Become a Creator today!Start creating today - Share your story with the world!
Start for free
00:00:00
00:00:01
Computational Curious with Dr. Isaac Ullah - Ep 130
225 Plays5 years ago
Curious about computational archaeology? Wonder how it’s different from digital archaeology? Us too! Dr. Isaac Ullah from San Diego State University tells us all about it and how he’s using open source software to do computational archaeology and reveal secrets about the history of the world.
Links
- Dr. Ullah’s site on GitHub: http://isaacullah.github.io/about/
Contact
- Chris Webster
- Twitter: @archeowebby
- Email: chris@archaeologypodcastnetwork.com
- Paul Zimmerman
- Twitter: @lugal
- Email: paul@lugal.com
Affiliates
Recommended
Transcript
Introduction and Sponsorship
00:00:00
Speaker
We're excited to announce that our very own podcasting platform, Zencaster, has become a new sponsor to the show. Check out the podcast discount link in our show notes and stay tuned for why we love using Zen for the podcast. You're listening to the Archaeology Podcast Network.
Episode Introduction
00:00:19
Speaker
Hello and welcome to the Archaeotech Podcast, episode 130. I'm your host, Chris Webster, with my co-host, Paul Zimmerman. Today we talk computational archaeology with Dr. Isaac Eula. Let's get to it.
Meet Dr. Isaac Eula
00:00:32
Speaker
Dr. Isaac Eula is an associate professor of anthropology at San Diego State University. He studies the long-term effects of human land use decisions and the bidirectional feedbacks between humans and environments. He is particularly interested in early farming and pastoralism and how these first food-producing socio-natural systems evolved over time.
00:00:51
Speaker
He's a computational archeologist and uses a combination of traditional archeological and geoarchaeological approaches in connection with GIS, computational modeling, and computer simulation to investigate the dynamics of these systems. His work is largely based in complex adaptive systems theory perspective.
00:01:08
Speaker
All right, welcome back to the podcast,
Impact of Current Events on Archaeology
00:01:10
Speaker
everybody. Paul, how's it going? Doing OK. It's a little crazy here in New York lately with all the rioting, but I'm safe and healthy, I think. So how are you doing, Chris? Yeah, I mean, I thought we'd be mentioning coronavirus every episode this year, but now we've got new fun things to talk about, and that's rioting. So it's June 2nd as we're recording this. I'm always mentioning the date this spring leading into summer because
00:01:33
Speaker
things in the world change so quickly these days. I want people to know what's going on in the world when we're recording this because we usually release about a week and a half post recording. So yeah, crazy times, but you know, that gives us time to do some interviews. We've got a lot of people accessible to us because they're either working from home because of the virus or other reasons.
Computational Archaeology Explained
00:01:51
Speaker
And that gives us a lot of good chances to talk to people.
00:01:54
Speaker
in this crazy time where most people are going to be starting field work. I think a lot of field work has just been canceled. So we might have some, some really good stuff coming up this summer as a by-product of that. So there's always a silver lining, but our guest today, as I mentioned in the introduction is Isaac Yula. Isaac, welcome to the show. Thank you, Chris. Good to be here. Hey, no problem. So thanks for coming on. Why don't you just fill in our listeners? What is your research about? What are we talking about? I mean, I read your bio, but what is your research that we're going to talk about today?
00:02:20
Speaker
Yeah, so I call myself a computational archaeologist, and I also do regular archaeology. I'm a geoarchaeologist and a specialist in landscape use, land use. But what I do a lot of is the use of computers and archaeology, and particularly what I focus on is
00:02:36
Speaker
simulation modeling and computation, which is I think kind of an interesting subset of the way computers are being used, have been used in archaeology and a little bit different from the general sort of digital archaeology that I think you guys talked quite a bit about to great effect on this particular show. So I'm hoping today to talk to you guys just about this slightly different approach to doing archaeology and using computers.
00:03:05
Speaker
So where do you typically get your datasets for this stuff? Are you out collecting field data or using published datasets from other researchers?
00:03:12
Speaker
Yeah, so it's a combination of all of that and kind of the interesting thing and a little bit about what's different about simulation modeling and archaeology is that you don't necessarily need to have traditional data sets the way I think most archaeologists think of them. You know, they think about tab delimited files, you know, numbers and spreadsheets and doing statistics and that kind of stuff.
Modeling Human Behavior
00:03:35
Speaker
can use that, but we use it to get inspired. So what we're doing typically is creating models from first principles, meaning we take what we think we know about the past and about human behavior. And we create these simulation models so that digital agents are programmed with
00:03:57
Speaker
Logic that derives from the theory of human behavior and we put them in settings and scenarios that are inspired by archaeological cases and then we actually let them play out making decisions using logic that we pre programmed.
00:04:12
Speaker
Acting on information in an environment that has been sort of parameterized from archaeological or ethnographic case studies and we sort of watch how these agents interact with themselves with their environment and see the behavior sort of unfolding in computational time, which is a sped up very simplified
00:04:31
Speaker
version of reality or real time. So we basically simplify the world, we put it into a computer, and we let it run so that we can see the behavior that would make an archaeological record. And then we go back to the real data sets and we compare.
00:04:47
Speaker
So that's interesting. You distinguish between computational archaeology and digital archaeology. Are you equating then what you're doing with modeling as basically computational archaeology? Or is there some broader meaning of one or the other that I might not be aware of and maybe our listeners would like more explanation of?
00:05:06
Speaker
I'm glad you asked that. I've actually thought quite a bit about this over the years. I have a sort of small blog on my website. I made a post several months ago, basically trying to disentangle those terms. If you do a Google search for digital archaeology, you'll find a lot of stuff on several pages of unique hits.
00:05:26
Speaker
You do a Google search for computational archaeology, you basically get one page. My blog post has moved up and the ranks obviously when I wrote it, it wasn't there. So there was basically five hits that you might get back when I wrote it. And I think the confusion comes from the use of computers, you know.
00:05:45
Speaker
So, everyone who's doing digital archaeology, everyone who's doing computational archaeology, we're all using computers and we're doing it in the realm of archaeological research.
Simulating Archaeological Hypotheses
00:05:56
Speaker
And I think there's certainly an overlap between them, but the way I distinguish them is that I see digital archaeology, if you need to put a boundary around it, is in sort of like the digital curation of archaeological data.
00:06:08
Speaker
And that can include 3D scans, that can include base GIS work, that can include like LiDAR survey, aerial photography, photogrammetry, all of the really fun stuff. And I do a lot of that stuff in my research as well. And then I see computational archaeology as the analytical part of that.
00:06:26
Speaker
as in we are either interrogating archeological data with advanced statistical methods that require computation. So I'm thinking specifically about machine learning, artificial intelligence, that kind of stuff, deep learning on massive big data sets, that kind of thing. And then I'm thinking about simulation modeling and manipulation and other kinds of modeling, including things like GIS modeling, predictive modeling,
00:06:52
Speaker
and basic algebraic regression models and that kind of stuff. I put that more in the realm of computational archaeology and I sort of separate the two because I think you kind of have to specialize a little bit. You definitely will do a bit of both if you're a digital archaeologist or if you're a computational archaeologist but there is a specialty that's sort of emerging in simulation modeling especially I think.
00:07:15
Speaker
Yeah, I think that's a good distinction to make too, because I've been a big proponent of what I've just called digital archaeology. And to be honest, what that really amounts to is just not using paper to record archaeological sites in the field sometimes. And that's what we've called digital archaeology. And I think that's a good distinction because some of the things that Paul and I have talked about in the past were the fact that
00:07:39
Speaker
At what point are we going to lose the word digital on archaeology? And it's just implied that archaeology is inherently digital in what we do. We dig in the dirt, we do things, we walk in the desert, we do all kinds of stuff. But all of that has some sort of digital component or tool set to it that we're using. And more and more often, as we're talking on this podcast,
00:07:59
Speaker
people are using computational methodology like you're talking about here to also just be part of their regular routine, not some one-off that a grad student is doing because they have to find original research.
00:08:11
Speaker
Yeah, I agree completely. And I'm fully on board with digital archaeology and digital data recording methods. I was an early adopter, you know, we were in the early like, I think it was 2009, even with with early iPads and file maker pro in the middle of Jordan, you know, trying to do that kind of stuff.
00:08:30
Speaker
So I'm fully in on that and I think one begets the other. So I really feel like the advances in digital archaeology have started to enable advances in computational archaeology.
Teaching Computational Archaeology
00:08:40
Speaker
I do feel like digital archaeology is more popularized or more ingrained, I think, in archaeological research generally. It's sort of ahead of the curve in terms of becoming normal.
00:08:51
Speaker
computational archaeology is getting there. I personally have witnessed the growth of it, so I started my PhD program at ASU in 2005. I had sort of self-taught myself GIS in my master's program, and I worked with my advisor at ASU, Michael Barton. He was really one of the
00:09:11
Speaker
three, maybe four archaeologists who really knew what simulation modeling was at the time. And I didn't even know what I was getting into when I showed up and I was in the lab and he put me on a computer that was loaded with GRASS GIS and we were doing bash programming in Linux back then. So there was all kinds of stuff that was new to me and he just sort of said he had gotten this grant, an NSF grant for the
00:09:39
Speaker
a couple of human environmental systems grants. And it was about, can we do this? Can we actually simulate human societies? And he sort of sat me down at the computer and said, this is what we're going to do. And I had no idea what we were in for. And we built our first model just sort of from the ground up with no archetype to go off of. And of course, we made tons of mistakes. And I think we could have
Getting Started with Simulation Modeling
00:10:01
Speaker
done a better job doing what we know now.
00:10:03
Speaker
But it was really, really interesting and we learned a lot, but it was a real small group of people back then and it's slowly gotten bigger and bigger and I think it's taking off and becoming more ingrained in archaeological research.
00:10:17
Speaker
Okay, that's interesting to me. You jumped in with both feet because you already had, I guess, one foot in the water and I'm probably mixing it in metaphors, but you're already into the GIS. Because so many of the, if we're going to use your heuristic of digital versus computational, so many of the digital methods are visual, are things like
00:10:39
Speaker
you know, 15 years ago would have been digital photography, nowadays it's LiDAR. These are visual things that really grab, you know, they're easy for archaeologists who tend to be very visual learners anyhow to understand and to gravitate toward. How would an interested student nowadays, you know, in 2020, they wanted to get into computational archaeology? How would they go about doing that? What programs are there? What kinds of data sets are available for them to use?
00:11:04
Speaker
What kind of tools are there out there that they might be able to play with in the same way that we all back in the day played with grass or QGIS on our own? Yes, that's a great question. In fact, this is something I'm struggling with as an educator right now. I've got a handful of grad students and a bunch of undergraduate students and I'm slowly introducing to them. I have actually now put a class through the curriculum on computational archaeology at San Diego State, which is pretty exciting for me.
00:11:32
Speaker
So, what I would say is that it depends a little bit on your previous experience. If coding is scary to you, and I fully understand that it can be for somebody who's not particularly computer-efficient auto person, then my suggestion is to start fairly simple and the classic piece of software that most people, I would say sort of the gateway drug,
00:11:55
Speaker
to computational archaeology is NetLogo.
Integrating GIS with Archaeology
00:11:59
Speaker
And it's specifically designed for making these agent-based models. It's pretty simple. You can code in it, but you don't have to code in it. There's hundreds of models that are just available.
00:12:11
Speaker
If you download the NetLogo software, they just have a library of models built in. You can just click through them. And also, there's a bunch available online. And I've been making some available through my website because they actually have a web player for NetLogo models. So if you'd like, I can give you a couple links.
00:12:30
Speaker
check them out and just sort of play around. And it's pretty straightforward. There are little digital graphical interfaces. You slide sliders to change the parameters. You can check check boxes. And then when the model runs, there's a little window and you can see the agents moving around, the environment changing, and then there's little live graphs that update so you can sort of monitor what's happening to your agents. And we have little models like that are just straightforward, like optimal foraging prey choice type models. I have a grad student right now who's sort of
00:12:59
Speaker
been just sort of got into it through that way and she's really started and I just started out doing most of the actual programming but she's taking over from that and she's changing one of these optimal forging models to talk about human caused extinction at the end of the Pleistocene just trying to see if different hunting preferences
00:13:16
Speaker
could drive prey species you know large prey species extinct or not and so that's really kind of an interesting way to start in computation archaeology
Advocacy for Open Source Tools
00:13:24
Speaker
because then you get into the what if scenarios and that's really what drives a lot of the research is well what if i move this slider over and i and i make i don't know giant buffalo
00:13:33
Speaker
you know, the handling times go way down. Does that induce the hunters to start going for these potentially large animals? And what would cause handling times to go down? Would it be a change in technology and increase in the efficiency of hunting weapons, that kind of stuff, right? So you can play around in these models. And I think that kind of hooks people into it. And as soon as you get hooked on the sort of simple, we sometimes call these toy models in computational archaeology, fairly simple ones.
00:13:58
Speaker
Then you start going down the road, you can learn a little bit of easy coding, the NetLogo coding language is pretty straightforward. I personally prefer programming Python and there are a few programs that you can use programming in Python.
00:14:14
Speaker
There's a couple of agent-based simulation models that even have graphical programming, where you can drag little widgets and connect them, the ins and outs of them. So there's a couple of fairly easy ways, I think, for people to get into this. And really, it's one of these things where I think you got to do it and see what it does. And I see in my students that as soon as they start to do it, things start to click. But beforehand, it seems a little strange, a little daunting.
Open Science and Transparency
00:14:39
Speaker
Okay. Well, I think, man, I had so many questions. I know I just kind of trying to wrap my head around all of it. So you mentioned in your, uh, in your bio that you're interested in, I'm just going to read directly here farming and pastoralism, um, and how these first food producing socio natural, uh, systems evolved over time. Yeah.
00:15:01
Speaker
First off, explain to our listeners, socio-natural systems. I think I can piece out what you mean by that, but I want to know how you're defining those terms. And then also just right after that, what kind of data are you bringing in? And then
00:15:17
Speaker
into these computational models in order to extract those. I'm curious more from a user standpoint, which I want to talk about later in the show, but what kind of things could people be thinking about from a data collection standpoint to be answering these kind of questions? But maybe start by explaining the process there. Yeah, so socio-natural systems, I mean, it is a bit of a jargon word, and I understand that.
00:15:39
Speaker
I use it simply because, you know, granting agencies, they like to use those kinds of terms. But I think it's pretty simple. Everyone has sort of an intuitive sense that human beings are not separate from the natural world. They're not separate from their environments. And, you know, there's a growing understanding of the nuance of how that interaction has changed over time.
00:16:01
Speaker
And I think, you know, if we can talk about all kinds of debates, you know, the Anthropocene debate, for example, but essentially we recognize that humans can be and are probably are big drivers of change in the Earth system, you know, the
00:16:18
Speaker
The weather system, the climate system, the biological systems of plants and animals, the water system, even in some cases the geologic system for surface flow and erosion and deposition of sediments. And so what we mean when we say socio-natural, it means that human social systems, human societies,
00:16:39
Speaker
are in fact embedded in natural systems, have impacts on those systems, and in fact those systems then feedback and impact human systems. So it's a bidirectional set of feedbacks. You can't take human societies isolated from natural things like climate and erosion and biological change. They all interact together and we have to study them together.
00:17:04
Speaker
So that's my explanation of socio-natural. The kinds of data that we collect, as I mentioned, we still do field work. We always do field work as a component of our research. And I'm a trained geoarchaeologist. I have a background also in the geosciences. So a lot of my focus has been on erosion rates and how human impacts to vegetation over time have impacted or changed
00:17:29
Speaker
erosion and deposition rates in parts of the world. So I do a lot of landscape scale survey and then we do a lot of soil coring and we do a lot of looking for eco facts and other proxy records for looking at past environments. So a lot of environmental archaeology, but we've also surveyed for land use patterns. That's a big, big source of our data just to get a sense of where people were and what they were doing.
00:17:57
Speaker
in different parts of the landscape at different points in time. I do a lot less excavation these days than I think most archaeologists might do, simply because I don't not think excavation data is unworthy. It's just that it gets really, really detailed and takes a lot of time to run excavations. And I just don't have that space in my research profile at this particular moment. And I trust my colleagues to do that kind of work. And I definitely use some detailed information
00:18:24
Speaker
So for example, you know, I did a lot of work early on, especially in the Neolithic in the Near Eastern Jordan. And I have done excavations. I worked with a team at the University of Toronto who's doing a lot of work there. We know now through their excavations, the kinds of activities that people were doing in the Neolithic, you know, what kind of farming, what kind of techniques.
00:18:44
Speaker
Generally speaking, how many people lived
Conclusion and Future Topics
00:18:46
Speaker
in these sites, you know, what their households were kind of like. So that information is available. And I pair that with the landscape scale information to try and understand from a quantitative perspective, what the kinds of what broad scale land use patterns might have been, what the effects might have been. And then we use the models to test those ideas to see, well, what if people were tenuring land in this way, right, passing it from
00:19:12
Speaker
generation to generation? Or what if it was relinquished when someone dies and it became a free grab again? What would that impact would have? So I think those are the kinds of information I'm using. Other people focusing on, let's say, hunter-gatherer societies would look at different pieces of information. But it's definitely, you can take fairly traditional archaeological data and bring it into these models. It all depends on how you design your model, the kinds of particular data that you need. Sure.
00:19:41
Speaker
All right. Well, that is a good spot to take a break and chew on all this information. And then we'll come back and continue this discussion with Isaac Yula back in a second. Chris Webster here for the Archeology Podcast Network. We strive for high quality interviews and content so you can find information on any topic in archeology from around the world. One way we do that is by recording interviews with our hosts and guests located in many parts of the world all at once. We do that through the use of Zencaster. That's Z-E-N-C-A-S-T-R.
00:20:09
Speaker
Zencaster allows us to record high-quality audio with no stress on the guest. Just send them a link to click on and that's it. Zencaster does the rest. They even do automatic transcriptions. Check out the link in the show notes for 30% off your first three months or go to zencastr.com and use the code ARCHEOTECH. That's A-R-C-H-A-E-O-T-E-C-H.
00:20:32
Speaker
Looking to expand your knowledge of x-rays and imaging in the archaeology field? Then check out An Introduction to Paleoradiography, a short online course offering professional training for archaeologists and affiliated disciplines. Created by archaeologist, radiographer, and lecturer James Elliott, the content of this course is based upon his research and teaching experience in higher education. It is approved by the Chartered Institute for Archaeologists as four hours of training. That's in the UK for those of you that don't know.
00:20:57
Speaker
So, don't miss out on this exciting opportunity for professional and personal development. For more information on pricing and core structure, visit paleoimaging.com. That's P-A-L-E-O, imaging.com. And look for the link in the show notes to this episode.
00:21:13
Speaker
Hi, welcome back to The Architect Podcast, episode 130. Today, we're talking with Dr. Isaac Ulla about his use of computational archaeology and what that means and various kinds of tools and things that he's doing. Now, you were talking in the last segment about social-natural systems and using the words modeling and testing, and I think it's in that last little bit.
00:21:32
Speaker
that the magic happens here. But I was wondering, what is special about computational archaeology as a general broad umbrella methodology that makes it so appealing to you? And I would guess you would probably argue useful for others as well.
00:21:50
Speaker
So what I would say to that is that when we do traditional archaeology, and I'm including digital recordation methods in that because, you know, we're just recording the archaeological record. What we're doing is recording.
00:22:05
Speaker
a fraction of the small amount of evidence that happened to get recorded and not deleted in the sedimentological record of a place that is just one place among many places and of a time that's just one time among many times. And so the story that we're recording is a fragmented snapshot of just one possible way that things could have played out. And just because it played out that way,
00:22:35
Speaker
in reality doesn't mean if you rewound the clock and start it again at any particular point in time, any particular place, that it would have played out the same way. And the reason is that there is a lot of stochasticity, a lot of randomness, and a lot of chance in the way that I believe the world works.
00:22:55
Speaker
And we have these issues called equifinality and multifinality, meaning that different processes can lead to the same result, that's equifinality, or the same process can diverge into widely different results if you play it over, and that's a multifinality. So that's a part of something called complexity science that really grounds the theory, I think, behind a lot of what we do in computational archaeology.
00:23:19
Speaker
And what we can do in the computer is to program a real simplified version of a reality of a case of a place of a time of a set of things that people were doing way they made decisions, and we can run through that scenario.
00:23:36
Speaker
hundreds of times allowing that randomness to happen. And what we notice is that depending on the way the system is designed or programmed and how the agents interact, some systems lead to a lot of multifinality, a lot of diversion, a lot of different possible results, and some systems don't. They tend to kind of converge on fairly stable solutions.
00:23:57
Speaker
But it's really hard to know which systems are going to do which until you do this simulation approach. So I think what happens in archaeological research a lot is that people get really invested in a case because archaeological research is hard and it takes a lot of time and money and effort and human hours to process all that data.
00:24:19
Speaker
we can tell stories and archaeologists love to talk and we love to tell stories about our valley or our town or our site that we excavated and we get really tied up in the story that we believe we can tell about what happened there and we want to connect that to a broader story of humanity and how
00:24:41
Speaker
human cultures and societies change over time and so we're in this sort of limbo land where we've got one tiny snapshot of one way that it could have turned out and we're trying to extrapolate from that you know how does human society work how do cultures change
00:24:56
Speaker
how does the world change in relation to humans interacting with their environments? And for me, computational archaeology and simulation science especially was a huge mind opener that we can take those little snapshots and we can get inspired by them, but we need to study all the possibilities in this sort of broader base that we can do in the computer that we simply can't do in reality. We can't run
00:25:22
Speaker
a hundred different experimental archaeology, you know, real life simulations of people farming, you know, over a hundred years. It's just not possible to do. But we can do that on the computer and we can see which systems tend to become stable and if it's stable in a good way, a way that seems to be sustainable.
00:25:39
Speaker
or if it leads to unsustainability, or if it's one of these things where it's like all kinds of possibilities. People can be fine, populations can go through the roof, or people can crash out, you know, expanding out all of the fertility in the soil, for example, after a couple hundred years. And we say, well, that's a real unstable space, and maybe we don't want to engineer future societies or, you know, to be in that space. Maybe we can use that study derived from the past to inform our present and future decisions sort of moving forward.
00:26:09
Speaker
Wow. So you're using it or you see it. I don't want to use the word utopian, but that's what I'm going to use. You see it is very idealistic. That's the word I was looking for. It can be used not just to better understand the past, but also to better model the future or propose models for the future.
00:26:28
Speaker
Yeah, you know, my research team has published a couple of papers on that, you know, like using the past to inform the present and to build a better future. And I'm involved in some of these sort of high-ish level think tank type organizations that are trying to do that. And it's not just archaeologists, it's modelers from across many disciplines who are trying to use
00:26:50
Speaker
our leverage modeling to be the tool that we can use to help engineer a better future and i think it's important that we use archaeology properly i think it's definitely there are grounds where we could misuse it and abuse it in this realm and it is a little optimistic it's a lot of work i'm still not 100 sure
00:27:09
Speaker
how we're going to go about this. I think obviously you all probably know there's a disconnect between scientists and politicians these days that we need to, I think, bridge that gap for our information and our methods to actually be making some impacts. That's not a trivial thing at all. But I am optimistic that we can at least learn about choices that we're making now
00:27:37
Speaker
through this method and using the lens of archaeology and history to contextualize the way people behave and the ramifications of those behaviors over long time spans. Okay. Well, we got about, I don't know, eight, nine minutes left in this segment. And one of the things I like to do with this podcast is teach people about things they could be doing on their own research projects or their own sites or whatever they're doing.
00:28:02
Speaker
One of the issues I've noticed with let's just call it computational archaeology, because we've established that on this talk on this podcast, is a lot of times we talk to let's say grad students, usually university setting people about these types of things. Because let's be honest, I'm in cultural resource management archaeology up here in Nevada, so CRM. And a lot of times
00:28:24
Speaker
You know, unless it's based in GIS, a lot of times we just don't have the time to do something else to the data and come up with fun things to do about it. It's done in our off time to produce a report for a conference or something like that. So I'm really interested in
00:28:39
Speaker
trying to create, let's say, more accessible ways for people to get into some of this stuff because it seems like every time you talk to somebody about some sort of simulation or some sort of computational thing, it's like, oh, I've got to go learn Python or I've got to learn this or I've got to learn that.
00:28:57
Speaker
Are we starting to get to the point where we can develop out of the box tools that we can drop a data set into and say, I want to know this? Are we getting to that point at all, where we don't have to be this other type of specialist to really understand what we're doing? I understand the need to really understand the nuts and bolts of how this works, but for this to gain widespread adoption, it's going to have to be out of the box at some point.
00:29:20
Speaker
Yeah, I think you're right about that. I think you're definitely right about that. So there are much more complex tools and simpler tools in agent-based modeling. But I'd like to talk about probably something that's maybe a little more accessible to people who are used to GIS. And I'm a big GIS person I mentioned before. I taught myself on ArcGIS 8 or whatever it was back in 2002.
00:29:43
Speaker
And there was nobody in the department in Toronto that knew how to do GIS. I had to go over to geography to learn anything. And so I've been doing it a long time and I switched completely to open source tools. So I personally use GRASS and QGIS exclusively these days. But I teach a class in GIS and I decided that I wanted to merge GIS and computational archaeology together because I think
00:30:07
Speaker
Most students come and they want to learn GIS and they can get a lot of access to resources to learn GIS. And I think we need to integrate GIS and computational archaeology better. And so I started to do that. I have a series of tutorials I put up on my YouTube channel.
00:30:22
Speaker
You can just search my name on YouTube on how to do some computational modeling techniques in GIS. And particularly what I focus on, I found a good sort of hook is predictive modeling and something, especially for CRM folks, I think that's really useful. And there's two sort of ways to go about this. There's sort of deductive approaches and there's inductive approaches.
00:30:41
Speaker
So the inductive approaches are probably the simplest. You have a site database, which you can get, you know, maybe your firm has one, maybe you go down in California, we have the information centers that have all the CRM, you know, database for all the sites that are excavated. And you can start to plug those in.
00:30:58
Speaker
And I show you how you can basically query them to get a fairly simple statistical understanding of the spread of site locations in relationship to other natural features in a landscape. And I show you how to derive some of those things like stream networks and ridge tops and we can do least cost paths and all kinds of stuff like that.
00:31:19
Speaker
And then I show a fairly simplified way to use map algebra, literally just the calculator Boolean logic, which is like if then statements and then a simple averaging regression to come up with a fairly actually in the end, depending on how many types of data and how good the data are nuanced.
00:31:38
Speaker
probabilistic map of where people were interested in locating sites. And so that's a good way to do it. The other way, again, is you can flip it and go from first principles and say, I think people like to put sites in this time period near water because. And so you can just build in your theoretical understanding of human site location behavior. And you can build these models.
00:32:01
Speaker
And so these aren't as dynamic as the simulation models, but you can start to change the parameters pretty simply, you know, say, okay, it's less than a kilometer from a stream. Okay, what if it's two kilometers from a stream, right? And you can start to see how the perception of a landscape and the way people will utilize landscapes can change depending on these things.
00:32:19
Speaker
And I've seen over the past couple of years as I've started to implement this with my class, the students are really getting into this. And we get a lot of students at SDSU who are going to go on and CRM. And simulation right now, it's not for them, but predictive modeling is something that they can sort of hook into.
00:32:39
Speaker
And if they get the logic of predictive modeling, then they intuitively are going to understand what simulation is all about. And eventually, if they want to pursue it, they can gain that sort of more specialized knowledge and skills to do some of the more technically challenging, maybe theoretically a little bit more sophisticated kinds of modeling. But it's not orders of magnitude different, I think, at that point. OK.
00:33:05
Speaker
Actually, I want to run back to something that you said in there, kind of in passing, but I suspect that it underlies a lot of your approach to it. Now, most of the guests that we have on here talk about different open source tools, and we talk about them ourselves quite a bit. We're big proponents of open source software. The only non-open source things that pop up a lot in our conversation are things like Photoshop or ArcGIS or Metashape.
00:33:29
Speaker
But you'd mentioned both qGIS and grassGIS a couple of times and you have on your blog different grassGIS modules, which just really makes me very happy to see because that's one of the things that a lot of users of open source tools
00:33:47
Speaker
seem to fail to do, which is contribute back into the open source ecosystem. You have a number of ones here for catchment, viewshed, a number of things that we've all seen before. Modules for Grass 7. Now, Grass is certainly not the most friendly of open source tools. It's getting more friendly, I would say. It's definitely getting more friendly.
00:34:07
Speaker
I did a big chunk of my dissertation and it ended. And I wrote in latex, which is like all these wonderful open source tools from the old school of open source, which was, you know, beat you overhead with it.
00:34:23
Speaker
Anyhow, could you maybe want to round out this discussion a little bit and talk about some of the different kinds of open source tools that you use, initiatives that you support, or approaches to teaching your students using open source tools? Because I think that that's an important thing to bring into the conversation.
00:34:39
Speaker
Yeah, so years ago, I went fully open source. In fact, I'm talking to you right now through a Linux laptop. I don't use even my operating system, right? It's fully open source. I try my very best to limit everything I do to open source software and I do try to contribute and I encourage my students to do that as well, whether it's just reporting, you know, making bug reports, feature enhancement requests, or if they get to the stage where you can do some coding, that's awesome.
00:35:04
Speaker
So I use, like I said, grass and KGIS. Those are like the core, I would say, of my workflow. Python, in terms of programming for me, is a Swiss Army knife, the glue that ties everything together. So all my programming, those modules you mentioned that I programmed, are all programmed in Python.
00:35:21
Speaker
And pretty much every open source project these days has a Python API to tie into it. So for me, that's the lingua franca that you can move from project to project with. Now, not all projects are like that. Some of them you need to know C or whatever it is. And I'm not that good of a programmer personally.
00:35:40
Speaker
So I have actually a blog entry on my website where I go through all the software tools. There's probably close to 200 software tools open source on there. So for specifics, everything from managing your photography workflow, instead of using Photoshop or Lightroom, you can use Darkroom.
00:35:58
Speaker
or raw therapy. You can use Geeky to view your photos. There's all kinds of stuff for digital content management to download your photos from your camera, from your drone, from all this kind of stuff. I use OpenDroneMap to do my 3D renderings from the drone flyovers. I render movies. I do everything like LibreOffice and
00:36:21
Speaker
text editors and that kind of stuff. So basically everything I do is open source, including my data collection. Now we're using the open data kit on tablets and we're processing all of our sort of field data collection through there now. So basically everything is open source. I think it's super important because we need to be transparent all the way down to our source code. If I make a mistake in my program, I want people to find that. I don't want people to blindly
00:36:46
Speaker
use my code behind some paywall, and maybe I'll make some money off of it, but the science is going to be wrong. And for me, it's more important that somebody says, hey, man, you made a mistake. You didn't carry the one or whatever it is. And your program is not doing the right thing. And then we correct that. And for me, that's the heart of open source. It's open science. It's making science actually transparent and better.
00:37:11
Speaker
That's a great answer. Yeah, indeed. So we're pretty much at the end, Isaac. Is there anything else that you would want to tell somebody who wants to get into computational archaeology or somebody who's, I don't know, computational curious? Yeah, computational curious. I'm going to steal that. I like that a lot. Yeah. What I would say is we get a lot of students who are a little bit afraid of computers. I think a lot of folks coming up in these younger generations are
00:37:39
Speaker
They know how to use devices, but they don't really grasp how they work. So I want to just say it's not that frightening. It's actually not that complex when you get to it. It's just logic. If you can make a logical argument when you're talking with a friend, you can program.
00:37:58
Speaker
And there are programming languages that are a lot more technical and then there are some that are much more human friendly. So you can start to learn something like R, you can learn Python, Ruby, you know, whatever it is, just start to learn something. Don't be afraid of it. Don't be afraid of using math and seeing numbers. There's nothing to be scared of.
00:38:19
Speaker
I wasn't a huge fan of math as a student. I struggled through calculus in college. I won't lie. But it turns out when you actually have an interesting question and you've got the data and you know that there's a method out there that can help you answer the question, math becomes a lot less scary when you're actually using it for something that's interesting.
00:38:41
Speaker
So don't be afraid to try and use it. That's what I would say. Just sort of open up to those possibilities and those skills. Even if you don't go on in archaeology, if you're an undergraduate student and you don't know you want to go on in archaeology, those skills, if you learn them, they're going to be important in whatever career you go down.
00:39:01
Speaker
And you're going to have a lot more fun applying them as a student in archaeology than you are in economics, right? Or something like that. So I personally think archaeology is the perfect realm for students to get really interested in these methods and to apply them. Because it's fun. It really is fun. It gets your curiosity flowing. I think we have an ability to do that that some disciplines, personally, I don't think do.
00:39:28
Speaker
Nice. All right. Well, thanks a lot, Isaac. Man, it would be great to have you on to talk more specifics about other projects and things like that. I love having these overview sort of conversations, you know, to talk about a technology, but really diving into some of the nuts and bolts of something where, you know, data were collected. You had these questions, you use these things to answer those questions. Then we come back and this is the, uh, this is the result. So maybe we can bring it back on and talk more details. Yeah, I'd be happy to just invite me.
00:39:56
Speaker
Hey, no problem. Open door. All right, Paul, anything else to wrap this up? No, no, that was great. Thanks so much for sharing your view on computers in archaeology. I think that, you know, I learned some things. I've been afraid of computational methodologies in general, even though I'm not one bit remotely afraid of computers. And so I think I'm going to try to take some of what you were saying at the end there to heart and actually start playing with some of these tools and see if I can broaden my horizons a little.
00:40:27
Speaker
Awesome. I'm really happy to hear that. Thank you guys for having me on. I've had a lot of fun today. Thanks for coming. All right. Thank you. Paul and I will be back in just a second to wrap up this episode back in a minute.
00:40:41
Speaker
You may have heard my pitch from membership. It's a great idea and really helps out. However, you can also support us by picking up a fun t-shirt, sticker, or something from a large selection of items from our tea public store. Head over to arcpodnet.com slash shop for a link. That's arcpodnet.com slash shop to pick up some fun swag and support the show.
00:41:00
Speaker
All right. Welcome back to the Architect Podcast, episode 130. We're going to wrap up this discussion that we just had with Dr. Isaac Iula. Paul, I am noticing a theme that came up with this episode where I mentioned that the difference between digital archaeology and computational archaeology, we asked him that question, but then also recognizing what you and I have talked about before about
00:41:24
Speaker
really starting to take the word digital off of archaeology because all archaeology really should and does have some sort of digital component, even if you're just talking about, you know, using a GPS. I mean, that brings a digital sort of aspect. And digital really just means,
00:41:39
Speaker
You know, electronics, it means computers, it means ones and zeros. So what does that mean? In 2020, that's literally everything. And I think it's more accurate. Yeah, I think it's more accurate with somebody like Isaac talking about computational archaeology. What are you doing?
00:41:55
Speaker
with this data that is born digital now and created this way. And I'm glad he brought that up too. It's better that people are collecting data using quote, digital archeology, where people think of tablets and stuff like that, because it allows you to take that born digital data and do other things to it. But I think that's, it's an interesting contrast, which is the point I'm trying to bring here to where we, when I first started this podcast, well, I didn't start the podcast when colleagues of mine started the podcast. And then I took it over.
00:42:20
Speaker
To me, digital archaeology really did mean let's talk tablets, let's talk methodology, let's talk how we're collecting data in the field. And while I know about things like some of the methods in computational archaeology and modeling and things like that, it wasn't even the first thing in my mind. And I think that's an interesting distinction that he brought up. Yeah, I think it's really useful for us. I think that it actually helps clarify, like you were saying, it helps clarify
00:42:43
Speaker
why we want to drop that word in digital archaeology. Ostensibly, it means a methodology or methodologies, a broad set of how you go about your work. But at this point, 40 years ago,
00:43:00
Speaker
Maybe typing out your report on a word processor would have been the vanguard of doing something digitally, but now everybody just does it by default. At this point, so many of our digital tools, digital photography in particular, but definitely a lot of databases, writing as I've just said, illustration, a lot of things are digital without even us thinking about them being digital.
00:43:24
Speaker
As a methodology, it's almost as silly as saying, yeah, I'm doing pedestrian survey by walking, using my legs. Yeah, totally. It doesn't add anything to the conversation. So computational archaeology, again, that is also a broad umbrella term for sets of methodologies and ways of approaching the data.
00:43:49
Speaker
But I think that he's right, that's where the future is going to be for those of us who are doing archaeology digitally. It's not going to be so much in caring about the nuts and bolts. I don't mean that in a bad way, to the point that it's trite way of how we collect and preserve our data.
00:44:11
Speaker
It's going to be in what then we do with it. And that's where the interesting stuff is going to happen. And that's why, at the end there, when I was thanking him, I've been afraid, frankly, of a lot of computational methods. I'm not afraid of math. I'm certainly not afraid of computers. But oh, agent-based modeling. I have no idea where to start with that.
00:44:34
Speaker
Well, now I've got a few things that I might go play with and try out and see what happens next. But I think that that's a good distinction. And maybe we ought to keep that in mind ourselves on this podcast going forward, which isn't to say that if somebody wants to talk about their excellent LiDAR survey, I won't be totally thrilled to hear about it.
00:44:54
Speaker
The really specialness of it is becoming a little diluted and that's a good thing and we have to move that specialness somewhere else.
00:45:06
Speaker
Yeah. And that's a good point because I've been, again, I know things, you know things, but like you said, some of these methods, they're just so far, for me, they're so far beyond some of my skillsets and what I actually have. I mean, I know a little about coding. I know enough to say that
00:45:25
Speaker
you know, I can understand a little bit about things. But if you asked me to actually use some sort of coding language, really any, you know, to do one thing, I mean, I'd have to look up how to do it. I don't just know how to do it, right? But that being said, I, it really comes down to like, I was asking him like, you know, we need an out of box solution for some of this stuff. We need somebody to say, listen, this is the kind of information you're trying to get an answer to.
00:45:48
Speaker
Well, in that case, here's the data you need to collect. Here's the way you need to collect it. Here's what it needs to look like. Here's how it needs to be formatted. Here's the number of data points you're going to need. That's also important to know. And then how do you put all that together? Well, you can hopefully drop it into this box and spit magic out the other side. But it does help to understand
00:46:06
Speaker
all of this as well because then you understand how to fix it. But I guess I quit owning a car, you know, 20 years ago, even 10 years ago, maybe, but 20 years ago, 30 years ago. I mean, I grew up with my dad just working on our junk car every day so he could get to work the next day.
00:46:22
Speaker
That's just that's just how that works, you know, but he knew how to do that just because he was born in the 50s You know, I mean he knows he knows how to work on cars They were basically relatively straightforward once you understood the components of an internal combustion engine You could get out there somewhat easily diagnose the problem. Maybe not easily, but you could diagnose the problem and even fix it yourself I mean the fact that we have those what were they like that the Haynes manuals or something like that for? Yeah, those are the best
00:46:49
Speaker
The Haynes and the Chiltons, but I like the Haynes ones. I have owned so many of those over the years. Yeah, but do they make a Chilton for a Prius? Do they even do that? Because people aren't going to get in and tinker with that as much as they did those older cars. And I think that's where we're going with archeology and some of these computational methods. They're so complex for some people only because it's really outside of your wheelhouse and you just can't conceive of the individual components that go to make up this thing.
00:47:19
Speaker
that I don't want it to get to the point where it's, you know, it's a very small specialized subset of people. I think, I think an out of box solution is, is something that someone should be developing. And I'm sure some of these do have something like that. Like there's a lot of tools in GIS itself, a complicated machine, but there are, there are out of the box solutions within GIS. If once you know that you can say, well, let me plug these things in here and do this analysis. That's why we have GIS, but you still need a lot of knowledge of GIS.
00:47:45
Speaker
Well, I'm just going to push that analogy that you were making with the older cars just a little bit further here. There's been a lot of lamentation amongst car geeks, and I have, at various times in my life, been one, about how new cars are so hard to work on. You can't do anything unless you hook it up to the computer, which is entirely true. But it also, I think, misses the point for the vast majority of people, which it's not just that you're being locked out of making modifications or doing things. It's that you don't have to.
00:48:14
Speaker
Long gone are the days where you had to tinker with your carburetor because he was running a little rich this week. That is a good thing. Nobody really lost anything. Yes, if I really want to get a classic car and tinker with the freaking carburetor jets, I can do that. I'm a weirdo. Most people want to get to the store and back without having to do that. And so that's what we have to get to with our various tools. And I do think that, you know,
00:48:41
Speaker
When it comes to closed source, that's one of the things that always gets pushed by the big software companies. It's like, oh, you don't have to think about it. It just does its thing. I mean, Apple, it just works, right? And then the other side of that is the equivalent of people who like tinkering with the jets in their carburetor.
00:48:59
Speaker
But it seems to me that there's an emerging property, and he hinted to this too with GrassGIS, which has been around for decades, of these open source tools getting friendlier, getting smarter about how most people use them most of the time.
00:49:15
Speaker
so that you don't actually have to go and tinker with everything. And that gets more than to what you want, which is the out of the box solution, something that is well tested and vetted and understood from the point of view of the engineers and the software engineers and the archaeologists that have put together these tools and the models.
00:49:32
Speaker
but can then be used in a very simple way. And I don't mean simple in a bad way. I mean, simple as an easy, uh, by somebody who doesn't want to have to deal with all that crap, but just wants a good sound answer to a question that they have. Right. Right. Indeed. So I don't know. I think we're going in a good direction. I think, uh, I think it's all getting much more exciting every year we talk about this, but, uh,
00:50:00
Speaker
What I'd like to hear from people listening to this is exactly what I mentioned at the end there. We're starting to come around to where we've talked about a lot of different techniques. And a lot of times we do talk about actual application of those techniques, but I think for people to really understand how they can use this stuff. And Isaac did a fantastic job of explaining some of this and how he's using it and how he's used it in the past.
00:50:24
Speaker
But it'd be nice to hear more examples, more examples of people that said, I have a question. Here's how I answered it using computational archaeology and understanding what the nuts and bolts of that was. If you're listening to this and you're in that situation and you see some seemingly simple method, even to you, because you're knee deep in it and you understand it.
00:50:43
Speaker
other people really don't understand how that works and they don't understand how it can benefit what's possibly a data set that they already have and i'll bring up serum again there's a lot of data collected with serum and it's possible that there are some computational methods out there that can be used on existing data sets just by going click click click oh man look what i just learned right because we don't know
00:51:06
Speaker
as a group, we don't typically know unless we're, you know, really have the time to do the research and read the papers and then try it for ourselves. But it'd be nice to hear some of these things. So other people listening could be like, yeah, that that's something I can use on my project, whether they're an academic grad student or, you know, serum or whatever. So that's my ask for our listeners today. So anything else to add to this, Paul, or are we going to wrap this segment up? Well, I'm being kicked out of the building here, so I think we have to wrap this one up.
00:51:35
Speaker
a good time to do that. So while Paul puts all his things in an office box and is shuttled out to the street. No, no, no, no, no. There's going to be a curfew in a couple hours. So everybody's leaving the building and I was asked quite politely to leave.
00:51:50
Speaker
Oh, that's right. That's right. Yeah, you got to get home. So all right. Well, thanks, Paul. And thanks again to Isaac. And I hope we can bring him on again sometime soon and talk about some of those projects he's working on. So again, contact us if you or comment here, wherever you listen to this podcast. And let's get let's get you on the show and talk about this stuff. Absolutely. Wash your hands.
00:52:17
Speaker
Thanks for listening to the Archaeotech Podcast. Links to items mentioned on the show are in the show notes at www.archpodnet.com slash archaeotech. Contact us at chris at archaeologypodcastnetwork.com and paul at lugall.com. Support the show by becoming a member at archpodnet.com slash members. The music is a song called Off Road and is licensed free from Apple. Thanks for listening.
00:52:42
Speaker
This show is produced and recorded by the Archaeology Podcast Network, Chris Webster and Tristan Boyle in Reno, Nevada at the Reno Collective. This has been a presentation of the Archaeology Podcast Network. Visit us on the web for show notes and other podcasts at www.archpodnet.com. Contact us at chrisatarchaeologypodcastnetwork.com.
00:53:04
Speaker
Thanks again for listening to this episode and for supporting the Archaeology Podcast Network. If you want these shows to keep going, consider becoming a member for just $7.99 US dollars a month. That's cheaper than a venti quad eggnog latte. Go to archpodnet.com slash members for more info.