Following on from the quick post on my thoughts last week, I’ve put together a Storify archive of the tweets from The Connected Past 2014 network and complexity science meeting with a little commentary included. I enjoyed the conference and found it very inspiring, so I’m happy to collate some of the tweets which hopefully give you a little taste of what was discussed.

If you’re interested in more information many of the slides are now up on Figshare, courtesy of the authors and the conference organiser, Tim Evans (Imperial). The meeting came out of the Connected Pasts community, so keep your eyes out for similar things in the future; I am definitely hoping there will be more meetings in the not-so-distant future!

Until Storify talks properly with WordPress I’ve put the bare-bones html below: if you want the more attractive full Storify experience, go over to the site. Otherwise, enjoy!

The Connected Past 2014// app:

The Connected Past 2014

Connected Past 2014 was a one and a half day multi-disciplinary meeting exploring how concepts and techniques from network and complexity science can be used to study archaeological data. It took place Monday 8th September and Tuesday 9th September at Imperial College London.

  1. Whilst the meeting did not formally start until Monday afternoon, Tom Brughmans (University of Konstanz) ran an introductory workshop in the morning. This included an hour or so talk introducing key concepts in network science, followed by an hour or so of working through a tutorial Tom put together on Cytoscape, utilising some nice data from Spain during the pre and Roman period.
  2. #tcp2014 20 mins in and I already want to do network analysis on ALL THE THINGS.
  3. The conference was organised locally by Tim Evans from Imperial, with the assistance of Tom Brughmans (Konstanz), Ray Rivers (Imperial), Anna Collar (Cambridge/Aarhus) and Fiona Coward (Bournemouth), but is part of a broader Connected Past community, lead by a multidisciplinary team.
  4. The workshop was really enjoyable and very inspiring, and very well attended. Whilst the majority of people were archaeologists, there were a few historians, geographers and a scattering of scientists. The papers in the afternoon saw a wider audience, incorporating a substantial number of computer scientists and physicists, but both showed the international nature of the meeting, attracting people from across Europe and the Americas.
  5. Attending the Connected Past @imperialcollege for #tcp2014, looking forward to lots of great papers on network science and archaeology!
  6. Prignano: work by Fulminante on this dataset suggested closeness was useless #tcp2014 http://t.co/JuoE7sff1Q
    Prignano: work by Fulminante on this dataset suggested closeness was useless #tcp2014 pic.twitter.com/JuoE7sff1Q
  7. I wish the physicists stopped saying ‘the archaeologist did this very simple research’. We don’t mention the ‘3 body problem’ to you, do we?
  8. After an introduction from Tim, Francesca Fulminante (Cambridge University) presented a paper coauthored with Sergi Lozano and Luce Prignano on Networks and Urbanization in central Italy (1175/1150-500 BC ca). This introduced something of a theme for the meeting, which featured several papers presenting archaeological data which had been analysed by scientists, and in some cases inevitably fell into the trap of downplaying the uncertainty and complexity of the archaeological data.
  9. Da Vela: Hypothesis of breakdown of network of north etrurian towns due to romanization, with triad census #tcp2014 http://t.co/TBBpBep7Vs
    Da Vela: Hypothesis of breakdown of network of north etrurian towns due to romanization, with triad census #tcp2014 pic.twitter.com/TBBpBep7Vs
  10. de Vela uses both roads/waterways AND viewsheds to create the links between nodes (settlements) in her network. #tcp2014
  11. After this, Raffaella Da Vela (Rheinische Friedrich-Wilhelms-Universität Bonn) presented a paper on The Network of the North-Etruscan Settlements during the Romanization, which showed an interesting approach to combining multiple types of archaeological data, but which did garner a few questions on how ‘style’ and ‘ideas’ could be accurately identified moving between urban sites.
  12. Mezza Garcia: ancient social hierarchies. proposal by large team of multi-disciplinary researchers #tcp2014 http://t.co/0bEt466t37
    Mezza Garcia: ancient social hierarchies. proposal by large team of multi-disciplinary researchers #tcp2014 pic.twitter.com/0bEt466t37
  13. Nathalie Mezza-Garcia (Universidad del Rosario, Bogata, Columbia) was up next presenting a paper coauthored with Tom Roese and Nelson Fernandez entitled Computational Aspects of Ancient Social Heterarchies: Learning how to Address Contemporary Global Challenges. This was a paper outlining a possible project which a multidisciplinary team in Bogata are hoping to undertake in the near future, and was followed by Miljana Radivojević (UCL Institute of Archaeology, London) and Jelena Grujić presenting a paper on Tracing metal networks in the Balkans at the dawn of the Metal Age.
  14. Very engaging double-act on network analysis and early metallurgy by Radivojevic and colleagues #tcp2014 http://t.co/kI5G9GQF4U
    Very engaging double-act on network analysis and early metallurgy by Radivojevic and colleagues #tcp2014 pic.twitter.com/kI5G9GQF4U
  15. Typical archaeologists...arrows going everywhere on our distribution maps :P #tcp2014 http://t.co/zxlGS7mjHd
    Typical archaeologists…arrows going everywhere on our distribution maps 😛 #tcp2014 pic.twitter.com/zxlGS7mjHd
  16. After a coffee break, we were treated to an hour long keynote talk by Prof. Sir Alan Wilson from CASA at UCL.
  17. Sir Alan Wilson giving the first keynote talk at The Connected Past #tcp2014 http://t.co/rhdFqUlTXd
    Sir Alan Wilson giving the first keynote talk at The Connected Past #tcp2014 pic.twitter.com/rhdFqUlTXd
  18. Wilson giving v. similar arguments to Jensen’s @sccs2014 talk: Boltzmann’s statistical mechanics can be applied to hist/archaeo. #tcp2014
  19. Wilson modelling settlements, wars, piracy, London riots … Giving us an overview of his work over the decades #tcp2014
  20. Wilson's concluding remark on working with archaeology and history and complexity science #tcp2014 http://t.co/4UQy8Pkcmv
    Wilson’s concluding remark on working with archaeology and history and complexity science #tcp2014 pic.twitter.com/4UQy8Pkcmv
  21. inevitable shameless question by @Iza_Romanowska “archaeologists are champions of shit data! Can that inspire method development?” #tcp2014
  22. Wilson’s talk was very well received, and it was fascinating to see how his involvement with network and complexity science has influenced the field over the last few decades. In particular whilst he showed a number of equations relating to his work modelling flow and use of shopping malls, all of them were well contextualised, making his talk accessible and interesting to all of the attendees.
  23. Interesting talks on network analysis in archeology and complexity science on day one of #tcp2014, now off for day two!
  24. The Monday wrapped up with a social evening at Imperial’s Union Bar, but everyone was back bright and early on Tuesday morning for the second keynote talk. This was given by Joaquim Fort, and examined the Neolithic transition and the various ways of modelling the spread of agriculture from the Near East.
  25. Joaquim Fort @SimulPast is giving the 2nd keynote at #tcp2014 on dermic diffusion, cultural transition and the spread of the neolithic.
  26. Kicking off Day 2 of The Connected Past meeting with Neolithic transitions and big data #tcp2014 #bigdata @imperialcollege
  27. Fort: isochromes of dated Neolithic sites in Europe, revealing different speeds of spread in km/y #tcp2014 http://t.co/g3XZsly0Rz
    Fort: isochromes of dated Neolithic sites in Europe, revealing different speeds of spread in km/y #tcp2014 pic.twitter.com/g3XZsly0Rz
  28. Fort: to reduce effect of v.early or late dates as compared to surrounding dates, interpolation is smoothed #tcp2014 http://t.co/MbmZ7nTRBx
    Fort: to reduce effect of v.early or late dates as compared to surrounding dates, interpolation is smoothed #tcp2014 pic.twitter.com/MbmZ7nTRBx
  29. This was followed by a paper by Xavier Rubio-Campillo (Barcelona Supercomputing Center) with five other coauthors on Comparative approaches to the use of archaeological data inside computer simulation.
  30. Rubio: new approach archaeologists + comp scientists doing eXtreme programming together to create the model. #tcp2014 Awesome!
  31. @xrubiocampillo lessons learned that are useful for all of us doing simulation modelling. Useful slide!!! #tcp2014 http://t.co/dBPjwj7K89
    @xrubiocampillo lessons learned that are useful for all of us doing simulation modelling. Useful slide!!! #tcp2014 pic.twitter.com/dBPjwj7K89
  32. Question for Rubio-Campillo: “How do you get archaeologists and computer scientists to sit down together?!” “We pay them”. Aha! #tcp2014
  33. Xavier Rubio-Campillo explaining comparative approaches to archaeological data in comp. simulation #tcp2014 #bigdata http://t.co/46fQTHlrkJ
    Xavier Rubio-Campillo explaining comparative approaches to archaeological data in comp. simulation #tcp2014 #bigdata pic.twitter.com/46fQTHlrkJ
  34. Following a coffee break we returned to hear Shumon Hussain (Department for Prehistoric Archaeology, University of Cologne) coauthor a paper with Adreas Maier about Path dependency: explaining higher order phenomena in the Lower to Upper Paleolithic record of Western Europe. This was a fascinating introduction to a theory which could have multiple applications to archaeological contexts, particularly technological systems, and really inspired me.
  35. Shumon Hussain explaining path dependency in culture change through QWERTY keyboard adoption #tcp2014 http://t.co/QriQBmoTNY
    Shumon Hussain explaining path dependency in culture change through QWERTY keyboard adoption #tcp2014 pic.twitter.com/QriQBmoTNY
  36. This was followed by Laura Perucchetti and Peter Bray (Archaeological Science, Oxford Univ.) coauthoring with Mark Pollard and John Pouncett to present a paper entitled Beyond provenance: Copper-alloy chemical signatures as proxies for human technological practice in the past. This took a similar problem as that tackled by Radivojević and Grujić’s earlier paper, but applied a completely novel approach to it to produce some quite simple but very interesting visualisations.
  37. Perucchetti and Bray talk about copper alloy signatures as proxies for technological practice #tcp2014 http://t.co/5LSAuXMmAv
    Perucchetti and Bray talk about copper alloy signatures as proxies for technological practice #tcp2014 pic.twitter.com/5LSAuXMmAv
  38. Christine Johnston (Cotsen Institute, UCLA) followed up, with a talk on Object Distribution, Networks, and Imperfect Datasets: An examination of market exchange at Ugarit, where she presented the only paper looking at relationships within a single site, by focussing on a small number of types of object.
  39. Johnston talking about object distributions, networks and imperfect datasets #tcp2014 http://t.co/5lSssIFZxR
    Johnston talking about object distributions, networks and imperfect datasets #tcp2014 pic.twitter.com/5lSssIFZxR
  40. This was followed by a brief interlude from Alice Oven talking about a new book being published by Imperial Press, and about her publishing house’s interest in works on network and complexity science.
  41. Alice Oven talking about @electricarchaeo 's new book on Big Digital History, lovely ideas being shown! #tcp2014 http://t.co/GoyuYXwXAP
    Alice Oven talking about @electricarchaeo ‘s new book on Big Digital History, lovely ideas being shown! #tcp2014 pic.twitter.com/GoyuYXwXAP
  42. After lunch we were treated to the final keynote talk, which was by Prof. Ulrik Brandes (University of Konstanz), who de-constructed a paper recently published in Nature to underline the importance of establishing the relevance of your network hypothesis before applying network analysis techniques.
  43. Ulrik Brandes giving the third keynote at the connected past in imperial college London #tcp2014 http://t.co/hcz1sX8dyS
    Ulrik Brandes giving the third keynote at the connected past in imperial college London #tcp2014 pic.twitter.com/hcz1sX8dyS
  44. Brandes: paper by Jeong etal 2001 was flawed from the very beginning because it was based on analogy rather than domain knowledge #tcp2014
  45. Here a summary of Brandes' critiques: data does not support hypothesis, no biological explanation #tcp2014 http://t.co/kdTkXkgmvU
    Here a summary of Brandes’ critiques: data does not support hypothesis, no biological explanation #tcp2014 pic.twitter.com/kdTkXkgmvU
  46. Brandes: different centrality indices perform differently across organisms. Indices are not universal or neutral #tcp2014
  47. Brandes: network science is no different than statistics other than that the data is different, network data #tcp2014
  48. Brandes: theoretical ideas need to always be part of the creation and selection of techniques #tcp2014
  49. Brandes: cross disciplinary collaboration is not socially desirable but necessary! AMEN! #tcp2014
  50. Fantastic keynote by Brandes #tcp2014! Raising key issues, explaining stuff in the clearest of terms and really engaging at the same time.
  51. … So what’s necessary is the need for cross-disciplinary collaboration guided by archaeological reasoning #tcp2014
  52. After Brandes very well received talk the meeting ajourned for coffee, and returned to hear Anna-Katharina Rieger (Max Weber Institute of Advanced Cultural and Social Studies, University Erfurt) talk about Region, religion and the question of the evidence: Is it possible to
    approach sacred spaces in Southern Syria with network analysis tools?
    This paper was more of a discussion of what might be possible with network tools given sufficient support by someone experienced in the analysis, but it was helpful in highlighting the importance of the context and meaning of archaeological data for any resulting network analysis.
  53. Now Anna-Katherina Rieger giving us a lot of archaeology related questions #tcp2014 http://t.co/EH2epho8lJ
    Now Anna-Katherina Rieger giving us a lot of archaeology related questions #tcp2014 pic.twitter.com/EH2epho8lJ
  54. Rieger is interested in exploring different types of spatial and social connectivity #tcp2014
  55. After this was a paper by Elsa Arcaute and Stuart Brookes from CASA, UCL, talking about Natural regional divisions of places in Domesday Book. This talk was a fascinating look at data taken from the Domesday Book and how to analyse this to tease out archaeologically meaningful information.
  56. Arcaute etal presenting natural regional divisions of places in Domesday book #tcp2014 http://t.co/kDA2rEW4Rm
    Arcaute etal presenting natural regional divisions of places in Domesday book #tcp2014 pic.twitter.com/kDA2rEW4Rm
  57. Arcauta et al good example of multi disciplinary team. Geographers from casa and archaeologists #tcp2014
  58. Arcaute etal: correlation cluster identified through method and weekly income per capita of regions England #tcp2014 http://t.co/DdmZ5nU499
    Arcaute etal: correlation cluster identified through method and weekly income per capita of regions England #tcp2014 pic.twitter.com/DdmZ5nU499
  59. Mid-Saxon political geography approximated by percolation process analysis, v. cool! #tcp2014 http://t.co/ebONdm2doI
    Mid-Saxon political geography approximated by percolation process analysis, v. cool! #tcp2014 pic.twitter.com/ebONdm2doI
  60. The last talk of the day was by Maeve McHugh (UCDublin) entitled Finding meaning in a digital landscape: GIS modelling and ancient Greek agriculture. This paper was the only one to draw at all on literary sources, though epigraphy and other written sources have been used in network science applications for some time.
  61. "I forbade the assdrivers, as I said, to touch the wood." *snigger* I am a child. #tcp2014 http://t.co/3o6fXotA5v
    “I forbade the assdrivers, as I said, to touch the wood.” *snigger* I am a child. #tcp2014 pic.twitter.com/3o6fXotA5v
  62. Overall the conference was an extremely profitable two days, filled with all the things that make meetings in new areas exciting; inspiring work by people who aren’t afraid to experiment, a friendly attitude, and a positivity towards the discipline that is really infectious.
  63. The concluding remarks from @tombrughmans and Tim Evans. Great conference, inspiring talks, fantastic event 🙂 #tcp2014
  64. The network science in history and archaeology meeting this week was overall epic. Young, early-career, exciting and novel. #tcp2014
  65. Most of the slides from the presentations can be found on line, and they’re well worth a look. I’d certainly recommend going along to the next meeting, if it’s near you; even if network and complexity science isn’t your thing, the questions being asked and the analyses undertaken in this area are really interesting and thought-provoking.

Advertisements

One thought on “Tweets from The Connected Past 2014

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s