Login
You're viewing the front-end.social public feed.
  • James BaillieJubalBarca@scholar.social
    Nov 6, 2026, 2:44 PM

    Much of the data is just wrong, clearly misinterpreting the relative independence of different places. Probably really quite a lot of it, since the spot-checks I've been able to do at a glance have found a bunch of issues.

    Here we have Czechoslovakia being treated as part of the USSR after the Prague Spring (no, still in the Warsaw Pact), South Vietnam in 1970 as a direct US territory (nope!), Scotland in 1616 labelled as England, and San Marino apparently losing independence in 1799.

    Terra tempora map showing the USSR in 1971 including Czechoslovakia.
    Terra tempora map showing south Vietnam in 1970 as an American possession.
    Terra tempora map showing the Republic of San Marino as only existing from 1797 to 1799, which would probably be news to the Sammarinese who are decidedly independent as of the present day.
    Terra tempora map showing Britain in 1616 with Scotland marked as part of the Kingdom of England, something it decidedly was not – nor was it the same country at all officially until the Act of Union most of a century later.
    💬 1🔄 10⭐ 3

Replies

  • James BaillieJubalBarca@scholar.social
    Nov 6, 2026, 2:45 PM

    I think this is an inherent issue - I could keep generating casual spot-check problems like that for weeks if I had the time to do so, the scope of human history is large enough that this sort of auto-processing will always produce these levels of errors. You do need historians to do history!

    I also think there are some even more fundamental issues with the data-vis, though - ways that this events-and-civilisations model distorts our understanding of the past more than supporting it.

    💬 1🔄 2⭐ 7
  • James BaillieJubalBarca@scholar.social
    Nov 6, 2026, 2:46 PM

    First, map gaps. Some of these are just nonsensical, like the example below where some parts of the 1616 Holy Roman Empire are apparently non-entities.

    Some are more troubling though. Oceania is a blank space before British colonisation in Australia: the pre-colonial histories of Australia and Aotearoa are relegated to "non-civilisation", erased entirely. Even some known cultures like the Hopewell are absent: the computer is unthinkingly replicating the full colonial model of who gets seen.

    Terra tempora map showing Europe 1616, with the southern Low countries labelled as part of the “Iberian Union” but also with several random holes in the Holy Roman Empire.
    Terra tempora map showing North America, 191CE. The map is empty. This should be the height of the Hopewell culture, which is not deemed sufficient to be included as a “civilisation” here.
    Australia 1799, with the British Colonial Empire occupying a near eastern half of it and the western half and New Zealan simply blank. This erasure of cultures that did not fit a state model was really popular in the colonial era for obvious reasons.
    💬 3🔄 37⭐ 87
  • James BaillieJubalBarca@scholar.social
    Nov 6, 2026, 2:47 PM

    The next problem to consider is what creates a 'top level' unit. "Independence?" What does that even mean? Many historical leaders were concerned with stability and extent of power more than "independence" in a modern-state sense.

    The approach here is totally confused. 1366 Touraine here is a "civilisation" despite being a (misplaced) county: Isaac Komnenos' independent Cyprus in 1188 is Byzantine which would have been news to the Byzantines who had recently lost a fleet trying to recapture it.

    Touraine 1366. It appears to be in the wrong place, but also, it’s a county, why is it labelled as a separate *civilisation*?
    Cyprus 1188 shown as part of the Byzantine Empire. Isaac Komnenos had ruled it independently since 1185.
    💬 1🔄 2⭐ 3
  • James BaillieJubalBarca@scholar.social
    Nov 6, 2026, 2:47 PM

    Even modern political structures don't work well on border maps, as we saw earlier with the Czechoslovakia problem. Premodern ones work far, far worse. English kings held land in France at times *as vassals of the French king with respect to that land*, so is that English land or French? Map-painting also obscures many ways of life: giant blobs for nomadic peoples with labels like "Kipchak confederation" do no justice to the actual ways that people interacted with or controlled territory.

    💬 2🔄 2⭐ 4
  • James BaillieJubalBarca@scholar.social
    Nov 6, 2026, 2:47 PM

    Or take the Mkhargrdzeli family - whose domains are just treated as part of the Kingdom of Georgia, and it's true that they were officials at the Georgian court but they also have inscriptions proclaiming themselves as kings despite being vassals of the Georgian monarchs, because that was a culture & period where kingship and "independence" did not need to be aligned (actually quite common).

    💬 1🔄 0⭐ 3
  • James BaillieJubalBarca@scholar.social
    Nov 6, 2026, 2:48 PM

    So why does all this end up being the case? I think firstly it derives from wanting to flatten history into something that's just too simplified. Categories like civilisations and empires and realms often do a lot of historiographical work, and visualisations like this present them as consistently *the same type* of entity, even if some are city-state collections and not unified polities at all. There is no one model of what a "civilisation" is and political power overlaps a lot in reality.

    💬 1🔄 3⭐ 3
  • James BaillieJubalBarca@scholar.social
    Nov 6, 2026, 2:48 PM

    There's an underlying problem with a data-scraping/AI approach to this though which is that we end up reflecting the data quantity biases in Wikipedia (largely written by western Europeans/N Americans & wildly skewed in its topic matter) and, where AI is involved, derived from general pop historiography because a large language model's weights will not default to more complex representations than the average, and the average by statistical weight is rooted in very old fashioned histories.

    💬 2🔄 3⭐ 4
  • James BaillieJubalBarca@scholar.social
    Nov 6, 2026, 2:48 PM

    I don't think that the author of this tool intended it to be what it is - but what I see in it is mostly a replicator that hardens existing biases in its sources and presents them as data-visualised fact. When it comes to things like erasing the complexity of recent political arrangements that affect people today, or wiping out indigenous histories in colonised parts of the world, then I think there is a real ethical problem with that as a way of approaching how we view and explore the past.

    💬 1🔄 4⭐ 5
  • James BaillieJubalBarca@scholar.social
    Nov 6, 2026, 2:49 PM

    Whilst I'm very far from the end of academia that thinks academia has all the answers, in this case we do have quite a few of them. First, tools like this need to come with a lot more clarity in what they do *and don't* show, making clear their limitations to public users. Engage with scholarship on what you're trying to do: good historical data-vis isn't a few days' project with Claude, it's real work. Finally, talk to a historian, or if you want to do global history, quite a lot of historians.

    💬 1🔄 2⭐ 5
  • James BaillieJubalBarca@scholar.social
    Nov 6, 2026, 2:49 PM

    Historical data isn't like scientific data, it doesn't come polished. Everything is skewed and incomplete, and we need data visualisations that help people understand that rather than pretending a sense of completeness that isn't there. This sort of tool doesn't make those recognitions and doesn't appear to have had any significant spot checks or discussions with experts, and the result is something that is frequently a misleading mess at best and runs into real ethics questions at worst.

    💬 4🔄 23⭐ 44
  • James BaillieJubalBarca@scholar.social
    Nov 6, 2026, 2:52 PM

    This isn't the first data-scrape/AI history attempt I've seen and I'm sure there'll be more. If this sounds negative, it's not to dunk on this guy in particular: rather, it's because I really think it's important for you, dear reader, to carefully think about these problems. We're going to face more of them in public history - and it's going to exacerbate existing problems with how we ensure good quality information gets a hearing. I hope we can learn to do better.

    Here endeth the thread.

    💬 3🔄 8⭐ 3
  • TottsGeri@veganism.social
    Nov 6, 2026, 6:42 PM

    @JubalBarca History grad here. A commendable thread. As your rightly allude to history is a question of interpretation but basic facts are basic facts - the framework of such interpretation.

    Worryingly, this sort of tool is a prescription to what history is for those who can't be bothered to search further, symptomatic of AI culture and reductive degradation.

    Xx

    💬 0🔄 3⭐ 22
  • 💬 0🔄 0⭐ 0
  • 💬 0🔄 1⭐ 1
  • 💬 0🔄 0⭐ 1
  • 💬 1🔄 0⭐ 0
  • 💬 2🔄 0⭐ 0
  • 💬 0🔄 0⭐ 0
  • 💬 0🔄 0⭐ 1
  • 💬 0🔄 0⭐ 0
  • 💬 0🔄 0⭐ 0
  • 💬 0🔄 0⭐ 0
  • Jonathan TJonnyT@mastodon.me.uk
    Nov 6, 2026, 3:55 PM

    @JubalBarca "Historical data isn't like scientific data, it doesn't come polished."

    Former biochemist here: A minor quibble but nor does scientific data. Especially so if it is the life sciences. It is (you would hope) more reproducible but it is not always polished.

    Adding because I wouldn't want people to think that doing the equivalent thing with AI for scientific data would be any less likely to produce a lot of slop. It inevitably would.

    💬 0🔄 7⭐ 43
  • Nicovel0 🍉Nicovel0@mastodon.social
    Nov 7, 2026, 9:39 AM

    @JubalBarca I’m in STEM, and I laughed at that bit. Some scientific data is well organised and annotated, but a huge amount is also an ungodly mess, with no raw source available, no clear ways of reproducing it, and on occasion produced through questionable means. So we’re all in the same boat really.

    💬 1🔄 0⭐ 0
  • James BaillieJubalBarca@scholar.social
    Nov 7, 2026, 10:03 AM

    @Nicovel0 Yeah, polished was a very overblown word choice, mea culpa.

    I think there are some real differences I was trying to get at re nominal reusability: even really very good historian-made data has a level of reader interpretation of sources that are simply not at all designed to provide that data or indeed be "truthful" as we would understand it, which adds some inherent reuse & data modelling problems.

    But I definitely skipped over that too fast and picked the wrong word, my bad!

    💬 0🔄 0⭐ 0
  • Nicovel0 🍉Nicovel0@mastodon.social
    Nov 7, 2026, 10:08 AM

    @JubalBarca oh it was not a criticism at all, it’s always fun to chat with my humanities/social sciences colleagues and get their view on what data they consider robust and why. And it is my strong belief that if more scientists did so they would be better scientists.

    💬 0🔄 0⭐ 0
  • Chu 朱chu@climatejustice.social
    Nov 7, 2026, 12:27 PM

    @JubalBarca

    I would argue that scientific data also comes skewed. I teach communications for scientists and one of the first things I teach is that science is political despite what they try to teach us. The data itself you can argue is apolitical but the decision of what data to collect is a political/social decision. Everything is political and all communication needs to understand the political nature of the ask. A lot fewer COVID deaths would have occurred if we understood that.

    Thank you for this thread!

    💬 0🔄 0⭐ 0
  • 💬 0🔄 0⭐ 0
  • benbenjamineskola@hachyderm.io
    Nov 7, 2026, 10:02 AM

    @JubalBarca this was always my issue with games like Crusader Kings where you’re either a vassal or you’re not, and there’s no model for ‘vassal with respect to this territory but not other territories’. (Brandenburg-Prussia comes to mind as another example and I’m sure there are countless others.)

    But it’s sort of forgivable in a game; they need to balance accuracy with playability/fun. It’s more concerning in something that purports to be factual.

    💬 0🔄 0⭐ 0
  • 💬 1🔄 0⭐ 0
  • James BaillieJubalBarca@scholar.social
    Nov 6, 2026, 8:02 PM

    @erikcats They could at least have ominously labelled the supposedly empty half of Australia Here Be Emus, Beware All Who Enter, or something

    💬 0🔄 0⭐ 2
  • 💬 0🔄 0⭐ 0
  • 💬 0🔄 0⭐ 0