Although there has been a dramatic growth in the development of autonomous vehicles and consequent competition between different companies and different methodologies, and despite the complexities of the task, the number of incidents remains remarkably small though no less tragic where the death of the occupants or other road users is involved. Of course, at present autonomous cars are not literally autonomous in the sense that a human agent is still required to be available to intervene, and accidents involving such vehicles are usually a consequence of the failure of the human component of the equation not reacting as they should. A recent fatal accident involving a Tesla Model X (e.g. Hruska 2018) has resulted in some push-back by Tesla who have sought to emphasise that the blame lies with the deceased driver rather than with the technology. One of the company’s key concerns in this instance appears to be the defence of the functionality of their Autopilot system, and in relation to this, a rather startling comment on the Tesla blog recently stood out:
No one knows about the accidents that didn’t happen, only the ones that did. The consequences of the public not using Autopilot, because of an inaccurate belief that it is less safe, would be extremely severe. (Tesla 2018).
So here’s a thing. A while ago, I asked whether there was any way to quantify the extent to which archaeologists were citing their reuse of data. I used the Thomson Reuters/Clarivate Analytics Data Citation Index (DCI) as a starting point, but it didn’t go too well … Back then, the DCI indicated that 56 of the 476 data studies derived from the UK’s Archaeology Data Service repository had apparently been cited elsewhere in the Web of Science databases (the figure is currently 58 out of 515). But I also found that the citations themselves were problematic: the citation of the published paper/volume was frequently incomplete or abbreviated, many appeared to be self-citations from within interim or final reports, in some cases the citations preceded the dates of the project being referenced, and in many instances it was possible to demonstrate that the data had been cited (in some form or other) but this had not been captured in the DCI. At that point I concluded that the DCI was of little value at present. So what was going on?
I’ve commented here and here about the question of data reuse (or more accurately, the lack of it) and the implications for archaeological digital repositories. It’s frequently argued that the key incentive for making data available for reuse is providing credit through citation. So how’s that going? I’ve not seen any attempt to actually quantify this, so out of curiosity I thought I’d have a go.
A logical starting point is Thomson Reuters Data Citation Index – according to its owners (it’s a licensed rather than public resource), this indexes the contents of a large number of the world’s leading data repositories, and, on checking, the UK’s Archaeology Data Service (ADS) appears among them. So far so good.
I’ve borrowed the idea of ‘deep-fried data’ from the title of a presentation by Maciej Cegłowski to the Collections as Data conference at the Library of Congress last month. As an archaeologist living and working in Scotland for 26 years, the idea of deep-fried data spoke to me, not least of course because of Scotland’s culinary reputation for deep-frying anything and everything. Deep-fried Mars bars, deep-fried Crème eggs, deep-fried butter balls in Irn Bru batter, deep-fried pizza, deep-fried steak pies, and so it goes on (see some more not entirely serious examples).
Hardened arteries aside, what does deep-fried data mean, and how is this relevant to the archaeological situation? In fact, you don’t have to look too hard to see that cooking is often used as a metaphor for our relationship with and use of data.
Solutions to the crisis in archaeological archives in an environment of shrinking resources often involve selection and discard of the physical material and an increased reliance on the digital. For instance, several presentations to a recent day conference on Selection, De-selection and Rationalisation organised by the Archaeological Archives Group implicitly or explicitly refer to the effective replacement of physical items with data records, where either deselected items were removed from the archive or else material was never selected for inclusion in the first place because of its perceived ‘low research potential’. Indeed, Historic England are currently tendering for research into what they call the ‘rationalisation’ of museum archaeology collections
“… which ensures that those archives that are transferred to museums contain only material that has value, mainly in the potential to inform future research.” (Historic England 2016, 2)
Historic England anticipate that these procedures may also be applied retrospectively to existing collections. It remains too early to say, but it seems more than likely a key approach to the mitigation of such rationalisation will be the use of digital records. In this way, atoms are quite literally converted into bits (to borrow from Nicholas Negroponte) and the digital remains become the sole surrogate for material that, for whatever reason, was not considered worthy of physical preservation. What are the implications of the digital coming to the rescue of the physical archive in this way?
The UK is suddenly wakening from the reality distortion field that has been created by politicians on both sides and only now beginning to appreciate the consequences of Brexit – our imminent departure from the European Union. But – without forcing the metaphor – are we operating within some kind of archaeological reality distortion field in relation to digital data?
Undoubtedly one of the big successes of digital archaeology in recent years has been the development of digital data repositories and, correspondingly, increased access to archaeological information. Here in the UK we’ve been fortunate enough to have seen this develop over the past twenty years in the shape of the Archaeology Data Service, which offers search tools, access to digital back-issues of journals, monograph series and grey literature reports, and the availability of downloadable datasets from a variety of field and research projects. In the past, large-scale syntheses took years to complete (for instance, Richard Bradley’s synthesis of British and Irish prehistory took four years paid research leave with three years of research assistant support in order to travel the country to seek out grey literature reports accumulated over 20 years (Bradley 2006, 10)). At this moment, there are almost 38,000 such reports in the Archaeology Data Service digital library, with more are added each month (a more than five-fold increase since January 2011, for example). The appearance of projects of synthesis such as the Rural Settlement of Roman Britain is starting to provide evidence of the value of access to such online digital resources. And, of course, other countries increasingly have their own equivalents of the ADS – tDAR and OpenContext in the USA, DANS in the Netherlands, and the Hungarian National Museum’s Archaeology Database, for instance).
But all is not as rosy in the archaeological digital data world as it might be.
At the same time, some of the assumptions behind Big Data were being questioned. It was no longer quite so straightforward to claim that ‘big data’ could overcome ‘small data’ by throwing computer power at a problem, or that quantity outweighed quality such that the large size of a dataset offset any problems of errors and inaccuracies in the data (e.g. Mayer-Schönberger and Cukier 2013, 33), or that these data could be analysed in the absence of any hypotheses (Anderson 2008).
For instance, boyd and Crawford had highlighted the mythical status of ‘big data’; in particular that it somehow provided a higher order of intelligence that could create insights that were otherwise impossible, and assigned them an aura of truth, objectivity and accuracy (2012, 663). Others followed suit. For example, McFarland and McFarland (2015) have recently shown how most Big Data analyses give rise to “precisely inaccurate” results simply because the sample size is so large that they give rise to statistically highly significant results (and hence the debacle over Google Flu Trends – for example, Lazer and Kennedy 2015). Similarly, Pechenick et al (2015) showed how, counter-intuitively, results from Google’s Books Corpus could easily be distorted by a single prolific author, or by the fact that there was a marked increase in scientific articles included in the corpus after the 1960s. Indeed, Peter Sondergaard, a senior vice president at Gartner and global head of Research, underlined that data (big or otherwise) are inherently dumb without algorithms to work on them (Gartner Inc. 2015b). In this regard, one might claim Big Data have been superseded by Big Algorithms in many respects.
[To interrupt the blogging hiatus, here’s the introduction to a recently published paper …]
Since the mid-1990s the development of online access to archaeological information has been revolutionary. Easy availability of data has changed the starting point for archaeological enquiry and the openness, quantity, range and scope of online digital data has long since passed a tipping point when online access became useful, even essential. However, this transformative access to archaeological data has not itself been examined in a critical manner. Access is good, exploitation is an essential component of preservation, openness is desirable, comparability is a requirement, but what are the implications for archaeological research of this flow – some would say deluge – of information?
Preservation by record is very much in the news at the moment in relation to attempts by ISIS to destroy cultural heritage in Iraq and Syria in places like Nineveh, Nimrud, Hatra, and the present threat to Palmyra. In some instances, the archaeological response has entailed excavations, in others it has been to begin to use crowd-sourced imagery to digitally reconstruct the heritage that has already been destroyed, or to use satellite and aerial imagery to map unrecorded and endangered sites.
Emma Cunliffe, from the Endangered Archaeology of the Middle East and North Africa (EAMENA) project, has suggested that “in some extreme, and particularly devastating, cases, the records may be the only thing left of a culture, in which case we owe it to them to preserve something, anything”. Hard to argue with that, and the article goes on to suggest that one approach to preservation of these sites is the use of archaeological technology to record monuments in high resolution in those areas which are still accessible (Foyle 2015).
In an earlier post I wrote about the importance of understanding the legibility, agency and negotiability of archaeological data as we increasingly depend on online data delivery as the basis for the archaeologies we write and especially as those archaeologies show signs of being partly written by the delivery systems themselves.
A simple illustration of this is the idea of filter bubbles. This term was coined in 2011 by Eli Pariser to describe the way in which search algorithms selectively return results depending on their knowledge of the person who asked the question. It’s an idea previously flagged by, amongst others, Jaron Lanier who wrote about ‘agents of alienation’ in 1995, but it came to the fore through the recognition of the personalisation of Google results and Facebook feeds (and is the counter-selling point of the alternative search engine, DuckDuckGo, for example). So can we see this happening with archaeological data? Perhaps not to the extent described by Pariser, Lanier and others, but still …