If you look more closely at the actual code, this exercise compares the performance of readFileSync (an operation that deliberately blocks) on 1 core vs 2 cores.
Initially they overlap which is one of the main cause of the jitter. Spreading out the initial locations would be a remedy, but then I would have to find a different way to explain "diffusion" to my daughter :)
This works as well (using jquery):
var a=0,t;setInterval(function(){t=$("#question").text().replace('x','*').replace('÷','/');t=t.slice(0,t.length-2);if(t!=a)$("#answer").val(eval(t)).keyup();a=t;},0);
There is a link to the source at the bottom of the page (http://www.straeto.is/rauntimakort/). I parse the JSON input, associate the records with previous ones (i.e. which bus is uniquely which) and stream output via websockets.
For continuous known functions we would look at the mathematical derivative (using stochastic calculus). For sparse sample data with an unknown derivative, we resort to calculating differences between consecutive numbers.
This analysis is interesting, however I have to comments on the methodology:
* Same measure
Comparing USD/INDEX with BIT/USD is a little bit like comparing apples and oranges, as the underlying measure for each of the time-series is different.
Multiplying BIT/USD with USD/INDEX gives us a time-series that is BIT/INDEX. Subsequently USD/INDEX can be compared with the new BIT/INDEX to get the correlation between USD and BIT with the INDEX basket as a measure.
* Correlation is only relevant for non-integrated series
It's well known that correlation for any unrelated random walks can easily be very high (positive or negative) in a spurious and random manner. The solution here might be to take the take first-difference (geometrical) of both series (de-integrating) and then check the resulting differences for correlations.
Not only that, regular algebraic operations also work with vectors/matrices as inputs, as long as you press CTRL-SHIFT-ENTER after editing. This means that most functions with complex MATCH and OFFSET can be replaced as well (i.e. SUM( (column_vector = cvalue) * (row vector = vvalue) * (matrix))
Matrix functions and named ranges are the most underused functionality of Excel. Collectively they can replace a large amount of the conventional spagetti VBA code, emphasizing Excel's true strength.
D3 is an excellent tool for managing a large array of SVG objects. Connecting it to leaflet is easy, just call function _initPathRoot() on the leaflet map object, followed by d3.select("#map").select("svg"). Then you can use map.latLngToLayerPoint(..) and map.latLngToLayerPoint(...) leaflet functions to calculate the X,Y coordinates in SVG space.
A great idea, however the web workers would have to be semi-persistent as the initial build of the crossfilter object takes some time (i.e. I wouldn't spawn a new worker on a new request). Here I simply increased the heroku nodes to respond smoothly to Hackernews traffic. I'm pretty happy with the results.