Archaeology. Technological advances in mapmaking, map-parsing and surveying (drones) are leading to a golden age in excavation, on a level unseen since the XIX century. They now have software that can literally just sit there spouting suggestions all day: “hey, this pattern around fields in Nolandia-sur-nothing looks a bit too regular, wanna investigate?”
And there is still so much that can be done, as digitalization is solving the “dusty archives” problem and neural networks might soon be able to suggest links between material (“this parchment fragment here might be related to that fragment in a museum on the other side of he world, which they’ve even forgot they had”).
Unfortunately the obstacle is always the same: money. There is little or no money to be made digging up old vases, and often plenty to actually forget about the vases altogether and let this or that new building be constructed on top of them (or destroying them).
What are the chances that we're on the cusp of a golden age of archaeological discovery as robots/drones/etc advance in capabilities and allow us to deeply explore ruins, underground lairs, underwater wrecks, and other remains of past, buried civilizations that previously required human-sized openings / human-supportable environments? Or is it more likely to be continued marginal advancements over time as we keep discovering the next-hardest stuff to discover? (It's been decades since Alvin explored the Titanic, but the technology is getting much better and much more affordable.) Either way, it's an exciting time to be alive.
I remember, e.g, Italy and Brazil had projects more than ten years ago where they used some sort of machine learning to find hidden historical buildings under terrain or jungle by looking at patterns in satellite/aerial image and it was successful in finding archeological sites in both countries.
Also a technology win in that sites that have had digs still revealed new archaeology with lidar mapping (aeroplane and drone).
Presumably we have enough tech now that the Dr who found the 27 new sites can train an ML algorithm to recognise sites and buildings (and do metrological analysis automatically?)? Is there a generalised image analysis system for aerial imagery that catalogues buildings/roads/ruins/foliage/etc.?
Note the juxtaposition of this self congratulation
> Yes, the stories about excavating with tiny brushes and grapefruit knives are true, even though shovels and backhoes would get the job done quicker. The stories about obsessive photography, data recording, sampling, artifact recovery, and general record keeping are also true. The underlying reason for these quirks is that archaeological search and recovery can often simultaneously acquire evidence and destroy its context. That being the case, we are honor bound to document the process well enough to allow some future researcher to use discoveries not just to test our conclusions but also to help answer new questions that have not yet occurred to us. [emphasis mine]
with this observation
> Technological advances have dramatically improved our abilities to detect and recover archaeological information
I fuzzily recall going to an archeology talk soon after PCR came out, and hearing how carefully they washed their recovered pottery shards. It reminded me of early paleontology using dynamite.
We can now do single cell PCR, and single molecule sequencing exists. We can do near-field spectroscopy of nanometer-scale samples. And the microrobots are coming.
Imagine a conversation a half a century from now. "We found a site untouched!" "Untouched? By looters, or by early 21st-century archaeologists?" "Given how much our technology has changed, and how rapidly it's changing now, there's little difference".
I'm not an archeologist but it seems to me that so much of the past has already been lost that every find can be viewed as potentially useful in reconstructing the past - if not now then later after new tools have been developed.
Archaeology is one of those sciences that is desperate need of attention from CS. While there have been many advances, they (generally speaking) are still at least 10 years behind the technology curve and lack the knowledge required to really leverage the advances that have come.
The entire field is ripe for the picking for cross collaboration of people who can bring technical skills to "big data" and other problems.
You can see many other sites that specifically allocate the majority of dig work to the future on the premise that our future technology will better be able to process and preserve our findings.
I suspect that tombs in particular are large tourism draws, or can yield attractive museum exhibits.
And there is still so much that can be done, as digitalization is solving the “dusty archives” problem and neural networks might soon be able to suggest links between material (“this parchment fragment here might be related to that fragment in a museum on the other side of he world, which they’ve even forgot they had”).
Unfortunately the obstacle is always the same: money. There is little or no money to be made digging up old vases, and often plenty to actually forget about the vases altogether and let this or that new building be constructed on top of them (or destroying them).
reply