 Mae gyr autochio rydyn ni hefyd. So mae'n gwybod nhw'n rhoi gyda Claren, Victoria ac Donnell yn dda i ddaw'r cyflug y tu coll gwrs. If you've not seen them- They certainly helped me to motivate a little bit about what I'm going to talk about today. I'm going to talk a little bit about the questions that we're interested in and then reasons for going to crowd-sourcing which initially were, I guess, more financial lle ahead o gyhoeddwyl, ond maeddemu ar lawr. Ond o'r cyfnod oedd y cyfnodd nesaf ayw ar nhw'r cyfnodol anhun uchydigol ac mae'n just yw y seithio ymlaen i gael bod y cyffredin a gwahau ei anzach yn eistedd dyma'r gweithio. Ond roeddwn ni'n gweithio'n cyfrif i'r cyfnodd, alech nhw'n rhan o'r hyn ystod o'i werth o gynnig â'r bêis. It is interesting both in terms of hearing about Senitaff and also about the work that we're going to be looking at here that we're in to Papa and we've got this wonderful display downstairs about Glipley. Both Senitaff and the work we've got has access to all the records of all World War I soldiers including those that served at Glipley and we know eventually we'll know a lot about the characteristics of our soldiers. journalists. So, wat was sort of reason why we got into wanting to measure the ANZACs? And in particular we are really interested in trying to use certain characteristics measures of … measures of BMI, measures of height and weight as alternative ways of measuring well-being. I'm an economist and we often get told that using things like GDP is a really bad way to measure well-being and one of the alternatives is to look at stature. Stature is really determined by the first 20 or 21 years of your life. It's based on net nutrition and effectively net nutrition is affected by what you eat and the diseases that you have had as someone between nought and 21 years old. So it's an alternative way of measuring life experience that is recorded in me obviously I had a very difficult childhood. I did but it doesn't really matter and it wasn't about food it was about something else and in large samples genetic variation which is a thing that normally people ask you about really gets washed out so we need access or we needed access to large database as on height and weight and the original sort of questions that we asked and I'll ask you the question now and see whether you can answer it. We wanted to know things like when did New Zealand does become obese? Anybody got any answers? Where would you go? We went to health scientists and they said exercise is good for you. That will help you stop being obese. I said that's not quite asked. You could have your stomach stapled. No no no I said that's not what asked and there was nowhere out there where we could find out when did obesity become a problem in New Zealand when it become an issue and some of the work that we've done helps answer that question but obviously quite a lot more questions as well. So reductions or improvements in net nutrition health what you eat the sort of stuff that you eat will affect what your potential and actual stature your height is and if we look at different groups different cohorts over this period we can track what is happening to the stature what is happening to the well-being of population over time. Now obviously we're not looking at the same people we're talking about people that got to 20 then another group that got to 20 but then we can link those together and give a time series as what's happening to heights and weights. We need large samples because we have to take out the genetic variation that we might find in the population. So to get large samples and this is not just a New Zealand issue this is a global issue you tend to rely very heavily on military and prison record sources organizations that routinely measured heights weights and a whole bunch of other characteristics about people including social backgrounds and data birth etc. So these were the three original sorts of specific questions that we asked or wanted to try and answer in terms of to what extent and how have the health and biological standard of living changed in New Zealand from in our case 1850 right through to the 2000s. In a country like New Zealand you are almost inevitably drawn into looking at whether there were differences between indigenous Maori and settlers and the typical story will be of course there were differences we all know that. The reality is that we have very little until now evidence to suggest how big those differences were and also how did these biological measures of living standards confirm or refute other sorts of measures that we have out there about well-being like GDP or consumption or wage growth. Remember I'm an economist we have to have some numbers sometime so when we look where do we get our data from well this is a set of New Zealand expeditionary force what they call attestation papers attestation for that read enlistment so these are a bunch of papers which were filled in by each person who enlisted in world war one and they have names they have addresses watch your job contact details I've served in the forces before do you swear to obey various other things these records are the things that ultimately we want into huge crowdsourcing for because as you can already probably see even in the first couple of roads roads we've got some where's my little pointer on here well if I could find a little point here we go very small this is all written in script in pencil oh script in pencil and it's very difficult to trans to transcribe that in an algorithmic way so we've got people's names we've got people's dates of birth etc etc they're written in old fashion script and they're often in pencil and difficult to interpret here's another aspect that's a problem we have sticky labels that get put on these files these history sheets give a wider range of things of interest perhaps to others than to us where did you do your service what wounds if any did you have what periods of sickness did you have what type of sickness was it and did you die were you missing were you a prisoner when were you discharged that information is currently available when we were looking at this it was archway well when we started looking at this it was paper records in archives then it became available through archway and I guess senator tough is another source of some of this stuff the second source of data that is commonly used and one that we use is from prison records and the particular project I'm going to talk about today does not look at prisoners just looks at world war one soldiers but here's an example of a prison record which gives details on hair colour height weight trades aliases and importantly you can see that the person here is showing all of their fingers as a way of identifying whether they've got any distinguishing marks like missing fingers and various other things again we're not particularly interested in that but these records have the potential to interest a whole bunch of other humanities and social scientists particularly when it comes to say examples of the prevalence of tattoos missing fingers and various piercings so while we found so far well so far we've spent about three quarters of a million dollars from various sources to look at 25 000 world war one soldiers records around 25 000 prisoner records and a smaller number of world war two soldier records and we find when we look at mean stature or height that we've got periods where there's common decline in stature between mary and parkihar in the 1890s through to 1900 1900 is probably the low point in population for the mary population divergence and persistent divergence from 1900 right through effectively to the 1960s so that's one set of questions that we could answer from these data that haven't been answered before secondly in answer to your question that i didn't give you a chance to actually shout out when did bmi when did obesity become a problem in new zealand well we can see again depending what you're defining obesity to be from these prison records we can look at differences in bmi between mary and parkihar and we can look at averages and we look at trends and again there's no other way to get this data no other way to answer this question i've missed that one out on so what have we found so far there certainly were health disparities that emerged between the mary population and settlers again you might say yeah we all knew that well we could be able to put a little bit more meat on the bones in terms of when that happened was new zealand different well there are some similarities between what we find in terms of us studies and some differences in that between world war one and world war two there seem to be a significant reduction in mortality risk associated with being overweight again something that wasn't known we've only got 20 25 000 samples not small but not big enough we haven't got enough interactions we don't know whether there are class differentials wealth differentials affecting these things because we haven't collected the data and able to link it to individual records we didn't collect stuff about combat exposure and it's interesting if you're interested in post-traumatic stress to look at some of the service records of these people that were disciplined for things that we can now clearly see as post-traumatic stress disorder we didn't collect that but it's available and we didn't do anything because our samples were relatively small about women so with all those problems we went off to crowdsourcing and we went to zooniverse why zooniverse was zooniverse is hosted at minnesota evan robertz is at minnesota and zooniverse is supposed to be and and donnell will be able to tell us otherwise one of the largest crowdsourcing citizen science sites around and as zooniverse describes itself it provides opportunities for people around the world to contribute to real discoveries in fields ranging from astronomy to zoology and zooniverse originally gets its name from the idea that i'll get ahead of myself zooniverse had its first project which was galaxy zoo which was really trying to categorise different types of galaxies so what are these well they're two different types there's a spiral galaxy and a non-spiral that's a predator and that's prey here's two types of ships but they are different and here's i think some cell biology i don't know that one here's some stuff i think which i um donnell was talking about in terms of transcribing ancient documents and here's various weather systems so zooniverse came about partly because at the time when i think even now the ability for algorithms to look at these sorts of pattern and pattern recognition capabilities are beyond or if not beyond certainly a very computer intensive and what zooniverse was put in place to try and bring to fruition was the idea that the best sort of algorithm that we have out there is the brain and let's have lots of brains trying to sort out this problem rather than lots of computer power which probably wouldn't have been enough anyway so here's galaxy zoo the original zooniverse crowdsourcing project started in 2007 classifying types of galaxies here's their first two days so we got classifications per hour a number of hours and one of the problems and the problem that we faced over the weekend was that the server fried in about three hours the interest was such that the technology couldn't keep up with the citizen scientists who were desperate to get involved in trying to classify galaxies now galaxy zoo was the sort of origin of this whole process moving beyond galaxy zoo we've got three examples here of the the type of interaction citizen scientists can be involved in going from passive seti type ideas to data collection which is in some sense partly what the the classic citizen scientists were originally involved in and we're interested in data analysis and that's where zooniverse really think we we think that really have the edge now we need citizen scientists in part because the financial implications of trying to do this sort of work are just huge but what do the citizen scientists get out of this this is a two-way street and basically if you're looking at why in particular galaxy zoo was successful the vast majority of reasons why these citizen scientists stated that they were interested in doing this stuff for free was that they wanted to make a contribution to knowledge some of it is about the beauty of the universe some of it is about fun which is interesting in terms of what danelle said but the vast majority of people wanted to be scientists and make a contribution so if we look at zooniverse now it's got over a million volunteers it has i think currently 32 projects of which are measuring the anzacs is is the most recent and it has around and i'm sure this number is much bigger now 60 peer reviewed publications which include acknowledgments to the citizen scientists who provided the data again an issue that was raised in the previous presentation here's what's happening to citizen science over time it typically started off with lots of purple dots which relate to space and categorising climates measuring number of planets those sorts of things and has moved a little bit further into a broader range of issues but still relatively few in terms of humanities and social science so here we've got i think measuring orcas 10 000 volunteers 150 000 classifications so far here's our ship's logs 25 000 volunteers a million pages transcribed ancient lives 300 000 volunteers and 100 million classifications and based on two hours ago measuring the anzacs had got 60 000 classifications five and a half thousand users and 461 completed documents so what does a zooniverse site look like zooniverse basically has this snapshot at the top which is the primary task interface as even wishes to describe it which is the project but there's some important aspects really important aspects that have to be part of a successful project one is that there needs to be abilities for forums and blogs and talks and peer help and interaction and there need to be this development of a community zooniverse is great because it also has zoo tools i hate these sorts of words has to have zoo in front of everything so it has zoo tools which are basically ways that the users who are the citizen scientists themselves can make use of the data to actually do some primary analysis if they want to it's great for the classroom because again we can use zoo tools for students to either map these things or do geospatial issues but the things like the discussion forum and the communications are really crucial it seems and we're finding this already to making a citizen science project a zooniverse project successful and these two at the bottom are just in beta versions so the two beta versions are things like citizen science journal in its own right so let me squeeze out of this so peer-to-peer mentoring is really important so what does zumanities give us well we want into transcribe 140 000 first world war soldiers records many images many pages i think there's two and a half to three million pages that are now pdf files that are able to be classified we want to in our own work capture some work from those attestational enlistment forms in particular to do with heights weights ages where you were born etc and we hope to finish doing this by 11 11 2018 which is for those of you that know anything about world war one the end 100 years since the end of world war one so here's those forms again here's the sort of things that we're interested in capturing in all we've got the possibility of about 147 fields or columns of material that could be collected from six different types of documents and we can once we've got that data entered we can do a whole bunch of things with it so here's measuring the anzac site at zuniverse.org you need you can either choose to mark a document and or to transcribe we're using the software scribe which is again i think something that came out of the zuniverse organisation itself here's a typical history sheet and those blue boxes are the marking facilities that our citizen scientists have done they're presented with a document there well after they've completed a tutorial they're presented with documents that they can mark they can they're asked identify the area associated with unit draw a box that box becomes blue identify the area associated with rank that box becomes blue name etc etc and in the second stage transcription is based upon those identified boxes so it says transcribe the highlighted area which is McNeish the name of the person and you can imagine going through the whole document being presented with this highlighted and in many cases expandable zoomable area this is the sort of data you can get from the back room in the zuniverse which is telling you what is happening when now in terms of what might be happening there's certainly things that we would like to get from this project but also we're very much aware that we only get things if we give things and what we're going to be giving is access to the data when that data is complete but importantly what we're going to be doing as an almost necessity is responding to the users who would say to us why aren't you collecting information on x x could be anything tattoos type of wound any of those things that we are not necessarily interested in so that the whole project becomes dynamic and the citizen scientists become genuine scientists they're becoming scientists in as much as they're asking the questions when they're being informed by the data that they're seeing we're providing access to the data they're providing access to their time and ultimately we're hoping that crowdsourcing is going to give us everything that we need and everything that other people might want to extract from these resources that we are making available to them so there's a whole bunch of things that we could be looking at one of the things that we've looked at by linking to birth deaths and marriages data is looking at the prevalence of suicide in New Zealand returning soldiers and although we haven't really written this up properly it's interesting to guess or maybe you can guess before I get pulled off the stage we have Vietnam vets New Zealand vets sorry Korean vets World War two vets and New Zealand World War one veterans who had the highest propensity to commit suicide World War one New Zealand returning soldiers why was that we don't know the answer but we know the answer in terms of the numbers many people when I presented this have said Vietnam the Americans give very poor support Vietnam is very low down the list of suicide rates for returning vets compared to World War one and World War two and part of it was if you had post-traumatic stress if you weren't shot oops if you weren't shot as a deserter you certainly weren't supported when you came back so thank you very much for listening I'll be around if you want to ask any particular questions there's a huge amount of stuff that we can do zuniverse.org measuring the anzacs go in there register and thank you for your time thank you Liz