Hoping I haven't missed something interesting inbetween the philosophical discussion, here's my take:
Your civilisation has its own weapons, and (as we know) other holes in its knowledge, like sometimes no High Boots. But 'foreign' weapons and clothing come your fortress's way via trade. If your respective craftsdwarf gets his or her hands on these items, then the secret of their construction becomes (at least partially) theirs and they can start considering making them.
"Getting their hands on them" could include personally belonging to them (including assigned in uniform) and, if/when damage levels are implemented to all items with a repair system in tow, during the repair process.
Obviously it wouldn't be an instant success. Perhaps it starts off some low multiple, like 10%, to adjust the quality level downwards from whatever would have been achieved for creating/repairing something from the more familiar pantheon of items. Attempts to make/repair that item thus results in low-quality tat even from a master of the trade, initially.
If the adjusted quality is below a certain minimum threshold, you would get either junk (non-biological equivalent to vermin remains, perhaps suitable for Melting, if a metal material) or nothing at all might end up being created in the process. Or a pre-damaged item, if lower than "base quality" but still above "destroyed" or "rubbish" limits. (Attempts can be made to repair pre-damaged items, but that still carries the risk of destroying it while trying to fix it.)
Each involvement (or each non-failing engagement?) with the non-native item increases the multiple until eventually normalised (or nearly so, could be made asymptotic to 100%, for always that little bit of uncertainty) with native item creation for the skill-level already gained. Perhaps when the learning process adds to the familiarity, it should mean a reduced increase on the craft's own XP gain from what a native item would give. (Or fully zero additional experience, or else even randomly varying from zero to actually a little more gain than usual, due to the challenge. I'll leave that to be an aspect of fine-tuning of the concept, or just to be ignored altogether in any hypothetical implementation...)
The reasoning behind the "any interaction" stipulation is that getting a heft on a weapon (as per the OP) or actually wearing a bit of unfamiliar armour should convey some experience. Although I do feel that assignment via uniform might be considered somewhat "danger room"-style exploity (especially if, in an actual danger-room, appreciation for a particular foreign armour item is built up alongside the usual skill-building you get, at the same speed), so maybe usage alone can only drag it up to (say) 50% proficiency in understanding the item, compared with a 100% limit on attempts to create and... repair? Speculatively place that at a 75% ceiling, but that's another matter that could be be adjusted (or ignored) accordingly.
Should this system encounter unusual items whose interactions are nominally in display, e.g. a stone fountain (variation on a statue), or coming across a cage made of bones, then observing in-situ would be the prime learning method, behind the ubiquitous method of creation, with sitting in stockpile, or otherwise unconstructed giving the lesser degree of familiarity. (Haulage? Same as 'hefting' the object, I'd say, but maybe nothing, to avoid stockpile-restacking exploits.)
Novel items with other uses (e.g., say, something like a chaise longue... in context, maybe an elvish bit of furniture that can be used as both throne/chair and bed) you gain experience with by the given interactions (and perhaps 50% max experience allowed by sitting on it, and 50% max experience allowed by sleeping on it, but added together). Note that the fountains or chaise longues ideas are just illustrative plucked out of the air and not intended to be additional ideas. (Although... on second thoughts, procedurally generated strangenesses of this kind might well be something worth suggesting...
)
Anyhow... The process is "Oooh, so you say this is called a 'scimitar'? Well, I suppose I could make a copy...", goes through a bit of "How
do they get the curve just right?" and "Well, I can try to file the nicks out, but I'm not sure I can reforge that edge correctly, just yet" until eventually competence is gained and you're churning out home-grown *Dwarven scimitar*s or better, like they were your bread-and-butter short swords like your mammy used to forge...
We already have Secrets of various other kinds coming up, I
think the damage-and-repair thing is, the supply is already there, the skill-base is there (just needs conditional adjustment, according to some hidden-or-otherwise additional skill value, or just tied directly to the knowledge of the Secret, itself, however that's supposed to work for the upcoming stuff) and there's plenty of opportunity to waste raw materials on each utter failure to replicate (to balance the potential for spamming the learning process and selling all the low-quality failures) or distruction of the prized toy in a botched repair attempt (again, so you don't just hand around a damaged exotic weapon to all your smiths and spam them all into familiarity... should be tuned so that a single practitioner has a good chance of creeping the learning curve up but multiple inexperienced hands would more likely destroy it).
And if teacher/student skills get used to enable craftsdwarfs to pass on knowledge, then (with the suitably low multiple also being involved, maybe even reduced further due to awkwardness of communication) the skill might well be passed from craftsdwarf to apprenticedwarf alongside general skills.
Anyway, that's probably putting more flesh on the bones than I should be, while still leaving it look like a zombie that's been wandering through a briar-patch. TL;DR; treat foreign weapon construction as Secrets, and "blacksmithes
[sic] should learn", as per the subject is the progression from that. But don't just restrict it to weapons, or indeed just items that blacksmiths of all flavours can make, but extend across the range.
(Perhaps it could even also apply to any re-introduced Dungeon Master, so that (somehow) working with a creature not even an Exotic-Pet, but brought along (tamed) by the elves, let's him do his own tracking and training of that species (and maybe closely related ones), regardless of the lack of any PET-style tag. But I wouldn't like to suggest how the failures to succeed, from the percentage-multiple factor, might represent themselves when dealing with Giant Desert Scorpions or whatever.
)
Poll
3 (37.5%)
Author
Topic: blacksmithes should learn (Read 1935 times)