- DONE assemble the sts references on number that I'm referring to -- Porter, Desrosieres, Hacking, Mackenzie, Verran, Scandinavia journal, Latour, Law, Callon
- DONE add all to zotero and set up bib files
- Write section on capital numbers - will need to include examples such as Facebook, perhaps genome, maybe iphones, etc;
-merged sss stuff into this branch
- need to check slides, but don't want to!
- starting changing mixtures slides -- put in knorr cetina on post-social; had idea that 2 facets of the paper -- github + ethnographic sensibilithy; under what conditions is it possible to write with numbers
- check guidelines for sss [https://uk.sagepub.com/en-gb/eur/journal/social-studies-science#MANUSCRIPTSTYLE]
- look at 1970s work on counting and citations and see how it matters today -- use WoS
- look at recent marres/gerlitz article in sociological review for relevance of interface methods (device specific research, etc)
- look up critical making -- Ratto, etc. -- see sss zotero folder for reference -- and see if it helps say what I'm doing
- read Mackenzie on gaussian copula for stuff on methods
- read verran paper on relational empiricism from care issue of sss to follow up on verran
- characterize github as only one of top 100 media sites that is about coding
- perhaps add timeline as organizing narrative of the paper? -- see timeline branch for latest version
- add some open source literature like Kelty & Coleman
- if using capital number, write a short passage describing it
- will probably be a combination of desrosieres and verran, with some boltanski/thevenot?
- list sections/minianalyses that need to be written
- in account of what people do with the data, make sure to mention that this recursive work on the platform is also what capitalises the number more wrote abstract
playing with bigqueries to look at the repo census again. Tried
- stus basic repo event count -- can use this to cut up repo count
- started some work on durations using Richard's repo census query which makes a wide table of events, etc.
Needed to modify queries as they were all using repo names not urls. Also saved repo event count top 1000 as csv to data in data_analysis. Can easily plot this to show power law distribution of the event counts. Could also fit distributions using poweRlaw package, but not sure what the point would be ... But the main thing is that the cumulative event counts show very quickly that most repos have few events. Could then add to that the number of repos that are forks.
- looking at abstract again - it seems to be doing too much. Maybe I need something simpler. Oh, but then I was looking at the old abstract. The new one is in plan.md
- did the number bibliographic work -- need to assemble the argument, tho.
- seems very little sthv but quite a bit in sss. Not sure why that is. Particularly relevant pieces would be Lampland on False numbers; Law & Callon on qualcalculation;
- also put refs from Distinktion into zotero folder -- nafus, lury, etc, should be useful
- had the idea of setting out a middle ground between science and technology, between literature and things inhabited by software and other such things
- also had the idea of reassessing counting in sts more generally -- how do we do it ourselves?
- drifing through Guyer on percentage and ordinals, Dumit on implosion,
- had the idea that numbers are a way to think about the bits of infrastructure that are forgotten or hard to study in infrastructural studies ... borrowed this from sss manuscript
- spent quite a lot time remapping keys
- looked up Derrida on capital -- if I'm going with the notion of capital numbers, then the different senses of capital -- liable to death (loss of head?), standing at the head (city or a name), accumulated wealth, etc
- these different senses suggest possibility of looking at names in terms of capitals?
- re-wrote new abstract again. It's still too long. Need to cut down to 5-6 sentences. At the moment, its 280 words.
publication/material_cultures_london2016