This barges right into and through the 20 year rule but I'm going to do it anyway because it is interesting and relevant to the question.

For the last few decades there have been an intense, somewhat panicked fear amongst the US nuclear weapons labs today that nuclear weapons designers have lost several important skills. No new nuclear weapons have been designed since the 1980s, no weapons have been tested at all since the early 1990s, and no weapons have been produced since then either. The scientists who made the weapons and tested them are all now of retirement age. Many have passed away. There is actually significant funding at the weapons labs for anthropologists, of all people, to go in and interview the scientists and to document the "tacit knowledge" (things known "with your hands" that are hard to explain on paper, like how to ride a bicycle) that these oldsters had.

The most famous example of this being a real problem is when the government decided a few years ago it needed to re-create an exotic, classified material inside many modern thermonuclear warheads known as FOGBANK. It turned out that the records of FOGBANK's production were not well-preserved and nobody who had ever made it originally was still around. They ended up having to spent $23 million on finding a replacement method for creating a substitute material with the same properties. (FOGBANK is thought to probably be some kind of exotic aerogel used as an interstage radiation channel in thermonuclear bombs, if you're curious.)

This kind of thing has led to discussions amongst historians and sociologists about whether nuclear weapons could be "un-invented." That is, if they are hard to make, if they require huge investments and lots of tacit knowledge development to make, what if nobody made them for a generation? It doesn't mean you couldn't make them again. But it would mean that making them again would involve a lot more than just a "turn-key" revival of weapons laboratories. You'd have to re-invent the bomb, which is a non-trivially difficult activity.

Anyway, I've always thought this was a pretty interesting case of lost skills and lost knowledge, given that this isn't even aeons ago, it's just decades ago. What makes them "lost" is that we aren't using them anymore and many of the skills in question are really quite specific to that task, so the skill-base doesn't get replenished readily or easily. It's a somewhat "artificial" lost-ness, because it is enforced by the international norms, treaties, and practices and the weapons themselves are so highly controlled that you aren't going to just have people randomly learning these skills.

Many hand skills are close to being lost- being able to completely hand-stitch a garment or a pair of shoes for example. One of the biggest problems for me as a historic shoemaker, however, is the lack of appropriate leather. There are techniques that are long since lost. For example, the tight boots that were so popular beginning in the 1780s for men were made from a particular leather that had an amazing amount of stretch and bounce back. The leather was tanned in a very secretive manner and no known records exist explaining how it was done. Today, leathers are often much more rigid, or if they do stretch, don't come back to their original size. The quality difference also means my stitching can't be as fine as theirs (15 stitches per inch is the best I've managed, finding Kangaroo to be closer to their calfskin).

1-5-3-6-2-4

Those numbers landed my grandfather a trip to Detroit, then a cruise of the Pacific. All because he knew how to tear down an engine and rebuild it from scratch. This is a skill that is quite uncommon in the current United States, but as the owner of a small family farm it was vital information for my grandfather—and most of the country. As a child of the depression, he couldn’t afford to pay a mechanic to fix his equipment, he had to fix it himself. His experience was so common that many histories of WWII mention how skilled US mechanics were in being able to keep their machines running in the field—particularly Jeeps and tanks.

So, how did the above numbers affect my grandfather’s fate? He was drafted and showed up for induction. He said that representatives of a number of different services were present. The recruiter asked him if he knew the firing order of a straight six diesel engine.

“1-5-3-6-2-4”

He had fixed one the week before. So, he was sent to the Packard plant in Detroit for further training, then assigned to LCVP’s and shipped out across the Pacific. He never saw combat (which may be why I am here) but was part of the occupation force in Japan. All because he knew how to fix an engine.

(I must note that I had to look up the firing order, and may have it wrong. My grandfather had no such difficulty)

Starting at 400,000 years ago spear throwers, or atlatls, appear in the archaeological record. Atlatls are typically a wood shaft with a cup/divot/spur at one end. A dart is placed onto and parallel with the atlatl, with the blunt end of the dart in the cup. The thrower holds onto atlatl and dart, cocks their arm back, then brings the arm forward to cast the dart. For more info see this video if you are confused by my terrible description.

By increasing the lever arm, the atlatl increases the range of a spear by four or five times the distance for just a plain arm toss. Admittedly accuracy suffers at such long distances, but the distance gained from such a deceptively simple piece of technology is amazing. Modern throwers can launch a dart more than 500 feet, with the current record standing at 850 feet (260 meters).

Atlatls were used throughout the Paleolithic and supplemented the emerging bow and arrow technology. The skill accompanied humans on their dispersal across the planet. Atlatls persisted in the Americas and Australia to, and after, contact, and we still use the Nahuatl word atlatl to identify the spear thrower.

Today the skill is mostly lost, though some colleges in New England and Upstate New York are trying to revive atlatl use, and many dog owners use similar technology to entertain their canines. The only U.S. state permitting atlatl use as a hunting weapon is Montana, though other states like Florida will allow atlatl fishing. If you would like to learn more about this lost, but reviving, skill check out the World Atlatl Association.

The illumination of manuscripts is one that I can think of. There are quite a few medieval professions that are enjoying resurgence in the hobbyist world (blacksmithing for example) but not that one.

One that always fascinates me is the handling of tall ships. Sure, plenty of people can (and do) still do it, and do it well, but the peak of excellence is not quite what it was two centuries ago. If you look at a modern picture of a tall ship under a significant amount of canvas, like this one of Etoile du Roy (ex-Grand Turk), you will often see one or more of the sails rippling or even flapping; in the above picture, a large ripple is obvious at the port side of the forecourse (at the bottom right of the rigging as we look at it). This is not really a big deal nowadays, quite honestly, but it does lose the ship a small amount of speed. You can bet that a Napoleonic-era captain engaged in a stern chase would not have been overjoyed to see his ship's sails doing that.

To repost from a similar feature question:

So in Judaism you can't talk about lost arts skills without talking about tekhelet. It's a blue dye used in various ritual objects. Besides things in the Temple, it was used in the tzitzit, a sort of fringe that goes on a ritual garment, the tallit (I'm going to avoid rambling off-topic about this--if you're curious, ask). It was lost in antiquity. After the Temple was destroyed and most of the community was exiled, a lot of the ritual infastructure for things like that died off. While the dye persisted for a few centuries after, it eventually was lost in the centuries immediately following the Talmud. However, Jewish texts give several important facts about it:

1. The dye comes from a Mediterranean snail

• The snail has a shell
• The snail is fish-like
• The snail is rare
• Its color is like indigo, though it is not made from it
• It is expensive

Because of its relevance in Judaism, people have tried to identify it. An incorrect one was the cuttlefish--it turns out the dye made from it has nothing to do with the cuttlefish, it's actually the Prussian Blue artificial dye, using the cuttlefish as a source of organic material. And the important bit about the dye is the animal it comes from.

However, this happily falls into the third category! Someone eventually tried a snail that fit the bill, the hexaplex trunculus. And more importantly, archeologists have found evidence of that snail used in Near-Eastern dye production. It's actually the same animal that made the ancient royal purple, but with a slightly different process. Importantly, it's known to have been used by other Canaanite groups, including the Phoenicians. And based of a bit of dyed fabric, it seems that tekhelet was a dark, almost purple, blue color. After all, it's said to be the color of indigo.

Whether or not people should use tzitzit dyed with this is an interesting question of Jewish law. But either way you can buy them now. Which is pretty cool.

Can we talk about obscure arts that aren't quite lost? Chant hand-signalling is part of Jewish liturgy that's in grave danger in most communities.

Flint knapping, perhaps. Of course there are a few people who make reproduction hand-axes and arrowheads, and some knapping is done for surgical tools, but I don't think that anyone makes microliths now. The ones I am thinking of were about 1cm long, and half that in width, and appear to have been bonded to the side of arrow heads as barbs rather than forming a single tip.

There are a lot of more recent mechanical skills which have died out. For instance you have probably seen metal-topped café tables with a sort of decorative scalloped design apparently etched into the top surface. This reproduces the effect of "scraping". Picture a tool like a chisel, but with a squared off end rather than an edged end. The worker would use this to scrape the distinctive scalloped marks into a flat metal surface. This was done to introduce microscopic irregularities which would take up a small amount of oil, so that two such surfaces in contact would slide past each other rather than stick. As far as I know, the craft of doing this by hand has died out.

I took some medieval history in my undergrad, and IIRC, didn't the Carolingians and their contemporaries used to memorize whole books?

That would have made my comps take a whole lot longer.

It's interesting to me to see the short term loss of skills due to rapid changes in technology. For example, I'm sure at one point in the mid-20th century there was a decent number of people who could competently work a punch card computer. I'm sure many of those individuals are still alive today, but at what point will we no longer have individuals who knew how to operate those behemoths?

My grandad's favourite thing to do was work in the forest. He'd spend weeks felling trees. There was an art on how to wrap the chains around the tree in order to have it fall the right way when the horses/tractor pulled. Wrap the wrong way and it could get caught up in other trees. Wrap the right way and it could spin around and come down right where it was needed.

Used to be a lot of men in my area that knew all the little tricks and techniques of timbering with hand tools and chains. Now there's just a few old guys like my uncle that know any.