When: 21 September – 14 November 2021
Duration: 8 weeks
Experience level: All levels welcome! No previous experience required as full training will be provided.
Requirements: Internet connection, and the ability to commit 3-5 hours per week
How to apply: Apply below by 09 September 2021
How do archaeologists discover new sites? And how many sites still remain unidentified? And what would happen if we combined Artificial Intelligence with the brainpower of real people to search for archaeology at a massive scale?
Funded by NESTA’s Collective Intelligence programme, DigVentures is recruiting a team of 100 Pastronauts to join Deep Time – an experimental mission combining archaeology, Artificial Intelligence, and the power of the crowd to identify previously unknown archaeological sites from LiDAR images.
Archaeological sites, historic monuments, and the ruins of ancient settlements are often spread across entire landscapes. Variously discovered by chance, ahead of development, or as part of a research project, each newly discovered site gets added to the UK’s Historic Environment Record (HER).
But what would happen if we take a more co-ordinated approach? Rather than discovering sites as they come, could we use Artificial Intelligence and the power of the crowd to explore archaeology at a massive scale, and to discover sites across an entire landscape? Could such an approach help to better understand and protect the UK’s irreplaceable historic environment? And most importantly of all, can it give local people a bigger voice in the process?
Led by DigVentures and our partner ArchAI, with support from NESTA and Discover Brightwater, this mission will combine archaeology and Artificial Intelligence, as well as the Collective Intelligence of a team of 100 specially-trained Pastronauts.
It’s our most exciting, ambitious, and groundbreaking project so far – and we want you to be part of it! In the immortal words of Doc Brown in Back to the Future, where we’re going, we don’t need roads…
Throughout our Deep Time experiment, we’ll be working side-by-side with ArchAI, an organisation founded by Iris Kramer to teach an Artificial Intelligence (or machine learning system) to identify archaeology from earth observation data such as historic maps, and LiDAR.
We’ve already put ArchAI to the test, and have discovered more than 2,000 potential new archaeological sites across two areas of Northern England as a result. That’s an uplift of up to 800% on the number of sites currently listed in these regions’ Historic Environment Records!
But how accurate are these results? And how do they compare to what real people discover through the power of Collective Intelligence? That’s where YOU come in!
It’s time to put ArchAI through its paces by bringing in our team of Pastronauts who will be trained to identify sites using the same dataset as ArchAI. We’ll then compare our findings and use the results to teach ArchAI, helping it to identify archaeology more accurately. By combining the power of Artificial and Collective Intelligence, our aim is not only to generate a list of previously unknown sites that can be fed back into the Historic Environment Records, but also to generate enhanced knowledge about an entire landscape, and to explore the different ways that real people – including local populations – can be part of the process.
This is a a real opportunity to gain new skills and knowledge in digital archaeology. Our Pastronauts will be trained by specialists in digital archaeology, and learn techniques that are so new they’re not even being taught in universities yet!
Pastronauts will need to complete a short online course comprising three training modules, including how to:
Once training is complete, Pastronauts will be ready to launch into the Deep Time mission, where you will:
Completing this mission will be a collective effort and include some elements of teamwork – it’s digital archaeology, with a sociable twist!
Let’s get this mission started! If you get selected as a Pastronaut, you will be invited to our welcome event. We’ll begin with a live event on Zoom, where you’ll get to meet the project organisers, ask questions, and learn more about what’s in store for you over the next 8 weeks. The event will be recorded, so if you can’t attend you’ll be able to watch the recording.
You’ll have one week to complete your first training module, introducing the basics of collective and artificial intelligence, their application to archaeology, and the big ethical questions involved.
You’ll have one week to complete your second training module, introducing the basics of Geographic Information Systems (GIS) and LiDAR (Light Detection and Ranging) and how they can be used to identify and record and archaeological sites.
You’ll have one week to complete your third training module, introducing the different types of archaeological sites that we’ll be aiming to identify in our experiment. You’ll learn how to tell a hillfort from an enclosure, and spot the difference between a moated manor and an ordinary mound.
The experiment begins! This week, you’ll use your newly acquired knowledge to put yourself to the test. Your aim will be to identify as many sites as you can, as accurately as possible, within a 10km grid.
You’ll have one week to review the discoveries and sites identified by another Pastronaut, while another Pastronaut reviews your discoveries too. It’s teamwork!
This week, we’ll take a break, giving ourselves a chance to catch up on any outstanding results. We’ll also share some of the discoveries we’ve made so far…
Back to the experiment! You’ll have one week to review and validate discoveries made by ArchAI. Who will have been more accurate, you or the AI?
This week, we’ll use our discoveries to teach the AI how to identify sites more accurately. This is where everything starts to come together!
It will take our team a little while to do a bit more number-crunching, but as soon as we’ve pulled everything together, we’ll send you the results – including an overview of all the new sites we’ve discovered, as well as what we’ve learned about the process. In the meantime, we’ll send you a certificate and a goodie bag.
The certificate will confirm your participation in the project, as well as the skills and techniques you’ve learned.
Your goodie bag will be filled with some lovely archaeological surprises to say thank you for your participation!
Oooh now that’s a big question! Once we’ve done our number-crunching, and shared our initial results, it’s possible that we may need to get hands-on and do some archaeological fieldwork to validate some of our discoveries on the ground!
That means there might be opportunities to help us excavate some of the sites that were discovered as part of the Deep Time mission. There may also be opportunities to join further Deep Time missions in the future. We’ll keep you informed…
Click ‘APPLY TO BE A PASTRONAUT’ and complete your application by 09 September 2021. You will then be entered into our list of applicants from which our system will select 100 people and we email you back by 14 September 2021 to let you know if you’ve been selected.
If you are not selected, don’t worry – you will automatically be added to our Deep Time Passenger List, which means you’ll still get to follow the progress of the Deep Time mission and see what our Pastronauts get up to.
No previous experience is necessary. Seriously! This experiment is open to anyone and everyone who is interested in the past, and wants to learn new skills, gain archaeological experience, follow their curiosity, and be part of something big.
The ONLY requirements are that you:
If you don’t meet the requirements (for example, if you can’ commit to 3-5 hours per week, or aren’t sure if you can), you can register as a Passenger instead. This means that you’ll get to follow the progress of the Deep Time mission, and stay updated about the discoveries we make, without having to play an active role. Yay!
We will be using LiDAR images from the Discover Brightwater project area – a landscape partnership scheme covering a 220 km stretch along the River Skerne in County Durham, UK.
The area includes towns like Sedgefield, Newton Aycliffe, and Darlington, as well as many well-known archaeological sites, historic monuments, and heritage attractions, such as Bishop Auckland, Bishop Middleham, Sedgefield Roman settlement, Legs Cross Bolam, and more. But we’ll be looking for sites that have not yet been identified…
Yes, participation is entirely online – including completion of all three training modules, as well as the mission itself.
To take part, you’ll need a device with an internet connection so that you can login to the Deep Time training portal, and the ability to commit the equivalent of one evening per week for five consecutive weeks, starting on 16 August so that you can complete the training, and take part in the mission.
We estimate that you will need the equivalent of one free evening per week, for 8 consecutive weeks. This equates to approximately 3-5 hours per week, or a minimum of 24 hours in total.
Anyone can join the Deep Time mission as a ‘PASSENGER’. This is ideal if you’re interested in the experiment and want to know what it’s all about – but can’t commit the time to fully taking part.
You’ll get regular updates about what the Pastronauts are doing, and a chance to test your intuition at identifying new sites. But you won’t have to do the training, or take part in the final challenge.
This mission isn’t just about finding new sites. It’s also about understanding the human impact of projects like this. As well as testing the ability of collective and artificial intelligence to identify new sites at a massive scale, we also want to test its ability to involve people. Once the mission is complete, we will also look at who applied, who took part, what they learned, how it affected their sense of place, and their understanding of archaeology.
We are therefore asking all of our Pastronauts to complete a survey before and after completing the Deep Time mission.
All personal and identifying data will be anonymised upon receipt, and will be deleted from our systems once evaluation is complete.
Our Pastronauts will be selected to ensure that we have a 50-50 split between those who live within the study area, and those who live outside it. Those who live outside it can be from anywhere in the world.
We also want to maintain an even balance between people who have lots of prior archaeological experience, and those who have little or none.
Therefore, when filling out your background survey, there is no advantage to answering in a particular way – the best thing you can do is be honest!
It’s like Space Camp – for archaeology! Once you’ve been selected as a Pastronaut, you’ll then enter our training camp. You’ll receive a weekly training module that you’ll need to complete before you can take an active part in the mission.
When you complete the mission, you’ll receive a Pastronaut goodie bag, including a certificate to acknowledge your contribution, as well as the new skills you’ve gained in digital archaeology, including GIS, LiDAR, recognising ancient sites, the application of collective and artificial intelligence to archaeology, and your ability to work with others to validate your findings.
We can only take 100 Pastronauts on this first mission. If you’re not selected, you will automatically be added to our ‘Passenger List’ instead.
This means you won’t be part of the mission this time around, but you will still get to follow along with regular updates about the Pastronauts’ progress, and maybe even get a chance to test your skills with a few mini-experiments.
You’ll also be on our reserve list in case anyone drops out, and in the running for any future Deep Time missions.
No, you don’t need previous experience. We’ll teach you everything you need to know to become a fully-fledged Pastronaut (see ‘what are the requirements for being a Pastronaut?’)
We will be using LIDAR images to view our study area to identify shapes and features that represent potential archaeological sites.
Together, we will help the ArchAI system learn to ‘see’ archaeology, and check whether the things it has identified are archaeological.
No. This experiment isn’t just about finding new sites, or testing the accuracy of AI. More than anything, it’s about exploring the ways that people can work alongside AI, and continue to be central in decision-making processes.
As AI increasingly becomes part of our lives, this is an essential question across many sectors. We don’t want people to be left out!
In short, this experiment isn’t just about how accurately our crowd, or ArchAI, is at identifying sites. It’s about the human impact of projects like this. Can collective intelligence and artificial intelligence work together for the benefit of local communities? Does it help more people feel more connected to their local environment? And can it help heritage to have a bigger voice in the planning process?
It’s a big mission, we’ve got a lot to learn, and YOU can be part of that journey.
Our wider aim is to revolutionise the capacity of the historic environment to inform the planning process. When it comes to new developments and huge landscape transformation projects, can artificial intelligence help people have a say in decisions about their local environment that deeply impact their lives? The experiment also positions archaeology as a frontline social concern, equal importance to many of the other important, and more widely recognised, issues Nesta is working to support.
Once the initial phase of this mission is complete, there may be future opportunities to take part in fieldwork as we begin to test our theories ‘on the ground’, as well as to join in on future Deep Time missions.
Sign up now and be one of the first to hear about how to get involved with DV’s archaeology projects – online and in the field!