In the future humans have used generational ships to explore the galaxy. It's taken centuries but we have finally found several habitable planets (they are not without their challenges of course).

On one such planet we discover a great machine, essentially a giant cylinder in the ground. The 'walls' are 100 meters thick and packed with machinery and what appear to be control rooms. It is clearly very old, the best guess is 50,000 Earth years but it's impossible to say for sure. The hallways and ramps were made for creatures of roughly similar size to ourselves, with ceilings around 5 meters high. There are some remnants of a civilization on the surface - tunnels, worked stone, channels in straight lines, etc - but nothing that provides any real clues as to who they were.

We conclude that some rooms are control rooms because they have flat sections on the walls and sliding panels with a rippled texture that seems to respond to touch - no buttons or knobs as we know it but they do seem to be some sort of input device.

For the purposes of the story, the ancient aliens wanted future civilizations to have access to this device. The question is, how would they leave instructions for working this incredibly complex machine such that any type of life form would be able to understand it in the far future?

My assumptions about the aliens:

  • Roughly our size based on the Earth-like gravity of the planet, say between 1-3 meters tall.
  • They evolved a way to physically manipulate tools with precision
  • They have senses to detect and interpret light and sound
  • Highly advanced technology with the ability to generate large amounts of energy and the ability to produce any material that we (humans) know of
  • They have explored large portions of the galaxy
  • They were aware of other life forms that evolved on various planets and could identify general common traits among these

As a bit of background, the device is used to fold space, allowing for instantaneous travel. The aliens used a type of Dyson swarm to capture energy and had incredibly advanced methods for storing this energy. The machine is fully intact and has access to stored energy to operate, also intact.

For the purpose of this question please assume that the science behind the generation ships, the space-folding device, habitable planets, and the Dyson swarm are all explained in the story. It is OK if your answer requires additional assumptions about the ancient aliens, their specific traits are not relevant to the story, merely the fact that they existed and left this device for us to find.

Note: I found some similar questions, like How could an ancient race warn the future in a universally understandable way? discussing leaving messages over long periods of time but I don't think it quite matches what I'm looking for. Apologies if it's too similar.

  • 2
    $\begingroup$ If the machine is designed to destroy the universe or just to make a half decent cup of tea, then just a big red button will do, anything else would be show. What does it do? $\endgroup$ Commented Mar 24, 2021 at 22:04
  • $\begingroup$ Do you want instructions to tell us how to operate the one they left, or how to build more ? $\endgroup$ Commented Mar 24, 2021 at 22:06
  • $\begingroup$ How does Ikea do it? Their instruction manuals for putting together their furniture-like products are famous for their complete lack of words. $\endgroup$
    – AlexP
    Commented Mar 24, 2021 at 22:59
  • $\begingroup$ LEGO manage pretty well without words as well. $\endgroup$ Commented Mar 24, 2021 at 23:03
  • 2
    $\begingroup$ Use whatever handwavium allows this machine to still function after 50,000 years. $\endgroup$
    – John
    Commented Mar 25, 2021 at 15:45

7 Answers 7


in Hindu belief, a manifestation of a deity or released soul in bodily form on earth; an incarnate divine teacher

Your machine would pay attention to the beings that entered the control chamber. It would quietly learn how they looked, moved, smelled and sounded. Then it would produce a being in the likeness of those it encountered. It would be different enough not to be weird. Maybe it would make up some purposeful differences. It would still be kind of weird.

The avatar would learn more from the people. It would learn their ways and language and modes of communication. Then once it had learned enough, it would teach.

This would be a fine way to teach aliens if you had AI good enough to learn from the aliens. Also an easy way for an author writing a fiction, because the avatar is a character.

short squib inspired by this scenario https://www.fictionpress.com/s/3355296/1/avatar

  • 1
    $\begingroup$ +1 If the civilization is advanced enough to build a machine capable of surviving for 50,000 years, it can write the code to learn an alien language. This complex was 40 millennia old when we invented agriculture. Either it already has a reasonably powerful AI performing maintenance, or it's been completely destroyed by time. $\endgroup$
    – codeMonkey
    Commented Mar 26, 2021 at 17:48

I an making the assumption that the aliens will communicate via visual and audio media. For example, alphabets, pictures, videos, sound. If they communicate via farts and dance then the same principles hold although it might take humans a little longer to understand instructions.

On that assumption the solution is probably 'via baby-steps'.

Imagine you wanted to explain to humans with no technical background how to use a fairly complicated piece of machinery - lets say a programmable multi-axis milling machine. You wouldn't just hand them a manual and let them go to it. You would start by training them in the use of hand tools, the difference in properties of aluminum, copper and tool-steel etc. Then you would introduce them to manually operated lathes, drill-presses and mills. At that stage you might give them a course on interpretation of technical drawings, expanding to CAD design tools. Finally, once they had all that under their belts, you would demonstrate the use of the fancy automated milling machine before letting them loose on it. I would expect the aliens to follow a similar path. So...

Step 1: a symbolic language based on maths, geometry and physics to teach you their terms for numbers, true/false, greater than/ less than/ left/ right/ up/ down/ hot/cold etc. At the end of this step you will have a simple understanding of a base language.

Step 2: using the symbolic language, expand the language into 'practical' communication via video, audio, showing simple scenes of the aliens communicating with each other. At the end of this step you should have a fairly detailed understanding of their language. Linguists and explorers used similar techniques to learn the languages of other people they met with no common language in the past.

Step 3: technical training and practice. Basically learn the concept of what the machine does, even if you can't understand why it works. And learn by controlling simpler non mission-critical equipment. "stroke this symbol to get water from this aperture", "adjust the weenix until the grobulator bisects the frabulation", "mind-meld with the..." you get my drift I am sure. So the aliens give you 'terminals' or control systems to practice on with feed-back. You need to pass tests before you are allowed to use the next more complicated bit of equipment.

Step 4: Finally: you now get the detailed manual and access to the matter transmitter or whatever. Initially it might have limited power until you show you can use it safely.

The only problem will be if the alien's way of thinking is so different from ours that we cant understand the way they think, or if control of the equipment is too difficult for us to grasp. But if the3y want us to be able to use it, and have had long experience communicating with other alien species, then they are probably capable of 'dumbing-down' the controls and concepts sufficiently for us bipedal apes to grasp.

  • 2
    $\begingroup$ "If they communicate via farts and dance" - ah, I see you've met my daughter. $\endgroup$
    – Vicky
    Commented Mar 25, 2021 at 13:13

Don't underestimate the difference in knowledge required to understand how to work a device, from that required to understand how a device works. Humans - even extremely non-technical ones - interact with a range of complex devices every day without scratching the surface of how they work. Computers and smartphones are the canonical examples, but when's the last time you worked an elevator, or an electric drill, or the traffic lights at a pedestrian crossing? In all of these cases the device has been explicitly designed with the goal of making it usable by people who do not need to fully understand how it works, which means a considerable amount of 'dumbing down' and simplifying the wide range of possible behaviours and making a 'sensible' subset of them available with a set of very simple, very obvious controls.

The challenge, then, is to clearly communicate 1) what the device is supposed to do, 2) what controls are available, and 3) how the controls map to functionality. The best method for all three is demonstration. Just like how many millennials will now instinctively turn to Youtube for a demonstration of how to use their new drill/camera/laser power tool/military hardware, rather than digging into the full user manual, so will your adventurous humans learn a lot more from a holographic projection of an Ancient Alien going through the motions of using the device, than they will from any written manual. Encouraging 'learning by doing' shifts the exercise from one of painstaking research to intrepid discovery. Whether the device is actually safe to experiment with in such a way depends on its design: an elevator is a lot safer to operate in ignorance than a press brake, despite them having broadly the same basic structure!

  • $\begingroup$ This is the answer I was going to add. Simulators could address your last point: From press bake demo, to practicing on a press brake simulator with realistic depictions of damage/harm for grave errors, to a real press brake with instructor standing close by to save the student from any dangerous mistakes, to signed off and ready to operate it unsupervised. $\endgroup$ Commented Mar 26, 2021 at 1:21

Assuming the aliens planned to leave the device behind to be discovered they have any number of options. The key point/question though is how far in advance did they make those plans?. Note - in all cases the stone artifacts etc were probably left simply to attract inquisitive aliens to the burial site and do not serve any other purpose.

1) Did they build the device in the first place with the intention of leaving it behind? If so then the machine will have been build with the best internal 'teaching' subroutine they could devise. The machine will be programmed to project and receive messages in as wide spectrum of frequencies, sounds and tactile sensations as possible and will have an elaborate AI to run it. Their version of SIRI in other words. It will also have multiple types/settings for its control panel options - which BTW may also explain why there are lots of control rooms (e.g press a panel in control room No. 3 and it floods with water and has acoustic controls).

2) If the machine was already built when the decision was made to leave it behind but time was not pressing. They could build a library of written instructions and place it on a suitable asteroid or moon. Say something carved into titanium or carbon plates and left perfectly preserved in vacuum. Then they just leave a map of the solar system inside the device with the orbit of the body concerned clearly marked and a giant sign post on the orbital body itself saying 'here lie the operating instructions. This also has the advantage BTW of confirming that the species which found the device is in fact space fairing and not a locally evolved, more primitive native species. Anyway instructions are in the form of pictograms including detail maps of the device, its control circuits and what the device does. In as many varied ways as the aliens can devise. (I did say library).

3) Lastly if it was buried in a rush. All they can do is place a similar but more compact set of instructions (e.g how to turn on the computer) inside the device and then program the computer to be as flexible as possible and the controls as simple to use as possible in the time available. Then hope of the best.


I know you've already linked "How could an ancient race warn the future in a universally understandable way?" - I'd encourage you to read through the "Expert judgment on markers to deter inadvertent human intrusion into the Waste Isolation Pilot Plant" report.

It makes some very interesting points about super-long-duration structures and documentation - for example, that you don't want it to be scavenged for building materials, and that future civilisations might not read left-to-right or might not know what an arrow is.

Of course, in a way you'll have an easier time if your world already has an explanation for a fully working 50,000 year old machine - it must be self-repairing and nigh-indestructible to have lasted that long, so there's no rush to teach users how to repair it.

I would suggest you probably want not instructions, but an interactive tutorial with gated capabilities. For example, at first there's only a single button, and it gates you to a fixed, safe location 3 miles away when you press it, and gates you back when you press it again.

To enable extra abilities, you have to go through challenges, where each challenge proves you've got what it takes to figure out the abilities for yourself.

Want to gate to a destination based on a map? Gotta prove you can understand a map, by following a map to retrieve a macguffin. Want to gate to somewhere without oxygen? Gotta survive 20 minutes in an airless room to enable that option. Mess with the machine's parts? Gotta show you know about precision machining first.

Some of the challenges could be foundation-building for later challenges: Perhaps you can gate 10 miles if you can answer grade school math problems in the aliens' notation, 100 miles for high school math and 1000 miles for degree-level math. That way, even if initially the future society doesn't know which end of an arrow is which, they'll gradually be taught what they need in order to understand the instructions.

  • $\begingroup$ The report's got good information about making a durable instruction manual, but it's rather less useful about making an understandable one. $\endgroup$
    – Mark
    Commented Apr 2, 2021 at 3:39

Assuming that they have eyes, a pictorial instruction manual might work. Start with pictures of the machine itself so there is a common reference point. And zoom into various rooms so it is clear where they are, then zoom into screens and provide pictures of button sequences with pictures of the out come.

Pictorial background information could also be provided to explain basic concepts like numbers, the relative positions of planets and stars, time intervals and more.

Even better provide a video of the builders actually running through all of the steps necessary in sequence to power up the machine together with the outcome.

  • $\begingroup$ Vision in general is tricky because not every species might see light of the same frequencies. If you look at a human picture in the infrared or ultraviolet spectrum there is not much to see. $\endgroup$
    – quarague
    Commented Mar 25, 2021 at 7:38
  • 1
    $\begingroup$ Initial instructions could be provided in a range of wavelengths and even in relief. Some species might not be able to see the video but given initial hints could probably build something to up shift or downshift the wavelength. $\endgroup$
    – Slarty
    Commented Mar 25, 2021 at 10:21
  • $\begingroup$ @Slarty, for the initial instructions, I'd use an incandescent light source shining through a shadow mask. As long as you're not dealing with a species that sees in the far ultraviolet or beyond, you're covered. $\endgroup$
    – Mark
    Commented Mar 25, 2021 at 22:49
  • $\begingroup$ @quarague That's not necessarily an obstacle, so long as a means of seeing in those frequencies can reasonably be expected to be within the technology of anyone getting there. Eventually someone will try it. It might take a little while, of course. For example we're only just now figuring out how some squid and octopuses communicate, because people have only recently thought about looking at them with polarised light, in spite of cephalopods and polarised light both being known for many years. $\endgroup$
    – Graham
    Commented Mar 26, 2021 at 14:37
  • $\begingroup$ Could even use some initial engraved images to help those using echo location $\endgroup$
    – Slarty
    Commented Mar 26, 2021 at 16:13

Show, Don't Tell

There have been plenty of discussion on this site about the difficulties in leaving warning signs or instructions for future generations. Even assuming that the person who sees your warning or message is the same species as you, when you are talking about thousands of years in the future you just can't account for cultural shifts that could make your warnings convey the wrong information or just be indecipherable. Add in entirely different species with possibly different sensors and manipulators and definitely different cultures, and you pretty much can't expect any written or pictorial message to be able to transmit the information you want. I would say that is especially true for something as complicated as a giant piece of machinery.

Luckily we are not stuck with just single images to convey information for the future. I would suggest that a video showing how the machine is used would be much more helpful for anyone who discovered it. You are still going to be making some assumptions about the sensory input of the people that discover your machine, but there is not much you can do about that and it might even be a good thing. If this ancient race knew about other intelligent species they might have purposefully left their messages in a format that only the "safest" (in their opinion) races would be able to interpret. Sure, they could have included smell-based instructions, but that race of light-blind hyper predators would be more likely to use the machine for war and conquest so better make it harder for them to understand it.

I would suggest having multiple types of recordings based on the kind of species that discovers the machine, and what they are doing with it. The builders would want to have a bit of a bootstrap sequence built into the setup, so that certain actions would unlock new recordings. The simplest would be a recording on a loop that shows how to activate the machine. If someone can follow the steps to do that then you can give them more recordings because you know that you are actually conveying information. Each subsequent set of recordings would have more information on how the machine is operated, as well as some gatekeeping procedure that unlocks the next level of recordings. At a really high level I would expect it to look something like:

How to turn on the machine -> How to use the machine -> How to customize/configure the machine -> How to build your own version of the machine

All of that really depends on what the machine does, but the basic concept should work for just about any functionality. You just need to add more or less steps as appropriate.

There is an added bonus to using these recordings as they can convey a pretty in-depth look at the original species from an anthropological perspective. Anyone who discovers the machine will be able to see what the builders looked like, what kind of design choices they favored (e.g. purely functional vs intricately detailed), what they thought was basic information, and what order they gave information about the machine in. All of that can help give a better picture of their society and possible reasons for leaving the machine behind.

A race of hyper-engineers might have a purely functional design for the machine, with no embellishments or designs, and their boot strap order might be "turn it on, configure it, produce parts, build a second one". Very straightforward with a specific goal in mind. On the other hand an advanced but artistic race maybe have more focus on the architecture of the machine and decorate every surface and their startup sequence would be geared more towards general understanding of the machine without a specific goal for its usage. "Turn it on, make something, change a setting, make something else, repeat." You could still work out how to build a second machine if you wanted to, but it wouldn't be the explicit goal like with the engineers.


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .