HNHacker News
TopNewBestAskShowJobs

tfgg

1,287 karma · joined May 25, 2013

timothy.green+hellohn@gmail.com
submissionscomments
tfgg··on Avoiding a pointless blockchain project
Most 'blockchain' proposals I've seen, particularly in civic technology, would be satisfied with a public git repository containing a .csv file, and also generally ignore the social issues around data ownership and quality.
tfgg··on Commuting Takes Its Toll
Locating the office in zone 1 also allows people to live in lots of outer London areas. If you site the office in zones 2/3, you have to choose which bit of zones 2/3, potentially meaning people have to move house or commute even further. It's a higher cost to the company, but gives them access to a larger proportion of the labour market.
tfgg··on Poor Sleep Gives You the Munchies, Study Says
It's the informal generic "you". More formally it'd be "Poor Sleep Gives One the Munchies, Study Says".
tfgg··on The Oxford maths interview, shaking up admissions
The mouse is walking along the band, while the end of the band moves away, however the mouse is also being carried along by the expansion of the band. I think the best approach is to think in terms of the fractional coordinate of the mouse along the band, and work out its effective velocity in this coordinate.
tfgg··on The Oxford maths interview, shaking up admissions
They recently (a few years ago) introduced A* grades for 90+/100 scores, where A was 80/100 before and so, as you note, not very discriminating. I'm not totally sure, but I think a standard offer for one of the elite universities is A*AA these days.

I'm not sure I like the idea. Only having to hit 80 reduces stress quite a bit, and it gave me time to do extracurricular things like programming, and way more maths modules than anyone should :) If I had to hit 90+ in everything, I would have probably been risk adverse and cut down. It also feels like it selects a bit too much for people who over perform in exams. That said, I did have to send in my raw module marks anyway when I applied ~10 years ago, and did hit 90+ anyway.

tfgg··on What My PhD Was Like
I agree with your analysis, I did notice that not many groups in the US were doing what looked like interesting new methods/theory, more 'cashing in' on existing code in a fairly short term way. There are people like Louie and Vanderbilt, though. I did my DFT methods PhD in a materials department group, full of Italian/German/English physicists, anyway :)
tfgg··on What My PhD Was Like
Eh? This isn't my experience of condensed matter DFT at all. It's a pretty hot area in the UK/Europe at least, and there are lots of postdocs being advertised, as well as a reasonable number of new faculty being hired in the area. Maybe if you're just hitting 'go' on someone else's code, sure, but developing high throughput analysis or higher levels of theory... I mean, until recently, the head of the Cambridge CMP theory group was a DFT guy.
tfgg··on The Superfluid Universe
This is a pretty amazing and compelling idea, it feels odd that no one has thought of it before - the article mentions some previous models, what were they?
tfgg··on Exotic four-neutron no-proton particle confirmed
Particles are sort of mathematically more poorly defined the shorter their lifetime. But yes, a nanosecond is a long time in this context :)
tfgg··on A bot has successfully appealed $3M worth of parking tickets in the UK
I think it's more that the letter only holds sway in the first place by the implied threat behind it. If the letter is from a bot, there is nothing to back up the threat.
tfgg··on Google Launches Fresh-Grocery Deliveries
> A supermarket takes up a huge area, and usually in areas where land tends to be expensive

I think this is a particularly good point about their success, especially in London. In London, there aren't really many huge supermarkets (the closest one to me is about 5km away through dense built-up areas), and most people don't own cars. Hence, if you want to do a big shop, online delivery is pretty much the only way.

tfgg··on Mern: Build JavaScript apps using React and Redux
The page keeps mentioning 'isomorphic apps' but I have no idea what those are.

From googling, they appear to be JS code that can run client and server-side? Not what I would have guessed!

tfgg··on New finding may explain heat loss in fusion reactors
I think the answer is "with great difficulty". When you're developing something like this, you might have an issue with:

- Experiment

- Model

- Approximation to the model

- Numerics

- Code

(- Compiler & hardware)

So, your answer might not match experiment. Maybe the experiment is wrong. Maybe it's an expected effect due to an approximation you used. Maybe your integration grid isn't sufficiently converged. You just have to systematically eliminate these, often by comparing to more expensive but verified simulations, or analytic solutions on simplified models.

tfgg··on What should I learn next as a programmer?
I'd say stop worrying about the technology you use and start worrying about the question you're trying to answer and problems you're trying to solve.

> Programmers should really take a leaf from the Mathematics & Sciences. In those subjects they are endlessly looking for new solutions and then crediting or discrediting each other, in a logical manner with the use of facts, calculations, proofs and actual data. Not gut feelings, random beliefs and twitter debates!

This problem is worse in the sciences, there are even more 'schools of thought' and people stuck in certain modes of thinking. "Science advances one funeral at a time" is a well-known quote for a reason.

That all said, yes, go explore new technology, try it out, do what you enjoy, I find it's only ever been beneficial :)

tfgg··on I no longer understand my PhD dissertation
During my PhD I found I had to occasionally lock myself in a conference room with a bunch of papers and all the whiteboards to do a sort of "deep dive", in order to get everything into my head and push things forward. Things that were a bit rusty quickly came back after a little bit of effort.

At the end I'd emerge with a bunch of photos of whiteboards with clear intermediate steps, and then I'd write it up as a short report in LaTeX. Of course, as the report turned into a paper, I took out all the useful 'trivial' steps - which always feels a bit wrong. I think it's also useful at this stage to drop everything for a month or two and come back and see if you still understand what you wrote.

I think the "deep dive", putting prolonged effort in, and writing it up are all key aspects.

tfgg··on Douglas Rushkoff: I’m thinking it may be good to be off social media altogether
A decade ago? Facebook.

I think it's interesting that despite Facebook being vastly uncool, it still fundamentally meets its original need, which is socialising in a university environment.

tfgg··on Physicists Detect Gravitational Waves, Proving Einstein Right
(None of my points below say the graviton is massless, just that it's not crazy. As another post says, this new observation probably confines the graviton mass to be less than 10^-55 grams)

> If the massive gravitron was leaving a black hole it would be slowed by the black hole's gravity.

A graviton wouldn't be able to escape a black hole. A photon can't, and it's massless. The gravity of a black hole is actually a self-sustaining effect of the curvature of the spacetime around the black hole.

> (1) We should see this as some inconsistency in how gravity scales with the mass of a black hole. The larger ones would have proportionately greater 'drag' on leaving gravitrons.

We don't know details of the gravitational field around black holes and the mass that created it, because none have been observed close up. To an extent, the mass of a black hole is defined by its gravity.

> (2) If they are massive, and therefore subject to slowing, shouldn't gravity waves leaving a black hole be subject to some sort of doppler effect? Should we be looking for red/blueshifts in these waves?

Again, photons are massless and subject to the doppler effect. Gravitons, massless or not, will be too.

> (3) If gravitrons have mass and are subject to gravity, what brings that gravity? What sub-gravitron particle regulates gravity going in/to/out of the gravitron? This would require a new set of particles be created by non-gravitron massive objects (ie black holes) alongside the gravitrons. Like I said, too strange to exist.

Force carying particles can interact with themselves, c.f. gluons in QCD. In fact, GR is a non-linear theory so there will be non-linear interactions (as far as you can describe them in the weak limit).

tfgg··on Physicists Detect Gravitational Waves, Proving Einstein Right
To be fair, that article is pre-announcement and doesn't have much detail on the actual discovery.

Actual paper here: http://journals.aps.org/prl/abstract/10.1103/PhysRevLett.116...

tfgg··on Twitter's new timeline feature
I agree. I, and I suspect quite a lot of their users, stopped using Twitter for quite a while because the velocity was just too high, the signal wasn't good enough. The stuff you might have missed feature has actually made me use it much more, now I can actually catch those quality tweets.

I'm sure someone is going to tell me that I should unfollow more people, but that seems like a pretty crude and destructive method to try to increase the quality of my feed. I still want to see good quality tweets from the people who tweet a bit too much. I appreciate that use case (follow just a few accounts and read every tweet) though and definitely think they should cater to it.

tfgg··on Show HN: I've been writing daily TILs for a year
This is great.

As an extension, I was discussing with a friend a while ago how great Stackoverflow is at capturing the contents of expert's brains. Is there a way to achieve that on a broader scale without requiring the question-asking side of things? Say you're a lab scientist and you have a small trick or bit of informal insight, how could you be prompted in the right way to share it?

tfgg··on Jack Dorsey: “We never planned to reorder timelines next week.”
Are they going to 10k sized tweets, or some sort of notes facility? It seems like the main use case crying out is people attaching massive blocks of texts as images.
tfgg··on Best Paper Awards in Computer Science Since 1996
I was surprised at how low University of Cambridge is. Is everyone good working next door at Microsoft Research?
tfgg··on Ask HN: Do you use an old or 'unfashionable' programming language?
I also worked on a quantum (DFT) code in Fortran 95. I agree on your points, once you get used to it, it's basically isomorphic to C but with better numerical support. Compared to C++ it's verbose but understandable and with little magic, which is an advantage when working on a team, similar to Java. Main disadvantages I found were rubbish string and IO handling, no standard library of algorithms/data structures (lists, sort, etc.), and (in my codebase) there was still a bit too much reliance on global variables. Most of those are effectively fixed as the high-level controlling code is moved to Python and the F95 code is libraryized.
tfgg··on UK government urged to use Bitcoin-style digital ledgers
Having read the intro the report, I don't think there's anything they're suggesting that can't simply be implemented with a spreadsheet in git repo. But that isn't as sexy.

Government has already shown reluctant to share data, c.f. the Open Data movement. National and local government still use Excel spreadsheets, and you still come across the occasional scanned-in PDF. Why do people think they're suddenly going to adopt some flashy distributed ledger tech? Why do people think that's the important problem to solve?

It's a solution in search of a problem. It doesn't even solve the actual problems of trust and getting good quality data released. Blockchains often seem to be an expression of technologists' distrust of government, as though it'd all be better if other people could be replaced with code, when in fact it would undermine democracy [1].

The idea of GDS being mandated to explore the use of a specific technology is ridiculous. They should be using the technology that they choose to solve problems that people actually have.

Ultimately, this sort of hyperbolic all-in report does serve to pull the conversation in a particular direction (openness, machine-readable data), which is good, but it simultaneously undermines it by being likely to fail.

[1] http://tom.loosemore.com/2015/11/01/blockchain-vs-democracy-...

tfgg··on When Are You Really an Adult?
Appears to be a UK "Freeman on the land" [1,2] alternative legal theory. Can't work out much else since it's pretty incoherent.

[1] http://www.fmotl.com/forum/viewtopic.php?t=1693

[2] https://en.wikipedia.org/wiki/Freemen_on_the_land

tfgg··on Moore's law hits the roof
Given that Moore's law is about transistor densities, no I don't think quantum computers would qualify. It also seems premature to talk about about a Moore's law equivalent for the cost/density of quantum gates given we can barely build more than a handful at a time (D-Wave 'quantum' computers aren't quantum gate computers and don't scale in the same way).
tfgg··on Experian chooses UK authentication startup for Gov.UK Verify
Gov.uk Verify uses federated identity assurance:

> Experian is one of nine independent identity assurance providers supporting the government’s Gov.uk Verify identity assurance service, which is one of the government’s biggest digital projects

So if their software sucks, there are other providers to try.

tfgg··on Barren islands that countries never stop fighting over
In a way, isn't that just it? They're still 'fighting' over them because they're actually not that important? If they were important, the fight would be over much more quickly.
tfgg··on The quantum source of space-time
Indeed. When I saw a novel proposal for embedding QM inside GR using closed time-like curves (rather than the usual vice-versa), the main problem seemed to be that a GR-computer is potentially _much_ more powerful than a QM-computer, so you have to explain where the extra power has gone.
tfgg··on Physics Forests
Reminds me a lot of various machine learning approaches to molecular dynamics - basically trying to fit a cheap function to the energy landscape. I wonder if the more classical (rather than quantum) nature of these systems makes it more approachable?
← PreviousPage 3 of 8Next →