@beep It does look a lot like a barcode, but in a next iteration I want to take into account multiple hits on the same pixel, which should make it draw darker.
@tonzylstra Nice direction, ‘temporal spectrum chart’ or something…
@adewale That is the closest yet, but I think the fact that time is the horizontal axis here is a solid differentiator. I need to get the Visual Display of Quantitative data anyway, but I fear it won't be much help here because this is not a quantitative dataset.
in R, where 0 is initial time and each second is 0.1, for 20 seconds:
<pre>
> b <- c(0.1, 0.4, 0.5, 0.6, 1.0, 1.7, 1.8, 1.9)
> hist(b, breaks=20)
</pre>
18 comments so far
Do you mean a histogram: http://quarknet.fnal.gov/toolkits/ati/histograms.html
10 months, 3 weeks ago by adewale
No, I'm pretty familiar with those. I'm whipping up something in Prototype.
10 months, 3 weeks ago by alper
This might help to find the canonical name.
10 months, 3 weeks ago by kaeru
Boxplot?
10 months, 3 weeks ago by Liekje
@kaeru: Nice, but no cigar.
@Liekje: A boxplot would come close if I wanted the data even further reduced and the distribution displayed on a single dimension.
I just whipped this up: http://www.flickr.com/photos/alper/3197163008/ If it doesn't have a proper name, we should christen it.
10 months, 3 weeks ago by alper
in that case I suggest the "Alper Barcode Chart" :-)
10 months, 3 weeks ago by beep
spectral lines? ;)
10 months, 3 weeks ago by tonzylstra
@beep It does look a lot like a barcode, but in a next iteration I want to take into account multiple hits on the same pixel, which should make it draw darker.
@tonzylstra Nice direction, ‘temporal spectrum chart’ or something…
10 months, 3 weeks ago by alper
This reminds me of the output from spectral analysis.
10 months, 3 weeks ago by adewale
http://en.wikipedia.org/wiki/Emissionspectrum and http://en.wikipedia.org/wiki/File:Sodiumspectrum.jpg are examples of this.
It looks like @tonzylstra was right. These are indeed spectral lines: http://en.wikipedia.org/wiki/Atomicspectralline
I cheated and asked a passing physicist
10 months, 3 weeks ago by adewale
@adewale That is the closest yet, but I think the fact that time is the horizontal axis here is a solid differentiator. I need to get the Visual Display of Quantitative data anyway, but I fear it won't be much help here because this is not a quantitative dataset.
Most time series visualizations are also concerned with quantitative data, though this might be an interesting replot: http://www.math.yorku.ca/SCS/Gallery/images/LAoz.gif (by day of the week).
10 months, 3 weeks ago by alper
I'd call that a one-dimensional scatterplot (using a vertical bar as the symbol). It's not like it's the first time I've seen one of those, after all.
10 months, 3 weeks ago by effbot
it's still just a histogram, time x (did occur), just munge the data to be 1's and 0's
10 months, 3 weeks ago by termie
in R, where 0 is initial time and each second is 0.1, for 20 seconds: <pre> > b <- c(0.1, 0.4, 0.5, 0.6, 1.0, 1.7, 1.8, 1.9) > hist(b, breaks=20) </pre>
http://term.ie/data/quick_hist.png
10 months, 3 weeks ago by termie
hmm that got escaped funny it was:
<pre> b = c(0.1, 0.4, 0.5, 0.6, 1.0, 1.7, 1.8, 1.9) hist(b, breaks=20) </pre>
10 months, 3 weeks ago by termie
(you've obviously do some stuff to make the lines thinner, and blacker, and have much more data, and hide the the y axis)
10 months, 3 weeks ago by termie
Thanks for the additions, I think I'd probably should re-acquaint myself with R (I used to have some proficiency in SPlus).
10 months, 3 weeks ago by alper
It's funny but ever since seeing this page: http://en.wikipedia.org/wiki/Baader-Meinhof_Phenomenon about the Baider-Meinhof phenomenon I've started seeing references to the R programming language everywhere.
10 months, 3 weeks ago by adewale