All entries for December 2018
December 17, 2018
Perches, Post–holes and Grids
Writing about web page https://arxiv.org/abs/1704.07342
I and my MSc student of two or three years back, Clair Barnes, produced an appendix, Perches, Post-holes and Grids, for a book being prepared by John Blair et al., arising from the project, Planning in the Early Medieval Landscape. You can find it on arXiv <https://arxiv.org/abs/1704.07342> of course! The appendix is aimed at demonstrating the application of statistical methods to the analysis of archeaological data, typically expressed in graphical form, with the objective of assessing the extent to which the spatial configuration exhibits planning by the original architects. Typical questions: did the builders use a common unit of measurement over a wide geographical region? to what extent is there evidence that they used a grid pattern when designing groups of buildings?
The name of the game is to contribute a statistical assessment to be mixed in with all sorts of other historical evidence. It's fun doing statistics in new areas like this: one learns a lot of stuff one didn't know before, and it provides a brilliant excuse to visit Anglo-Saxon Kingdoms <https://www.bl.uk/events/anglo-saxon-kingdoms> during the working week.
Crowd–sourcing data
Some really good ideas being implemented recently about and around the idea of crowdsourced data, For example:
- Seresinhe, C. I., Moat, H. S. and Preis, T. (2018), "Quantifying scenic areas using crowdsourced data" <https://journals.sagepub.com/doi/10.1177/0265813516687302> (may need institutional sign-in);
- "Cloudy with a chance of Pain" <https://cloudywithachanceofpain.com/>, involving a Warwick Statistics PhD student David Selby.
Of course the cartoonists got there first:
Noise to Signal: Rob Cottingham
<https://www.robcottingham.ca/cartoon/archive/2007-08-07-crowdsourced/>
Gold Access and so on
Writing about web page https://www.xkcd.com/2085/
Lots gets said about the importance of open-access publishing. Researchers are under pressure to publish papers which are "Gold Access" (translation: they pay the publisher quite a lot of money so that the paper can be accessed freely by all and sundry). Many people discussing this, and/or making policy decisions, appear not to have noticed that in many research fields new work is invariably released as a freely available preprint using the wonderful arXiv <https://arxiv.org/>, for which the publication cost is extremely low (mostly met by academic institutions). For example virtually all of my work of the last 14 years can be found there using <https://arxiv.org/a/kendall_w_1.html>.
The web-comic xkcd makes the point well <https://www.xkcd.com/2085/>.