andrewducker | Interesting Links for 14-2-2011

andrewducker

The pernicious effects of When Harry Met Sally (part 7143 of "Why rom-coms are awful dating advice")

(tags: relationships friendship romance)
Did Nokia take Microsoft to the cleaners by getting billions in Windows Phone 7 deal?

(tags: nokia microsoft mobile_phones)
Fairtrade gold: not going to work I’m afraid

(tags: gold)
Nice visualisation of programming language popularity

(tags: programming language)
Enhance romance by going out with other couples

(tags: relationships psychology)
Liberal bias: science writing's elephant in the room?

(tags: politics science)
Fairtrade Gold.

(tags: gold trade rights)
Free Monty Python Videos on Youtube Lead to 23,000% DVD Sale Increase

(tags: youtube montypython copyright)
Rage comics: Women - what are they like?

(tags: women comic)
Danielle Tunstall - gorgeous, twisted, photography.

(tags: photography portraits art)
9 out of 10 specialists working in sexual health would give their daughters a private vaccine instead of the NHS one

(tags: vaccination uk nhs)
Intro video for LA Noire's gameplay. Looks pretty awesome.

(tags: gaming rockstar video)
The best description I've seen of why nothing can go faster than the speed of light.

(tags: physics universe light)

Crossposts: http://andrewducker.livejournal.com/2333123.html

Threaded | Top-Level Comments Only

From:

andrewducker

Cheers to

despotliz who sent me this:
http://www.dataists.com/2010/12/ranking-the-popularity-of-programming-langauges/
which is the origin of the graph used above (together with the methodology used to create it).

From:

gonzo21.livejournal.com

I'm not terribly surprised by the liberal bias in science writing. It illustrates one of the fundemental differences between rigid hide-bound conservative mindsets and free-thinking inquisitive liberal mindsets.

Fascinating stats on the Monty Python sales though.

And LA Noire looks awesome.

From:

simont

Nice visualisation of programming language popularity

I must say, I was disappointed by the article, given the graph. When I saw a graph that plotted number of github projects using the language on one axis, and number of Stack Overflow questions asked about it on another axis, I assumed they were going to focus on the difference between those two axes, not the general positive correlation.

I thought the interesting thing would be the languages which lots of projects use but about which very few questions get asked. Those are the languages in which an unusually high number of people are able to get their code to work without help – i.e. the easy to use ones, without complicated gotchas. The ones where you just write down what you want to happen, in simple and natural terms, and it happens.

(Of course, that might be due to things other than the language's quality. For instance, it might be selection bias based on the fact that nobody tries to use a given language to do anything difficult – perhaps because they give up on working out how to before even asking on SO...)

From:

andrewducker

There's also the selection bias of SO users (more Windows biased, I believe) and GitHub (more Open Source biased).

There don't seem to be that many that fall far from the mean - I assume that there are a lot of newbie c# questions out there, while emacs/lisp people know what they're talking about. Delphi seems like an odd one - loads of questions, but possibly doesn't mesh well with Git?

And coffeescript and supercollider are clearly languages to watch - presumably still in the hands of the bleeding edge devs, but if they're getting that much use then they'll start to go (more) mainstream.

From:

momentsmusicaux.livejournal.com

> "This study suggests that if your romantic relationship has a case of the doldrums, having fun with another couple may help make your own relationship more satisfying," said Slatcher.

... though I am given to understand that some people take this a little too far...

From:

momentsmusicaux.livejournal.com

Yes, that's kind of what I was thinking -- it should also tell you about relative complexity and pig-awkwardness.

For instance, we can perhaps infer from the graph that PHP is harder than perl because it has more SO questions.

Though I would say that in this case, it's rather because perl projects have old hands whereas every newbie wants to hack around with something in PHP.

From:

simont

I didn't know SO was more Windows-biased. Fair enough.

Delphi: now you mention it, off the top of my head, I'd expect any Windows-specific (or even Windows-biased) language not to get high scores on github in particular, just on the basis that Windows git is not nearly as impressive as Unix git. (I've been using git cross-platform between Unix and Windows at work recently, and it's pretty clear which is more polished. Not to mention that git has no support I can see for sorting out line-ending issues between the two.) And indeed, C#, F# and VB are all above the line too.

I suppose another question is about size of projects. A language typically used for trivial one- or ten-liners rather than 100,000 line epics might well get more projects per KLOC but the same number of questions and problems per KLOC...

Edited Date: 2011-02-14 12:11 pm (UTC)

From:

momentsmusicaux.livejournal.com

> Now, there are some mysteries here. Why can four-velocity vectors only rotate, and never stretch or shrink? There is an answer to that question, and it has to do with the invariance of the speed of light.

Great up to this point... but isn't this now circular?

From:

simont

The point where I started to get confused by that article was the point where it talked about "moving through" the time dimension. Moving surely means that your position changes over time – so, what, they're saying that there's a "time direction" in which your position changes relative to, er, time? Surely in fact the way the time dimension works is that a moving object exists on multiple points in the continuum, such that for a range of different time coordinates there's an associated set of space coordinates for the object.

And, following on the same thought, the natural expression of "can't go faster than x" in this metaphor is that your 'arrow' can't slant more than (say) 45° away from the time-only direction. By the time you get to an actually horizontal arrow, you're looking at something existing simultaneously in several positions at the same instant, which definitely isn't right.

I'm pretty bad at relativity myself, but I'm vaguely aware that these are indeed issues one has to consider – in particular, the distinction between "timelike" and "spacelike" directions, being separated by that critical 45° slant – and that explaining those drags back in a lot of the relativistic oddness that this explanation purports to do without.

So yes, I think this has failed to simplify into comprehensibility by instead oversimplifying into incoherence.

Edited Date: 2011-02-14 12:26 pm (UTC)

From:

spacelem.livejournal.com

Nice speed-of-light discussion, very informative! I approve and feel that I learnt something.

P.S. Tachyons.

From:

andrewducker

I've definitely heard it said that from a beam of light's point of view it does exist at all points along its length at the same time.

From:

andrewducker

You mean, have too much fun with it :->

From:

andrewducker

Also, Git doesn't allow you to check out small parts of a project (from my understanding). So a SVN repository might contain several related chunks of code, with the understanding that most people would check out the client, or the server, etc. but a Git one would tend to split those into different repositories.

What's the difference between git on Unix and Windows? I'm only using it in a tiny way, and entirely through an IDE, so I don't keep up with that kind of thing.

From:

momentsmusicaux.livejournal.com

Asda car park next Saturday as usual then?

From:

andrewducker

Not trashy enough. I say we meet in the alley behind Lidl.

From:

zornhau.livejournal.com

It's also true that an academic career is often incompatible with pursuing a conservative lifestyle.

Whatever the origin of the problem, in a democracy, it's a problem if the forces of reason can't make themselves understood to a big % of the population.

From:

simont

Also, Git doesn't allow you to check out small parts of a project (from my understanding).

Yes, that's one of my biggest peeves with it and all its ilk. No other DVCS is any better for that, as far as I can see. And if you work around it as you said, by dividing your project into multiple git repositories, then that loses you atomic commits across several of them at once.

I don't think there is much actual difference between Unix and Windows git. It's just that git is performance-tuned for Linux (not surprisingly given its origins), so on that it runs unbelievably fast whereas on Windows it's generally a bit sluggish; the per-file metadata for git is expressed in terms of Unix file permissions and Windows just has to make them up as best it can; git stores files as binary blobs, with no metadata indicating whether they're conceptually textual and should have line-ending conversion done on them; woe betide you if you try to do anything tricksy with case in filenames (I got into trouble last month renaming between "makefile" and "Makefile"). And I don't think anyone using Windows git here has managed to put their hands on a Windows analogue of gitk, without which it's a lot less convenient to find your way around the confusing tangles of branching that sometimes come up.

From:

eatsoylentgreen.livejournal.com

re: sexual health practitioners who would examine their daughters, I think that's true for most doctors. How many general practitioners would examine their kids for a sore throat? Probably all of them.

I personally wonder how many daughters would allow themselves to be examined by their dads!

From:

andrewducker

I've been using EGit, from inside Eclipse, and that seems functional, but there's plenty of stuff it doesn't quite do yet. People seem to be working on it though, so I'm hopeful that it will get more functionality in the near future.

From:

andrewducker

I don't think that they were saying anything about examining their daughters - just which vaccinne they should have.

From:

eatsoylentgreen.livejournal.com

oh whoops

From:

gonzo21.livejournal.com

Yeah, I think about that often when I consider the problem the Obama administration has in making itself understood to a large sector of the population that willfully invests and believes in utter staggering falsehoods.

But because Fox says it, it must be true.

From:

zornhau.livejournal.com

Ah - as I understand it - this is a slightly different problem... but maybe not.

There's this book called "The Political Brain", writen by a liberal psychologist, that uses proper studies and experiments to explain why liberals are so bad at getting their point across, that the general populace consistently vote against their own self interests.

Ultimately, that "large sector of the population that willfully invests and believes in utter staggering falsehoods" are the people you've got to convince. Probably best to learn how.

From:

pete stevens (from livejournal.com)

But there is one way to know, conclusively, whether you're moving. That's just to sit there patiently and wait. If the train's sitting at the station, nothing will happen. But if it's moving, then sooner or later you're going to arrive at the next station.

Er no, you can't tell if the stuff outside the window is moving and you're stationary. That's the first postulate of relativity.

You can't understand relatively until you realise that your intuitive understanding of the world is wrong. My favourite question to ram this home is,

I have two trains 1km long. I have a passing place 800m long. How fast do my trains have to travel in order to pass each other without crashing.

Until you can persuade your brain to stop going 'that's impossible' you don't understand relativity.

From:

d-c-m.livejournal.com

Rage comics: Women - what are they like?
*snicker* Yeah, pretty much sums it up.

From:

matgb

I read a rave review of that book once. Put me completely off even getting it from the library, the quoted exerpts and review of the thrust of the argument, all of which the reviewer meant to be positive, had me actually shouting at the screen due to the stupidity and misrepresentation.

It's a shame as I suspect hte book itself isn't as bad as the reviewer inadvertently made it appear but...

The general populace doesn't always vote against their own self interests. A significant minority do, but the case being made simply isn't proven.

There is, ultimately, very little chance or point of persuading those whose minds are already made up. You need to persuade those who are undecided. That's a much bigger chunk of the population than those that are alreayd opposed.

From:

zornhau.livejournal.com

Well, Political Brain guy would beg to differ, and cite evidence accordingly.

The thrust of his argument is that when the right go for the gut, the left go for statistics, and get drubbed accordingly. It's not duplicitous to state your values clearly, and powerfully. The left and liberals should do it more.

From:

matgb

13 years of Labour rule, only defeated as a result of actual, demonstrably, incompetence while in power, with a huge proportion of those not voting Labour voting for another avowedly left wing party, with the right getting less than 40% pretty much guaranteed, seems to indicate that the evidence may be a little too US centric and not true over here.

Then again, both Blair and Clegg got plaudits while campaigning for being able to state values clearly and powerfully, so perhaps the lesson didn't need to be learnt over here.

Then, so did Clinton, he did remarkably well electorally, and his strategy was based on the principle Brain appears to want to refute? I'm reaching on the last as I've not actually read it, obviously, only read about it more than once.

(caveat as you don't know this, I studied rational choice theory as part of my degree, and it was one of my preferred study areas, I'm now learning economics. I believe rational choice/game theory to be a bloody useful analytical tool, but it's just that, a tool, one of many that should be used, emotional resonance is also essential, but resonance on its own gets you nowhere)

From:

andrewducker

Lots of the left seem unhappy at the idea of selling ideas to people, and unwilling to use the tactics which work, because they feel underhanded to them. This definitely causes them to be less successful, IMHO.

From:

zornhau.livejournal.com

Not just unwilling. Look at the angry contempt.

From:

andrewducker

Well, yes. They actively consider those methods to be evil.

From:

zornhau.livejournal.com

I meant the angry contempt for all the stupid people. It's the sort of olympianism that the right accuse the left of.

From:

andrewducker

People on both sides seem to view others that don't see their view as either corrupt, stupid or just in it for themselves.

The difference seems to be that people on the right are happy to deliberately try to use use propaganda to move people in their direction, and the people on the left just rail against their stupidity.