What Can SV Teach an Evil Sorceress? (abandoned) Original - Fantasy (2024)

Greetings, Bianca. It's Just Write again.

Anyway, you have asked how to defend against cannon. The answer is pretty simple: packed earth fortifications. I don't mean that the entire structure be made of packed earth, merely that the top and front parts that could plausibly be exposed to incoming cannonballs should have a very thick layer of packed earth atop them. The reason this works is that the packed earth absorbs the impact of the incoming projectile just as effectively as a large enough mass of stone, but unlike stoneworks or brick construction it will not shatter if exposed to extreme levels of force. Of course, a person out in the open who takes a cannonball hit to anywhere remotely vital is almost certainly going to die no matter how good their armor is; even if by some miracle their armor isn't breached, the extreme levels of force will transfer through the armor and bludgeon them to death.

You have also requested knowledge of how to build a finery forge, along with knowledge of why a blast furnace doesn't truly decarburize the iron but a finery forge does. The second question is much simpler to answer than the first: basically, in a blast furnace all the carbon-rich fuel and iron ore is mixed together and therefore a lot of that carbon winds up bonded to the iron. This isn't optional, since all that carbon is necessary to react with and remove other impurities in the ore. In a finery forge however, the fuel and the molten pig iron are kept separate, meaning that the process of melting the iron is not in and of itself constantly putting more and more carbon into the mix.

The crucial details of any furnace to refine pig iron are therefore that it keeps the fuel and iron separate, that it can heat the metal enough to actually melt it, that good airflow over the molten metal can be achieved, that the molten metal be agitated such that the entire volume of molten metal be exposed to air, and that it can extract the molten metal when it is done.

Finding details of a finery forge you will be able to construct was surprisingly difficult for such a foundational technology, but I believe I have found plans for a workable design known as the reverberatory furnace. The firebox and the hearth for the pig iron to go in are arranged side by side in this setup, with the flue staying low as it runs over the top of the hearth and then goes up a chimney. Aside from that, the top of the flue is curved and dips down over the hearth, so that the gasses leaving the firebox are forced into contact with the iron for more effective heat transfer; this curvature also serves to reflect the radiant heat from the firebox directly onto the melt. Also the Hearth should have a spigot near the bottom allowing the molten Iron within to be drained out once it's been properly refined.

An important note: some Limestone will need to be added to the Pig Iron you wish to melt, especially if you are also planning on melting down iron scrap for re-use. In combination with the Oxygen from the air, Limestone will react with various non-carbon impurities in the Pig Iron, separating them from the material. These impurities are referred to as slag, and are less dense than molten iron, meaning they will float to the top of the melt. As such, if the melt is drained from the bottom, the slag will come out last and can be separated with relative ease.

As a brief aside, radiant heat is another term for heat carried by light; in a furnace most of this light is outside the range of colors perceptible to the human eye, but plenty of visible light is also involved. Because radiant heat is so important to a reverberatory furnace, the best fuel is the sort of coal or charcoal that burns with a very brightly visible flame.

A set of large bellows will be needed to get the firebox up to a sufficient temperature, and the air coming from them will need to be pre-heated. Fortunately, there's a trick that can allow this to be done without a separate furnace; simply route the air pipes through the chimney so the hot gasses leaving the furnace serve to warm the air on its way to the firebox.

For best effect the pipes in the chimney should either divide into multiple pathways before re-merging, or should bend around inside the chimney a fair bit before exiting. In either case this serves to increase the surface area of pipes exposed to the hot flue gasses, and thus improves the pre-heating efficiency. Under no circ*mstances should the air pipe be routed to the firebox while still inside the flue; that's a recipe for frustration when the pipe inevitably cracks due to uneven thermal expansion. Instead it should enter through one side of the chimney, then exit through another side of the chimney before coming at the firebox from below or to the side.

Coming from the side with a downwards slope at the end seems best, as that avoids the problems of the fuel messing up the pipe and being a major chore to clean.

Anyway, since it's important that fresh oxygen be added to the metal to remove the carbon impurities and much will be consumed by the flames of the firebox (along with the need to agitate the melt), there are two options for getting fresh air to the entire volume of molten iron. First is to have a second air pipe that goes up over the level of the molten Iron, then comes out the bottom of the hearth where the iron is melted through a number of small holes. This serves to blow bubbles through the molten metal if forced with a very large set of bellows, and will do the job of removing the carbon content and agitating the metal very effectively. However, air absolutely MUST continue being pumped through this pipe until all the molten iron has been drained from the furnace, or else it will solidify inside the pipe and be nearly impossible to remove without completely taking apart and rebuilding it. This works almost the same as the Bessemer Process, however it has the notable difference of being much less efficient in terms of fuel; with the Bessemer Process you start with metal that is already molted, while with a Reverberatory Furnace you are letting the Pig Iron cool to solidity before expending a great deal of time and fuel to melt it all over again.

The other option is to just have an opening in the hearth through which workers can stir the molten iron with very long poles, along with a pipe blowing down unburned (but still heated) air from above. While this may seem like the obvious correct choice, I implore you to only consider it if all attempts at forcing air bubbles up through the molten iron fail. The gasses from the furnace will wreak havoc on the bodies of those who work it in this manner, dooming them to drastically shortened lifespans and utterly miserable suffering.

In addition, either method can be used to produce steel with careful control of the timing. This will require very careful timing of how long the molten iron is kept in the furnace for. The goal in this case is to stop the process at just the right point in time when the Iron has slightly more carbon in it than wrought iron, but less than pig iron. It will probably require a great deal of trial and error with each furnace to figure out how much time it takes to make the best steel.

That said, I still think a Bessemer Converter would be superior if it can be constructed. It will massively save on both time and fuel when making steel, and can do so in truly massive quantities limited only by the scale at which it is built. Taking a Bessemer Converter to its absolute simplest extent, it is simply a limestone-lined vat for molten iron that is taller than it is wide, with holes in the bottom for bubbles to come out. The pipe feeding these bubble-producing holes needs to raise up over the highest possible level the molten metal will reach before coming back down in order to avoid molten iron flowing back out the pump and ruining the series of bellows used to pump the air. The opening at the top receives molten iron directly from a Blast Furnace, while the spigot of the Bessemer Converter itself outputs molten Wrought Iron or Steel depending on usage.

To avoid issues with the weight of molten metal pushing air back through the bellows (especially if multiple sets of bellows will be used for larger converters) it may be wise to fit the pipes with ball valves to ensure air only flows one way with ease. A ball valve is a widened section of pipe with a ball in it that firmly plugs the normal diameter pipe when pressed against one end. Protrusions inside the widened pipe section keep the ball from pressing fully against one end of the widened pipe, and also keep it centered relative to the hole. Air flowing from the direction that the ball cannot make contact with will shove the ball against the hole and obstruct its own flow. Air flowing from the direction the ball can make contact with will push the ball off the hole and flow around it largely unimpeded. This is highly useful for connecting multiple sets of bellows to the same air pipe without them all causing problems for each other, allowing the blowing force for all these new iron-melting furnaces to be massively increased.

Anyway, back to the converter itself. Molten Pig Iron goes right in the converter from the Blast Furnace without any time to cool down, saving greatly on the fuel that would be needed to re-melt the Iron. Blowing bubbles through then serves to rapidly remove carbon from the metal, the length of time that would be needed to completely convert a volume of Pig Iron to Wrought Iron in this manner is about one fortieth of the time the world needs to rotate. During this time, the Oxygen stripping Carbon from the iron actually releases more heat to keep it from solidifying, while the limestone lining of the converter causes the phosphorus impurities to float to the top as a stony slag. However, if you manage the timing correctly the Bessemer process can produce not just Wrought Iron, but good high quality steel. This is accomplished by stopping the process while there is still some Carbon left in the iron, but much less than in Pig Iron, and far fewer other impurities as well.

Given that the limestone lining of the Bessemer Converter takes an active role in the process of turning Pig Iron into Steel, it will likely be slowly consumed over time. Given that fact, a Bessemer Converter will need to be re-lined every once in a while.

The timing for this is very tricky; it is easily doable to blow into the iron for too long and wind up with wrought iron instead of steel. Stopping partway through decarburizing the Iron will require much practice and careful paying attention to master. Fortunately for the Bessemer Process the total batch time to convert a load of molten pig iron into good steel is relatively constant.

That said, if you can master the timing needed to convert Pig Iron into steel in either the Reverberatory Furnace or the Bessemer Converter, the reward will be great. Either process if mastered will allow the Ten Nations to produce vast quantities of a metal so durable that even in our world there is no true substitute.

As for a tool to help with the precise timing required to make steel in either a Reverberatory Furnace or a Bessemer Converter, I would suggest a water timer. This is a simple bucket with a panel of clear glass making up one of the sides so that the water level can be seen at a glance; this glass part is then marked at regular intervals. Lastly, a small hole is put in the bottom of the bucket with a stopper fitted to it. To use the water timer, just fill it with water, pull out the stopper, and the water will flow out at a regular rate. Counting the markings above the water line will allow an operator to easily and rapidly tell how much time has passed since the water timer was started, without the sloppiness inherent to human perception of time.

However, if the timing of making proper steel in a Reverberatory Furnace or Bessemer Converter still eludes you after some time, I do know of another method for producing steel on a small scale. Crucible Steel is a type of steel made from a specific ratio of Pig Iron and Wrought Iron, along with a small amount of ground up Limestone to remove any remaining silicate impurities. These ingredients are then sealed inside an airtight clay vessel referred to as the crucible. It is absolutely crucial for this process that no cracks exist in the crucible; it only works because the ingredients inside the crucible are isolated from the outside world in a way that prevents any air or carbon from getting in.

Once the crucible is sealed, it is to be placed inside a Blast Furnace, though perhaps one loaded with only fuel and lacking in Iron Ore. The temperatures inside the Blast Furnace melt the Pig Iron and Wrought Iron inside the crucible, causing them to mix into a single alloy with a carbon ratio in between what was originally put into the crucible. More Pig Iron in the mix means that the resulting steel will be harder and take more force before breaking, but also means it will be more prone to cracking when forced to bend. More Wrought Iron in the mix does the opposite, producing steel that is very resistant to being bent out of shape, but taking somewhat less force to do so.

Either way, once the Blast Furnace containing the crucible is extinguished, it's a fairly simple matter of lifting the Crucible out and breaking it open to get at the ingot of steel inside. This ingot will either need to be heated and worked into shape in a forge, or might be melted and cast into a mold with a properly designed furnace. Crucible steel is a very useful material for making weapons and tools, but the difficulty of scaling up its production process excludes it from the large-scale applications that become possible if you master the timing on the Reverberatory Furnace with the bubbles. For example, it would be rather difficult to get together enough crucible steel to make a cannon.

As for a method of finding good Iron Ores, I have to ask if you've heard of Bog Iron? The basic idea is that certain waterways contain dissolved iron oxides. When these waters reach relatively stagnant wetlands, the actions of tiny life and some interesting chemical reactions serve to concentrate the dissolved iron into solid lumps of ore that can be spotted from above the surface of the water due to a silvery sheen also produced by the same tiny life that concentrates the ore. While Bog Iron is a pretty good source of small amounts of ore that even replenishes itself every few tens of years, it does have limits.

Fortunately, if you can find a bog with iron, you can use it to find where that bog gets its iron. The waterways feeding the bog will only contain dissolved iron ore if they pass underground before they reach it. Therefore, if you follow those waterways upstream until you find a place where the stream is coming from underground, that place stands a decently good chance of being a good place to mine for Iron, if perhaps a bit wet.

That said, if you want to be absolutely sure that you're digging in a good place, you can check if the local rocks tug on a lodestone. That's something only high-grade iron ores do. That said, given the relative rarity of lodestones this might not be an option in all cases.

Actually... I think I might know of a way to make an artificial lodestone. This will be difficult, but I think it should be possible. Remember that trick with the staff and the lodestone making an electric current? That will be needed. First some groundwork for how this works. Materials containing high quantities of Iron and a couple other metals that are currently irrelevant are always inherently magnetic. However, their magnetic fields vary greatly in orientation on an incredibly tiny scale so it all basically cancels out by default. To make an artificial lodestone requires straightening out all these tiny domains of magnetization so they're all pointing the same way, then intensifying them.

It goes like this. Get a lodestone and mount it on a turntable inside a coil of insulated copper wire; as it spins it will produce a linearly flowing electric current, or simply a difference in electric potential between the ends of the wires as long as the lodestone is spinning. Route this electric current through another coil of wire, and that second coil will produce a much stronger magnetic field than the lodestone itself.

To make an artificial lodestone, there are two separate steps requiring this generator of electric current. First, the chunk of iron that you intend to magnetize should be shaped as needed ahead of time; for most magnets a bar or bent shape is highly effective, but for some other specialized tasks other shapes might be desired. Then you heat the magnet to be until it glows red hot, and while it's cooling put it inside the coil of wire with an electric current running through it. For a bent magnet, it's probably wise to put it in sideways, since having one end be 'north' and the other be 'south' is more convenient than having both ends be 'north' and the middle of the bend be 'south'.

The heating of the iron loosens up the magnetic domains inside it so they're more prone to changing directions, and the fact that it's already inside a strong magnetic field ensures that the piece of metal will have them all pointing the same direction once it cools down. The piece of iron is now a weak artificial lodestone; to strengthen it, the generator needs to be hooked up differently. Instead of sending its electric current through a magnetic field coil, attach each end of the wire to one end of the artificial lodestone and crank the generator as hard as you possibly can. The harder it gets cranked, the more electric current goes through the lodestone, and the stronger its magnetic field gets.

Once this is done, you should have a reasonably powerful magnet that you can do as you like with; pure iron isn't an optimal material for this on account of a tendency to demagnetize after a while, but it should work well enough. There are various materials that can be added to iron to reduce its tendency to demagnetize, but I'm not quite sure how to explain identifying or purifying them, so this will have to do for now.

Anyway, hopefully the improvements to iron production from blast furnaces and reverberatory furnaces will make Iron cannon more feasible to construct. If you can melt down Pig Iron and decarburize it properly, it can probably be simply cast into cannon in a similar manner to what is done for bronze guns. A bit of a tip for longer-lasting iron cannon is to have the front end be a bit wider than the part immediately behind it, as well as smoothly rounded; this prevents the formation of cracks from the stresses of firing. and ensures that the cannon will last longer.

I will now be explaining the proper use of cannon aboard ships, with no regard for what material those cannon are made from. First, for as long as sails are the predominant way of propelling a ship through the water, one of the best places to put cannon is in a special structure known as the broadside gun port. This is a hatch in the side of a ship through which a cannon can be aimed, and which can be closed while the cannon is being reloaded. Further, to either side of the cannon is a set of strong ropes with pulleys, affixed to both the structure of the gun port and the gun carriage for the cannon.

These ropes have two main functions: first, they restrict the cannon's ability to launch itself backwards when firing, reducing the likelihood of it injuring the crew. Second, by varying the length of the rope bundles on either side by using the pulley system, the cannon can be aimed from side to side. As for the carriage that the cannon is attached to, it should have four relatively small wheels to keep it level relative to the deck. Further, the protrusions of metal on either side of the cannon by which it is mounted to the gun carriage should be in line with the path traveled by the projectile, so as to reduce bucking.

With regards to ignition, cannon on ships absolutely need the use of a flintlock ignition mechanism for accurate fire. The ship is constantly moving and rolling, meaning that it requires rather precise timing to fire the cannon with enough accuracy to hit reliably at any significant distance. Flintlocks allow much more rapid ignition than touching a burning length of cord to a hole, and present much less of a fire hazard as well, which is very important on a ship where all the wood has been coated in highly combustible pitch to waterproof it.

The basic elements of a flintlock have already been largely explained, but I do believe there was one part left out: a small amount of fine-grained bursting dust needs to be added to the touch-hole and a small pan into which the flint and steel will cast their sparks. This is needed to ensure proper ignition, or else the relatively small amount of sparks produced by a flintlock will be insufficient to reliably set off the main charge of bursting dust inside a cannon.

With regards to that charge, it often isn't needed to put as much bursting dust in a shipboard cannon as it is one on land. Beyond a few hundred paces of range naval gunnery becomes incredibly unreliable barring complicated equipment you won't be able to produce for some time, and wooden ships require much less force to damage than the heavy-duty fortifications a land-based cannon may be called to demolish.

Reducing the charge also lets a shipboard cannon be built a bit lighter than a land-bound one, which can be important. Furthermore, most injuries to a ship's crew from cannonfire are actually from the splinters produced when a cannonball smashes through a ship's structure, rather than the cannonball itself. Since a cannonball that impacts too fast can often produce less splinters than one that impacts slower, reducing the bursting dust charge can counter-intuitively increase the carnage caused by firing cannons at a ship.

Anyway, here are some specialized projectiles that are particularly useful when shooting at ships. First is chain shot: this is simply two cannonballs connected together by a length of chain. When fired, the whole assembly spins through the air at its target with great rapidity. The nature of the projectile means it is extremely effective at damaging a ship's rigging, cutting massive gashed in sails, severing ropes, and sawing down masts.

Next is heated shot, which is exactly what it sounds like: a cannonball heated red hot before being rammed into a cannon and fired at a ship. This requires some extra wadding between the ball and the bursting dust to avoid it setting off the charge prematurely, but a successful hit that lodges in an enemy ship can be devastating as it can easily cause the vessel to ignite. For reasons of the massive fire hazard inherent to having a furnace for making heated shot aboard a ship, this is better employed by land-based cannon that happen to be shooting at ships to defend a harbor or similar.

Canister shot is a thin container full of small metal balls that gets shoved down a cannon's muzzle in place of a cannonball. When fired this unleashes a roughly conical blast of high-velocity projectiles that are devastating against unprotected people, making it very effective at clearing all the crew off a ship's deck. For this reason canister shot is also useful on land when attacking armies that aren't in a fortified position.

I also have some tips with regards to the use of ships themselves. For example, did you know it's actually possible to use sails to drive a ship upwind? I don't mean sailing directly into the wind, that's obviously not how it works. The correct method is known as tacking upwind, and it works like this:

First, point the ship at an angle somewhere between pointing straight into the wind and pointing sideways to it. Then angle the sails to be roughly in line with the hull. The basic idea of doing this is that the wind will push the ship sideways relative to the direction it's blowing, and then the hull of the ship will convert that sideways motion into motion upwind by pushing against the water. Of course, if your goal is to sail directly into the wind this isn't quite sufficient, so a ship tacking upwind will need to repeatedly turn to either side of the wind in in order to sail in a zig-zag pattern that approximates a course directly into the wind. Of course, this requires a mast and rigging design which allow the sails to be rotated with relative ease.

Anyway, at the moment we have some valid reason to believe that the sea bordering the Ten Nations is mostly enclosed by land, though we could easily be wrong. However, if your ships should ever reach the open ocean here's an interesting trick that can make cross-oceanic travel much faster and more predictable. Basically, when a sufficiently large body of water is left to sit, the rotation of the planet tends to produce massive currents that move in a predictable fashion; sailing with these currents makes a ship go faster, trying to sail against them does the opposite. Near the eastern coast of a continent, these gyre currents are typically pointed roughly towards the poles. Near the western coast, the opposite is true. In the open ocean, the gyre currents tend to go west near the equator, and east closer to the poles.

On to a completely different topic, I will now be speaking about matters of medicine. In particular, I will start with a somewhat bizarre occurrence known as the placebo effect. As I'm quite sure you are aware, people are very capable of tricking themselves into believing all sorts of things. In particular, just them taking something they believe to be a medicine which will cure their ailments can convince them that they are healed and their body will listen, at least to a degree. This even works if the person knows about the placebo effect, and deliberately performs a known to be useless act to feel better knowing it will work anyway. Minor aches and pains can be treated this way and even serious conditions can be staved off for a short time, but it obviously has limits. The effect works in reverse too; simply convincing someone they have been poisoned can cause all sorts of amusing symptoms to appear for a while.

The reason I'm telling you this isn't because the placebo effect is particularly useful, but instead because it can get in the way of healers improving their craft. When testing a new medicine to see if it works, the patient could show signs of improvement on the short term due to the placebo effect, only to suddenly take a turn for the worse when the problem goes beyond the point where the body tricking itself can keep the problem under control. Even worse, a 'medicine' could turn out to be doing permanent damage to the patient, with symptoms not showing up until it is too late because the placebo effect has been covering them up!

It's this reason why I present a methodology for developing and testing the effectiveness of new medicines. The general idea is that when testing a new medicine, a number of patients are randomly assigned either the medicine being evaluated or a fake designed to do nothing when applied but very closely resemble the real medicine. Neither the patients being evaluated or the people directly administering the medicine are supposed to know whether the medicine being administered is the real stuff or the fake. Only the person running the study is aware of who is getting which medicine, and is kept regularly informed about how all the patients are faring. This is known as a double-blind clinical trial, and this methodology is critical for keeping wishful thinking and the placebo effect from making a mess of new medicines being developed.

Generally, a double-blind clinical trial proceeds in a few crucial stages, with each one getting more people involved and taking longer to complete. Stage one is generally a few dozen patients at once, and is simply concerned with making sure the medicine is safe to use at all. Figuring this out is generally pretty quick, and if the medicine is safe the clinical trial moves on to stage 2.

Stage 2 is where it starts becoming relevant that the medicine actually works. Stage 2 usually involves a few hundred patients being evaluated to see if they show signs of improvement from the new medicine being tested; it's also a good time to figure out what sort of dose is best in terms of effectiveness. A medicine that proves it can actually work to at least some degree in stage 2 then moves on to stage 3.

Stage three is the final stage of a clinical trial before adoption; this is an evaluation to see if the medicine can actually treat the ailments it's supposed to help with in real-world conditions. A stage 3 trial generally involves up to a few thousand people being looked at, and can last several months to a year depending on if it's for a medicine intended as either a preventative or to treat a chronic condition. If a medicine is shown to be reasonably effective over long periods with minimal side effects, it can be considered to have passed phase 3.

A medicine that has passed phase 3 of a clinical trial can be added to the standard medical practice used by healers. However, those using it should still be observed for a few years after initial adoption to see if the medicine has any unexpected long term effects.

In particular, there is one medicine I think would probably pass these clinical trials, and would greatly improve the lot of the sick in the Ten Nations. It is known to me as Bald's Salve, and it is used to treat bacterial infections; to the best of my knowledge it is extremely effective at killing harmful bacteria and causes minimal harm to people. The following ingredients and tools are needed:
-Equal amounts of garlic root and onion leaves
-Bovine bile salts of equal quantity to the garlic
-Wine, again of equal quantity to the garlic
-A quantity of redewed water
-A bronze pot
-Mortar & Pestle
-A fine cloth for straining liquid through

Anyway, the garlic and onion leaves are finely chopped, then ground with the mortar and pestle. Then the wine and ground garlic and onion mix are all added to the bronze pot. The bile salts and redewed water are added to the pot at this time. Then everything is covered up and left to sit for nine days, chilled but not to the point of freezing. When the nine day waiting period is up, the salve is strained through the fine cloth to remove any particulates, and is then ready for topical application to infected wounds and sores.

Anyway, the only ingredient I'm not sure you know about is the wine. It's an alcoholic beverage made by fermenting the juice of a fruit known as grape. It seems somewhat likely that you would have access to it through trade at the very least, but I've never heard it mentioned. You could maybe substitute for the wine using some quantity of redewed alcohol and vinegar, but there are biological molecules in the fruit that I'm not sure if they're needed for the salve to work at full effectiveness. If possible I'd recommend trading for grapes and trying to cultivate them using fruit walls, but I hope that you can get a working salve out of this regardless.

Anyway, I've now got some advice regarding the proper production of paper. From wood, not rags. The first thing needed is wood, obviously; this should be made into long thin strips either by being peeled apart along the grain, or through use of a planing tool. These long thin strips of wood then need to be softened; there are a couple different ways to achieve this but the easiest is to just leave them soaking in a pot of water for seven to ten days, followed by several hours of boiling.

This softened wood then needs to be pulverized into a pulp by repeated mashing; a monjolo seems an appropriate tool here, though a grinding millstone may also prove suitable. Once the wood has been pulped, add some lime as a bleaching agent and mix it in thoroughly; this will whiten the paper and ensure that writing on it shows up with more contrast. Then use rollers to flatten this bleached wood pulp as thin as possible onto baking trays, and bake at low temperatures until dry. This is one process by which writing paper may be produced, though there are others.

Anyway, once you've got significant paper manufacture going I must once again implore that a printing press be developed and put into service. A printing press allows the mass production of identical books, which has several beneficial effects. First and foremost, printed books allow those skilled in a specialist field to more readily spread any improvements of knowledge they make to others in their field, and more rapidly receive knowledge of such updates as well. This drastically increases the pace of invention in a society, especially if there are rewards in place for coming up with useful things and then sharing that knowledge. It works like this:

Let's say an ironworker figures out an improvement to case hardening that drastically increases the durability of tools. This ironworker goes to a table-ruler to file for a patent on the technique; a patent being a legally enforced agreement that for a number of years (no longer than ten, or else it stifles innovation) anyone simply using the technique verbatim without themselves making improvements to it has to pay some small amount of tribute to the original inventor. The catch with a patent is that it cannot remain a secret; the inventor has to provide a detailed account of their new invention, how it works, and how to make it to the table-ruler.

This table-ruler who had a patent filed then takes it to the printers. The printers make many, many copies of the instructions, which are sent to all corners of the Ten Nations. This allows anyone skilled ironworking and able to read with the ability to use the new technique, though unless they are able to come up with a significant improvement on the technique (which they would themselves then be able to patent), those other ironworkers will be paying minor tribute in order to use the improved methods. In some contexts the idea of paying minor tribute to use an invention doesn't make sense (such as in an army or something similar), but the general principle of rewarding people for coming up with useful inventions and then sharing them still holds true regardless.

A robust patent system thus massively encourages people to not only come up with new and exciting innovations, but to then strengthen society by sharing their innovations. And for a patent system to work properly, it needs the ability to make loads of copies of the same text that a printing press allows.

The other major impacts of printing presses also largely have to do with accelerating the spread of knowledge, and by doing so increasing the number of inventions produced to strengthen society. A properly written instructional text may not be enough on its own to confer mastery of a difficult subject like healing, but it can vastly improve the effectiveness of teaching when the student has a book of knowledge to study in addition to learning from one who has already mastered the skill.

The last major impact of printing presses is to make individual books less insanely valuable; unlike manually scribed books which may only have a few copies period, printed books can be produced in truly vast numbers. This means that if a library happens to catch fire it's less of a catastrophic loss of knowledge, since copies of almost all the books inside were probably printed out and sent to all the other libraries. And yes, there really should be more libraries throughout the Ten Nations than just the one near your great house. Redundancy is important for making sure a single setback doesn't turn into a major disaster.

I will now turn to the topic of spycraft.

First and foremost, I'm about to explain methods of sending messages in such a way as to make it extremely difficult for anyone but the intended recipient to read them. This is the field of study known as cryptography. All of the ciphers I am about to share are moderately complex to implement, but are highly effective at concealing the contents of a message. But first, I must speak of a tool that is required for all of them, known as the cipher wheel.

A cipher wheel is two circles of thick writing material, each of which is different in size and with the two connected by a pivot in the center. Each circle has the entire alphabet and some common punctuation arranged around it in the same order, so that at any given shift each symbol on the inner circle lines up with a given symbol on the outer wheel, with the possibility of lining the wheels up so that all the symbols correspond to themselves. The cipher wheel is critical for all three codes I am about to discuss.

First is the Viginere cipher. How it works is as follows: both sender and receiver agree on a keyword or key phrase ahead of time. When the sender is writing the message, at each symbol they rotate their cipher wheel so that the first letter of the alphabet on the inner circle corresponds to the symbol at the matching place in the key phrase on the outer circle. Then they find the symbol on the outer wheel that currently corresponds to the symbol they would normally write, and write that symbol from the outer wheel. When the sender reaches the end of the key, they repeat the key and continue writing the message.

When someone receives the message, they do the same process in reverse to decode it. They use the agreed upon key phrase to rotate their cipher wheel to the correct orientation at each symbol in the message, find the symbol on the outer wheel matching the current symbol of the ciphertext, and the corresponding symbol on the inner wheel is what was originally written. The viginere cipher is reasonably secure, but can still be broken by a sufficiently determined and clever individual.

The next cipher is one I will refer to as pattern shifts. Similarly to the Viginere cipher, the pattern shift cipher relies on the sender and recipient both having agreed on a key ahead of time. The difference for the pattern shift cipher is that instead of a word or phrase, the key is an initial orientation for the cipher wheel, followed by a repeating pattern of left or right shifts on the wheel for each symbol written. For example, shifting the wheel three spaces left after the first symbol, then two spaces right after the second, then a quarter turn right, then completely inverting the orientation of the cipher wheel, then repeating the pattern. Aside from that, encoding and decoding works much the same as the Viginere cipher. I would rate the pattern shift cipher as somewhat more secure than the Viginere cipher, though perhaps more error-prone. However, like the Viginere cipher, the pattern shift cipher is theoretically breakable by someone sufficiently clever and stubborn.

There is exactly one method of encoding a secret message that is impervious to all attempts at codebreaking, and that is the one-time pad. A one-time pad is an entire page of cipher wheel alignments longer than an entire message, with each alignment at each space in the message being selected by random means such as specially made dice. Since the series of cipher wheel alignments is entirely random with no rhyme or reason to it, it is impossible for the original message to be retrieved by reasoning without a copy of the one-time pad used to encode the message.

Only two copies of any given one-time pad are produced: one for the sender, and one for the intended recipient, with which pair a one-time pad is part of being written somewhere as an identification number. This identification number is added to the message as the only non-encoded part of the message, allowing the receiver to know which one-time pad they need to use to decode the message, which works the same as for the Viginere and Pattern Shift ciphers. Once used the one-time pad is to be thoroughly destroyed, so as to prevent its re-use from compromising the absolute security it offers; burning it is an effective method.

Yes, I am aware of the cumbersome nature of one-time pads for sending secret messages, and in most cases they are unnecessary when simpler codes will suffice. However, when you absolutely must be sure that no-one but the intended recipient reads a given message the one-time pad is the most secure possible option. Perhaps the best use for the one-time pad would be sending lists of message keys for the Viginere and Pattern Shift ciphers, so as to make sure which keys are being used cannot be deciphered from the message containing those keys.

That said, there is a way to get most of the security of one-time pads without quite as much cumbersome bookkeeping, though some will still be required. The method I am about to share is referred to as message-derived key generation, and is a form of pseudorandom algorithm, meaning it produces outputs that look random while in fact acting in a completely predetermined manner based on initial conditions. It works like this: the unencoded message (referred to as the plaintext) for the first message is encoded via one-time pad. Once the recipient has the message and decodes it, they then produce a version encoded using the Viginere cipher based on a list of key phrases possessed by both sender and recipient. The sender of the original message can of course simply encode the plaintext to get the viginere-encoded version without needing to decode it from the one-time pad.

Either way, after the first message (which was entirely secure due to being a one-time pad message), both parties should have identical viginere-encoded versions of it. These viginere-encoded versions of the first message are then used as the key to encode and decode the second message, with each letter in the viginere-encoded key being read as a cipher wheel alignment. Once both parties again have the plaintext, the next viginere or pattern shift key on their list is again used to encode the plaintext and generate the next key. To the best of my knowledge, this system is highly secure; while not perfect, it should be an extremely onerous task to retrieve the plaintext without the full key. Even if an adversary does somehow acquire the plaintext version of a message it doesn't fully compromise the security of subsequent messages thanks to the Viginere part of the code, but it does make breaking the code of subsequent messages vastly easier.

Given that fact and how the odds of it occurring get higher and higher the longer the string of messages goes on for, it would be wise to reset the string of messages every once in a while with a fresh One-Time Pad. Which you'd need to do anyway to refill the list of viginere keys when it ran out, assuming you couldn't have people meeting up to do that in person.

Next is the subject of spies themselves, and how you can effectively use them to undermine your enemies, ferret out secrets that others do not want you knowing, and prevent others from doing the same to you. I must preface that spycraft is by no means my specialty so I will only be able to describe this next section in generalities. Still, I hope the ideas I am about to share will prove useful to you.

The first topic I will be covering is infiltration, the practice of deceiving your opponent to believing that someone working for you is actually working for them. There are several ways of going about this, however in all cases they require what is known as a cover identity.

A cover identity is the pretense under which an agent is living and working in their assumed role. It can be thought of as an act that the agent performs while under scrutiny, except that they need to be able to improvise it as needed since real life doesn't follow a script. As such, agents need to be knowledgeable and competent with any wisdom and skills their cover identity is supposed to have. If an agent is claiming to be a cartwright for example, they need to be at least somewhat competent at making carts. Failing to maintain a believable cover identity is a near-guaranteed way to ensure that an agent is found out, and almost certainly dealt with by whoever they're supposed to be spying on. In some cases this simply leads to an agent being killed, but if your adversary is smart they may not let your agent know they have been found out, and will instead feed them misinformation that they will dutifully pass on to you.

The three main types of agent that I will be discussing differ mainly in how they build their cover identity, and how long it needs to last for.

An operative is a type of agent who is inserted a relatively short time before they are supposed to undertake whatever action they are supposed to perform, then is supposed to escape before they can be captured. While they need to be able to act their cover identity convincingly, it is usually not necessary that they be able to keep it up for years. It is a pretense that allows them to get in, do what they need to do, and get out.

By contrast, a sleeper agent is trained without any specific mission in mind. They are inserted into a target region with the intent being that they will live there as if they were not connected with you for several years if necessary. However, they are equipped and trained to perform whatever underhanded task is required of them when contacted again. Given the long time that a sleeper agent needs to maintain their cover identity it is crucial both that they be capable of acting every facet without arousing suspicion, but also that they are emotionally capable of betraying those they have spent years living among when ordered to. It is often useful to recruit those native to the target region to act as sleeper agents, as they will be more able to assume the customs and manner of speech of a local, thereby arousing less suspicion.

Lastly, agents-in-place are unique in that they come with cover identities already made. An agent-in-place is a person who already works for your target when recruited; the idea is that since they are already working for the target, their betrayal won't be anticipated. There are generally four methods of turning an agent-in-place, referred to as Money, Ideology, Compromise, and Ego.

Turning someone with money is pretty straightforward; if the target for recruitment has greed overwhelming good sense, just pay them with something that appeals to their greed in exchange for them doing something underhanded that benefits you. Ideology is a bit more complicated; effectively you need to convince the target that your ideas and customs are better than those they currently follow, to the degree of justifying betrayal.

Turning someone with Compromise is situational, and relies on the target's cowardice. It requires proof (or a very convincing fabricated account) of a misdeed on the target's part. Possibly even a misdeed you bait them into committing, if you cannot find evidence of an already existing one. Threaten the would-be agent with revealing evidence of their misdeeds and thus subjecting them to punishment if they do not work for you, and you stand a good chance of getting their compliance. Especially since the more they do for you, the more compromising material you have to potentially ruin them with.

Lastly, turning someone with Ego is a bit bizarre; it's basically exploiting someone's stupidity while convincing them that they're getting a chance to be smarter than their current employer. In our world there's a widespread misconception that being a spy is inherently glamorous to play off of, but I don't think that holds true in your world. Maybe best not to attempt this method unless such a common misconception takes hold.

Regardless of types, all agents who will be operating in a target region for a prolonged period need a Handler. A Handler is a person who acts as a message relay between agents and your spymasters. Handlers pass orders to your agents, and also ensure that the information your agents retrieve makes it back to you. Since Handlers will need to be in the target region to perform this role, they will need to have a cover identity; since they don't need to infiltrate any specific power structure aside from the general location, it is important that the cover identity for a Handler be as innocuous as possible. Farmers, common city-dweller professions, and others of that are rarely spared a second thought therefore make excellent cover identities for Handlers.

One very useful technique for Handlers and agents to pass information and items between each other is the "dead drop". This is a technique in which both the sender and receiver agree on a location and a set of times. The sender visits the location first and deposits an innocuous container holding whatever item or message they need to deliver. Some time later (or possibly simultaneously) the recipient comes to the location and retrieves the container. The sender and recipient do not directly converse, nor do they interact any more than absolutely necessary to ensure the dead drop takes place successfully.

This is useful because it keeps agents and handlers from knowing too much about each other, a practice known as information security. If every agent knows about every other agent, then just one agent being captured and successfully interrogated could cause your entire spy network in an area to be erased in extremely short order. By contrast, if each agent only knows the bare minimum needed to perform their roles effectively, it can be possible for an agent network to continue operating despite individual agents being detected and neutralized by the target.

It is due to the nature of information security that it is critical for Handlers to have such non-suspicious and well maintained cover identities. By definition, Handlers need to know more in order to perform their roles effectively; they know of a significant number of agents each, and if they are identified and interrogated every last one of those agents is placed at risk. It is also due to the importance of Handlers that it is wise for agents to be trained to resist interrogation; since those agents stand a good chance of being able to identify their Handler, every moment that an agent can resist interrogation for buys their Handler time to escape and make arrangements for the agents they are in contact with to avoid being detected while a replacement Handler is arranged.

I will also be telling you of some basic methods with which to detect spies sent by your own adversaries and foil their efforts.

Most obviously, if your opponents' agents aren't any good at keeping a cover identity it can be pretty easy to identify them as not quite fitting their assumed role. However, even if they have very good cover identities there are still some techniques that can catch them out, often by exploiting the times at which they have to break cover in order to perform whatever underhanded task they have been assigned.

One example of a tell I can think of is to keep an eye out for people who meet suspiciously often without seeming to have a good reason for it; that can be a sign of a Handler and Agent getting in contact.

Another method to trap spies is to keep watch of good locations for dead drops or secret meetings. If such is observed happening, it can provide a very obvious method of finding agents in action.

A much more active spy trap is to deliberately try and bait an agent with the seeming promise of uncovering a very big secret for the benefit of their masters. Then if someone happens to come and try to spy on it, it is fairly easy to identify them.

Of course, the absolute best way to find out how many spies an opponent has sent your way is to try and get an agent into the institution they use for keeping track of their spies. Still, this is easier said than done if your opponent is at all competent.

One last note on spycraft: It is a good idea to have an active spy agency all the time, not just during times of war. A permanent organization of spies can build up expertise and extensive networks over time, and provides an apparatus for discovering crucial information that can be used the moment it is needed. Constantly dismantling and creating spy agencies will just lead to confusion and reduced effectiveness.

And now for some more food related topics.

Some time ago we discussed pressure canners, and you opted not to put them into use because I failed to explain the proper construction of the safety valve needed to keep them from building pressure to the point of bursting. However, I think I've figured out how to do that now: just make a small hole in the lid and ram a plug of soft wood in there until it stays lodged in place. If the pressure builds too high, the plug will be shot out the top and safely vent steam straight up.

Also, seed drills. You expressed some skepticism regarding the burying of the seeds, which is fair. They don't need to go all that deep; perhaps only the length of a person's thumb at the deepest. Probably best to experiment on a small plot to see which seeds respond best to which depths, though.

And now for a topic best referred to as a curiosity, at least for the time being. You asked what use it was knowing that sound is a series of ripples in the air? The answer is that it can allow the blind to navigate the world almost as good as if they could see, or allow anyone to navigate in total darkness. The trick to this is that sound travels at a known speed, meaning that if someone makes a sharp 'click' sound with their tongue and listen for how long it takes the click to bounce back in different directions, they can get a sense of how far away various obstacles in their surroundings are with decently good accuracy.

A sighted person will probably need to spend significant time blindfolded while they practice this skill of navigating with sound, but if someone is blind they really don't have anything better to do. Either way, echolocation does have the potential to be a useful skill for some people.