12 minute read

Well, it was just a matter of time before you saw some natural language model AI output on this blog, wasn’t it? I’ve been playing with ChatGPT since it became available, and I just got my invitation to use Google’s Bard in my inbox yesterday, so I thought it would be fun to compare them. But what subject to use? My previous blog post comparing computational archaeology and digital archaeology has been one of the most read posts I’ve put up, so I figured I better check to see if the AI’s are going to put me out of business!

UPDATE There are some other free-to-use large language models out there, so I decided to add a couple of them into the mix as well: Chat-LLM-streaming and you.com YouChat. I have opted NOT to include Microsoft’s Bing Chat, because they force you to use their Edge browser, which is not available for Linux (which I use exclusively).

Here’s the prompt I used for both all of them:

Write an academic blog post comparing computational archaeology and digital archaeology

Simple, right?! Let’s see how they did.

ChatGPT

ChatGPT provides one draft that you can then modify with follow up commands if you like. Here, I am opting to provide just the initial, unmodified result:

Archaeology has always been a field of study that is both multidisciplinary and constantly evolving. As new technologies are developed, archaeologists have been able to adopt them to enhance the way they study and interpret the past. Two of the most recent developments in this regard are computational archaeology and digital archaeology. In this blog post, we will compare and contrast these two approaches and discuss how they have impacted the field of archaeology.

Computational archaeology is a subfield of archaeology that utilizes computational methods to study and analyze archaeological data. These methods include statistical analysis, computer simulations, and machine learning. Computational archaeology focuses on the quantitative analysis of data and uses mathematical models to test hypotheses about the past. For example, computational archaeologists can use algorithms to analyze patterns in the distribution of artifacts across a site to understand how people used and interacted with that site.

On the other hand, digital archaeology is a subfield of archaeology that focuses on the use of digital technologies in archaeological research. This can include the use of 3D modeling, virtual reality, and GIS (geographic information systems) to create digital reconstructions of archaeological sites and artifacts. Digital archaeology focuses on the visual representation of data and aims to create immersive and interactive experiences for researchers and the public. For example, digital archaeologists can create a virtual reconstruction of a historical site to show how it looked in the past.

While these two subfields share similarities, they have different approaches and goals. Computational archaeology is concerned with the quantitative analysis of data, whereas digital archaeology is focused on creating immersive digital experiences. However, both fields share a common goal of improving our understanding of the past.

One area where these two subfields overlap is in the use of 3D modeling. Computational archaeologists can use 3D modeling to create simulations of ancient structures and landscapes, which can help to test hypotheses about how people interacted with their environment. Digital archaeologists, on the other hand, can use 3D modeling to create immersive digital experiences for the public, allowing them to explore archaeological sites and artifacts in a virtual environment.

Another area of overlap is the use of GIS. Both computational and digital archaeologists use GIS to create maps of archaeological sites and to analyze spatial relationships between artifacts. However, computational archaeologists may use GIS to create models that predict where artifacts might be found, whereas digital archaeologists may use GIS to create interactive maps for the public to explore.

In conclusion, computational archaeology and digital archaeology are two subfields of archaeology that have emerged due to advances in technology. Computational archaeology focuses on the quantitative analysis of data, while digital archaeology is concerned with creating immersive digital experiences. While these two subfields have different approaches and goals, they both aim to improve our understanding of the past. As technology continues to evolve, it is likely that these two subfields will continue to grow and expand, providing archaeologists with new and innovative ways to study the past.

Pretty good, I suppose! A bit dry, and not particularly nuanced, perhaps, but it got the basic descriptions of the two areas reasonably correct. It also identified the potential overlap, and highlighted GIS as an important region of coincidence. Not bad! The ending is a bit lackluster and generic, however, and there isn’t much of a “take away” message. I’ll give it a B+!

Bard

Well, what about Google’s Bard? Bard gives you three drafts by default, but as with ChatGPT, I provide only the first draft here:

Computational Archaeology vs. Digital Archaeology

Archaeology is the study of human activity through the recovery and analysis of material culture and environmental data. It is a multidisciplinary field that draws on a variety of scientific and humanistic methods. In recent years, there has been a growing interest in the use of digital technologies in archaeology. This has led to the development of two distinct subfields: computational archaeology and digital archaeology.

Computational archaeology refers to the use of computers to analyze archaeological data. This can involve a wide range of tasks, such as:

Data processing: Computers can be used to organize, clean, and format archaeological data. This makes it easier to store, retrieve, and analyze the data. Data analysis: Computers can be used to identify patterns and trends in archaeological data. This can help archaeologists to understand the past in new ways. Modeling: Computers can be used to create models of archaeological sites, objects, and landscapes. This can help archaeologists to visualize the past and to test hypotheses about how it was used.

Digital archaeology refers to the use of digital technologies in all aspects of archaeological research, from data collection to dissemination. This can involve a wide range of activities, such as:

Data collection: Digital technologies are used to collect archaeological data in a variety of ways, such as:

  • Using digital cameras to record archaeological sites
  • Using GPS to track the location of artifacts
  • Using drones to create aerial maps of archaeological sites

Data storage: Digital technologies are used to store archaeological data in a variety of ways, such as:

  • Using databases to store artifact data
  • Using GIS to store spatial data
  • Using 3D scanning to store 3D models of artifacts

Data analysis: Digital technologies are used to analyze archaeological data in a variety of ways, such as:

  • Using statistical software to identify patterns in data
  • Using GIS to create spatial models
  • Using 3D modeling software to create 3D reconstructions of archaeological sites

Data dissemination: Digital technologies are used to disseminate archaeological data in a variety of ways, such as:

  • Publishing articles and books online
  • Creating websites and blogs
  • Using social media to share archaeological findings

Computational archaeology and digital archaeology are two complementary subfields of archaeology. Computational archaeology focuses on the use of computers to analyze archaeological data, while digital archaeology focuses on the use of digital technologies in all aspects of archaeological research. Both subfields are essential for the future of archaeology.

First impression: bullet points are boring! I guess it basically covered the topic, but I’d say drier and even less nuanced than ChatGPT’s result. In fact, I’d say it barely returned reasonable coverage of this comparison, and didn’t identify any specific areas of overlap. I’ll give Bard a solid C.

Chat-LLM-streaming

Chat-LLM-streaming is a smaller large language model with an open interface and several choices for underlaying model engines. I used the newest one (selected by default), and here is what it returned:

Computational archaeology and digital archaeology are two closely related fields that have both seen rapid growth in recent years. While there are some similarities between the two, there are also important differences.

Computational archaeology focuses on the use of advanced technology and algorithms to study past human societies and environments. This field is largely focused on using data from different sources, such as satellite imagery, archaeological sites, and other historical records, to build detailed models of the past. By studying these models, researchers can gain new insights into the past, including the locations and types of settlements, the economic activities of people, and the impact of environmental factors on human behavior.

Digital archaeology, on the other hand, is a subfield of archaeology that involves the use of digital tools and techniques to study past human societies and environments. This field includes the use of archaeological data from various sources, such as fieldwork and aerial photography, to create digital reconstructions of the past. By using digital tools, researchers can explore the history of a site in more detail and generate new insights into the culture, beliefs, and activities of past people.

Overall, computational archaeology and digital archaeology are two closely related fields that have both had a profound impact on our understanding of the past. While there are some similarities between the two, there are also important differences that have allowed them to advance in different ways. As the field of archaeology continues to advance, the use of computational and digital tools will continue to play an increasingly important role in our understanding of the past.

As much as I admire this project for trying to keep things as free and open as possible, it’s pretty clear that their model is pretty anemic compared to the bigger ones. This reads like a Wikipedia entry on the subject that has not been updated since 1997. It’s barely informative of the basics at best. I give it a C-, and that’s being generous.

You.com Chat

Now, here’s an interesting new player on the scene. You.com is a relatively new search engine that seeks to be more personalized and less extractive than the big search engines. It has recently released it’s own large language model AI interface. Let’s see how it does!

Computational Archaeology vs. Digital Archaeology: Understanding the Differences and Synergies

Archaeology has undergone a major transformation in the past two decades with the advent of digital technologies. Today, archaeologists have access to a range of digital tools that allow them to do everything from collecting data in the field to analyzing artifacts in the lab. Two terms that are often used to describe these practices are computational archaeology and digital archaeology. Although these terms are often used interchangeably, there are important differences between them that are worth exploring.

Computational archaeology is a subfield of archaeology that focuses on the application of computational methods to archaeological data. This includes everything from statistical analysis to computer simulation. The goal of computational archaeology is to use these methods to gain new insights into archaeological data that would not be possible through traditional methods. For example, a computational archaeologist might use machine learning algorithms to identify patterns in ceramic data that would be difficult to see with the naked eye.

Digital archaeology, on the other hand, refers to the use of digital technologies in all aspects of archaeology, from data collection to interpretation. This includes everything from using drones to create aerial surveys of archaeological sites to using 3D printing to create replicas of artifacts. Digital archaeology is not just about using technology for technology’s sake, rather it is about leveraging digital tools to enhance traditional archaeological methods.

Despite their differences, there is significant overlap between computational archaeology and digital archaeology. In fact, many practitioners in the field use both methods in their work. For example, a digital archaeologist might use computational methods to analyze data collected from a remote sensing survey.

The synergies between computational archaeology and digital archaeology are many. By using digital methods to enhance traditional archaeological practices, archaeologists can collect and analyze more data than ever before. This in turn allows them to gain new insights into past societies and cultures. Computational methods, in particular, can help researchers to identify patterns and trends that may not be visible to the human eye. For example, using machine learning to analyze DNA data from archaeological samples can help researchers to identify population movements and genetic relationships between ancient communities.

In conclusion, while computational archaeology and digital archaeology are different practices with distinct goals, they both play important roles in modern archaeology. By working together, computational and digital archaeologists can push the boundaries of what we know about the past and gain new insights into the cultural, social, and technological evolution of

…of what?!! Don’t leave us hanging like that!!! Just tell us! The evolution of what?!!! Sheesh. Well, up until that point, it was doing pretty well! It even gave us a snazzy title to use. It seems like it wanted to write more, but all of these models are limited in the number of “tokens” (word parts) that they can produce. Clearly it ran out of space. Bummer. Up until then, it was giving ChatGPT a run for it’s money. One thing to note is that it tried to give a concrete example (about ancient DNA analysis), but didn’t cite the source. That’s a no-no, for sure. Because of that, and since it didn’t finish, I can only give it a B-.

Conclusions

So what do we make of all this? Well, I am not fully sure to be honest. I suppose that one take away, at a minimally practical level, is that ChatGPT seems to do a “better” job at writing academic blog posts than does Bard the other models. Neither None of them does did an amazing job, however, and I think my future as a computational archaeology blogger isn’t in much imminent jeopardy from an AI takeover.

That said, the broader discussion around AI in computational archaeology is certainly intriguing. I attended more than a few talks and sessions at the recent CAA 2023 meeting focusing on the use of “AI” techniques. A lot of this isn’t particularly “new,” however, but it’s making a resurgence due to all the recent press about AI. For example, Machine Learning techniques made up the large bulk of these applications in archaeology, and we’ve been using Machine Learning for a quite a while. The tools for Machine Learning (and other “AI” types of data mining) are quite a bit better nowadays, and there are a lot more of them now than there used to be, and our comptuers are more capable of handling them. There is also much more awareness of the potential for these kinds of approaches by the general public and by archaeologists, so these techniques might be experiencing a bit more mainstream acceptance than in the past. That’s all a good thing, in my opinion. If it draws public engagement, new people, and new funding into the discipline, then by all means lets do it!

I suppose, then, that my main take away from this admittedly facile stunt with two of the big natural language AI models is this: branding and public perception actually do matter. If we want to push computational and digital archaeology further, then we might need to get on this current “AI” bandwagon and really start “selling it” to our colleagues and to the public. More engagement with digital and computational tools and approaches can only be a good thing. And if we don’t do it for ourselves, well, then the AI’s are gonna do it for us!

Leave a comment