barneyharris

eclectech, 23 days ago to random

It's time to get organised for the Friday Sparkly Dance Party!

Here's what you need to do if you'd like to join in:

1. Pick a dance (all 50 so far are below)
2. Film yourself doing the dance step a few times
3. Upload the clip to Mastodon, tagging me in and noting the dance number
4. Have a laugh with it, it's just for a bit of silly fun 🕺 💃

1 / 4

#FridaySparklyDanceParty

A vertical grid of 15 blocks of different colours, each numbered (1-15). Each box contains a simple animated black and white figure doing something resembling a dance move.
A vertical grid of 15 blocks of different colours, each numbered (16-30). Each box contains a simple animated black and white figure doing something resembling a dance move.
A vertical grid of 15 blocks of different colours, each numbered (31-45). Each box contains a simple animated black and white figure doing something resembling a dance move.
A square grid with 7 blocks of different colours numbered 46-52. Boxes 46 to 50 contain a simple animated black and white figure doing something resembling a dance move. 51 and 52 haven't been made yet, so are just question marks.

barneyharris, 24 days ago to uk

#UK, it's time to get the #ToriesOut

barneyharris, 6 months ago to Futurology

noob question about #reproducible #research #docker #rstats #python #academicchatter #foss

If I'm looking to use Docker to assist with making my research reproducible then should I rely on both Dockerfiles and Dockerimages or one or the other? As I understand it, the Dockerfile specifies how the Docker image will be built but an image can contain the scripts, original data, and correctly versioned software / libraries etc. So why not just / always publish the image?

barneyharris, 6 months ago to python

trying to get more serious around #reproducibility in research -- can any #rstats #python #gis folk chip in?

I use #macosx (first mistake, perhaps) and primarily R for my research.

Per project, I was thinking:

1. Create micromamba environment
2. Install base R
3. Install renv
4. Install R packages using renv + micromamba for dependencies (e.g. gdal)
5. Do research
6. Use exported micromamba env and renv lock files to rebuild the overall env in a linux Docker image
   ...

barneyharris, 6 months ago to Archaeology

Fantastic time visiting the Max Planck Institute of Geoanthropology in #Jena for project meeting and attendance of the superbly organised Big Historical Data conference https://bhdc.earth/ Particularly enjoyed learning about the powerful {pastclim} #rstats package. even spotted an old #ucl colleague on a poster.
#archaeology #climate #data #histodons

A poster for a lecture series at the Max Planck Institute of Geoanthropology, Jena. The poster shows an array of CGI objects including pots, rings, landscapes and charts.

barneyharris, 7 months ago to random

Interesting linear earthwork name #103: "Nutshambles", in Surrey. Interesting etymology too; Dorothy Nail (1965) "The name given there to the [nearby] hedge is 'Motschameles', which almost certainly has its roots in two Old English words 'mot' and 'sceamol [...] 'mot' or 'gemot' [...] is connected with an assembly of people, especially concerned with judicial matters, and is found frequently in combination with words for hill, such as
'hlaw' and 'beorg'." See https://eehe.org.uk/?p=29898

barneyharris, 8 months ago to random

#gischat #cartography peeps -- thoughts on 'overflow' panels for long linear features that would otherwise require an enormous scale (and loss of detail) to present contiguously. Should I add overview maps perhaps?