, when this new model is in place in the UK, you will perhaps notice that UK taxpayers and UK charity donors are making UK research free-to-view for a global audience, but the arrangement might well not be reciprocated. Some may be unhappy at such a situation.

Lastly, I find that I am inclined to feel that if a journal charges authors, they should also pay the reviewers (such as myself), who currently do it for free. Should peer review turn into a paid-for process?

So: the simple proposal you advance (and with which I generally agree) is not necessary so simple execute in a fair manner, there may well be unintended consequences, and it isn't "free", even to taxpayers & donors.

Japanese quantum boffins 'may have the key to TELEPORTATION'

Paul Kinsler

Re: It's one of the many reasons @ Mike Bell

Yes, but don't think just by calling yourself "Mike" is going to let you off - we all know you personally are responsible for the inequalities, John.

Paul Kinsler

Re: How the hell do you squeeze a vacuum?

Strictly speaking, a squeezed vacuum isn't a vacuum - it has photons[1] in it. It's called a vacuum because it has a zero amplitude, but in fact its quantum uncertainty has carefully arranged correlations, for which some excitation above the true vacuum are required.

Or: a squeezed vacuum not a vacuum, we just call it that because the field amplitudes measure (in average) as zero. But a photon number measurement will be non-zero.

[1] Assuming we're talking about an electromagnetic (squeezed vacuum) state.

New Doctor Who's new costume newly REVEALED by Beeb

Paul Kinsler

Re: for true innovation / dress to blen in

hmm ... so maybe some episodes in pre-revolutionary France?

Dusty old supernova could reveal answer to life, the universe and EVERYTHING

Paul Kinsler

good old 1987a

... my favourite ever supernova. Because I volunteered to be sent up a hill in the south island to help (very slightly, and in a not very critical role, as an MSc student) with the building of the JANZOS cosmic ray observatory. I think I might have moved some thingies from here to there, or possibly put one sort of thingy on top of another sort of thingy, or joined thingies together, in somewhat chilly weather.

How Britain could have invented the iPhone: And how the Quangocracy cocked it up

Paul Kinsler
Happy

Re: comic books, novelty ties and ironic t-shirts.

Oh, I don't know. You'd be surprised at how artistic theoretical calculations, computer modelling, and (e.g.) high-vacuum cold atom sculpture is. Especially once it had been properly funded.

When the lights went out: My 'leccy-induced, bog floor crawling HORROR

Paul Kinsler

Re: Oh yes, power cuts in the '70s

These days I might just hope that at least if the power was to cut out, that it was on a cloudless night, so that for a change I might see the stars properly.

What's wrong with Britain's computer scientists?

Paul Kinsler

Re: CS graduates are choosing unemployment over becoming teachers?

Remember, they'd have to retrain as teachers - so its not a trivial job switch, and they may well reckon they'll get a more standard CS job before finishing (and paying for) teacher training.

Our TINY flying robot moves like a JELLYFISH, say NYU boffins

Paul Kinsler

Re: a bit fast to be called flapping.

What do hummingbirds do with /their/ wings then?

I think the "flapping" debate might just as well be applied to slower flying jellyfish; it's more about the action than the speed.

Hello Warsaw: Greenland ice loss will be OK 'even under extreme scenarios'

Paul Kinsler

on the subject of an icy north (or not)

... this is an interesting semi-technical read:

http://www.see.ed.ac.uk/~shs/Climate change/Arctic%20ice/Jeffries%20Overland%20Arctic%20new%20normal%23.pdf

WHO was it that TAMED the WOLF? Heel, Rex! No! Aarrghh!

Paul Kinsler

Re: It may be a dumb question but ...

I've just skimmed the paper. Although its not my field at all, they have thought about alternatives.

E.g. this quote: "We consider this scenario unlikely as it would require a common recent coalescence of these ancestral wolf and dog sequences from geographically disparate areas. Nevertheless, ..."

But there are plenty of caveats scattered through the paper -- as with lots of this sort of science, their eurodog hypothesis is the most reasonable conclusion >>given the available evidence<<.

RED planet, INDIAN mission: Space probe prepares for voyage to Mars orbit

Paul Kinsler

Re: cash to waste

Other reporting[1] has pointed out that India's social aid budget vastly dwarfs the money spent on this Mars shot, and that India's poor would benefit more from more effective (and less wasteful) use of their existing budget rather than adding by some tiny fraction to it by cutting sci/tech funding.

[1] Either or both the Indy or Gruniad articles I read over the weekend, for example.

Ex-Beeb online news guru joins Brit speech-analysis startups

Paul Kinsler

Re: the worst way of disseminating information.

maybe I didn't look around the site enough, but why is it you think that's how they're proposing the courses are run?

Instead, I would assume some set reading and exercises to set up the lectures; as in e.g. http://arxiv.org/abs/1309.0852

Deploying Turing to see if we have free will

Paul Kinsler

Re: nothing contradictory about both those propositions being true at once

As I understand it (Seth Lloyd's paper), to determine whether or not you have free will or not you need a third proposition:

3. It is physically possible that another (not-you) entity or system could have a model of you that reliably predicts your actions /before/ you so act [1].

If this (3) is true, then you fail Seth Lloyd's "free-will" test; if false, you pass, and might have free will after all.

[1] Note: A prediction generated using prior information, but without looking at the outcome, but which is calculated after you act, does not satisfy (3)

Paul Kinsler

Re: "Free will" is rather poorly defined.

Physical models explicitly disallow free will. The most free-will-like thing they can do is add in some statistical uncertainty to either the initial conditions or the dynamics; and then consider this uncertainty to be a proxy for whatever free-will might be. Note that chaos cannot help here, since it is still strictly deterministic, although it can be a handy way of making things more complicated when constructing a pseudo free-will fr your models.

How do you build a mathematical (or physical) predictive model of something with free-will? The notion of a predictive model of free-will is an oxymoron. If I've got a good model, the decisions no longer look free, they are just consequences of the state of the system. Hence, there are (can be) no predictive models of free-will.

Hence the article's focus on "practical unpredictability" -- i.e. "we cannot in-principle predict in advance" -- as a way of assessing the likelihood of free-will being present. It's about as close as we can (currently) get to understanding free-will.

Paul Kinsler

Re: confused, moi?

I once read some Bertrand Russell, and had to read each sentence twice in order to understand it and its in-context meaning. Some time you just have to keep banging the (mental) rocks together and hope it eventually becomes clear.

But if you'd prefer a more IT angle, why not try this instead?

"When does a physical system compute?"

http://arxiv.org/abs/1309.7979

Alarming tales: What goes on INSIDE Reg hack's hi-tech bedroom

Paul Kinsler

Re: try your body clock

This sort of thing mostly works for me, although I wake seconds earlier, not minutes. If not, then the light from my trusty old Zaurus's screen (as it wakes) wakes me, and if not that, then it's really rather pleasant alarm chirp.

The most exciting radio alarm event I've had was when Mark F decided to play Ministry's "Jesus built my Hotrod" on 4ZZZ at 6:55 one morning, just in time for my usual wakeup. Thanks, Mark!

Microsoft wants to 'move beyond' the Cookie Monster

Paul Kinsler

Re: Who uses bing?

Now that a large proportion (maybe even most) searching is done online, we no longer need a special word to denote online searching. Thus we can use that one we all used to use before we got all excited about computery stuff. You need only say -

"Just search for it"

Boffins have constructed a new LIGHT SABRE. Their skills are complete

Paul Kinsler

Re: "wholly out of light"

The only way you get light to not travel at lightspeed is to make it interact with something. E.g. in water, visible light interacts with the water molecules, and it slows by about 30% (hence the refractive index of 1.5). To make light really slow, you have to get it to interact very strongly - in this case, using a specific frequency of light tuned to specific transitions in carefully managed rubidium atoms.

This "slow light" therefore, is a rather misleading name - it isn't just light - it's a strongly coupled light-matter system, which some would prefer to call a polariton.

I can therefore only assume the down votes were for the poor quality joke.

Page:

Biting the hand that feeds IT © 1998–2019