During some downtime last week between finals and graduation (also - yes I am graduated! Now taking a break at home before job and lease start), I decided to take up Nethack. It turned out to be a wonderful hybrid between Dwarf Fortress and TSR D&D, and I had quite a bit of fun (including DF-grade 'FUN') right up until I found and ID'd a wand of wishes and suffered a crippling case of decision paralysis. But the whole experience did tickle some brain cells on a problem that has bugged our ACKS group before - how to identify items?
Really the problematic items here are the command-word items (wands, staves, swords of charming, those sort of things) and some of the edge-case passive items like rings of fire resistance. Potions are easy enough to get with Alchemy, scrolls are quickly deciphered with Read Languages, and we tend to let people determine the bonus of a suit of magic armor or a weapon after sparring with it some (we assume fighter-sense and combat experience provide a fairly good sense for how much better than normal those sort of things are, and I'm way too lazy to keep those secret in use for any longer than I have to). But I think the keys to handling charged item identification are suggestive descriptions and blind activation.
To say that Nethack's wand descriptions are suggestive would be quite misleading, as they're random, but they are at least consistent and got me thinking about charged-item description, which we have to far slacked on pretty hard. Perhaps all detection wands have a small crystal ball embedded on the end of a stick whose type is appropriate to the type of detection, or offensive wands appeared charred, or so forth. This lets players get a general idea of the type of a wand, but not its exact function.
Blind activation is something which Nethack does do well, and it is one of the standard ways to identify wands. ACKS, however, has no clear facility for it. Upon a re-reading of the magic item identification rules and the wands, staves, and rods rules, it's not entirely clear whether the user needs to know a command word or just sort of points and clicks. Given the sort of hilarity that building experimental wand-testing setups can result in, I'm inclined to rules in favor of blind activation (particularly with the caveat that a blindly-activated staff will use one of its capabilities at random). Limiting blind activation to Magical Engineers and Arcane Dabblers (both at -4) might also be reasonable (also wow, I never realized bards got Arcane Dabbling for free... that's a pretty serious proficiency at mid-to-high levels). In either case, there's still a trade-off; running a full identification will tell you how many charges it has, but will take time and money, while a blind activation takes maybe a turn but burns a charge and has a chance of frying your whole party if it turns out the copper stick with a piece of amber on the end that you thought was a wand of treasure detection is actually a wand of lightning...
Other thought - Alchemy II is kind of a bummer, and given how potion miscibility works in ACKS, it would be cool if there were a way for players to pre-identify the 'potion countdown' (potency?) of a given potion so that they could better plan their imbibing. I'm sort of thinking potency identification on 18+ for Alchemy 1, then +4 extra per further rank (as usual) so that taking Alchemy II lets you more than double your ability to do it. On the other hand, alchemy is already quite good on the whole, and allowing methods to pre-determine potency generates more paperwork / record-keeping.
I also sort of want to build an extensive themed potion descriptor table for parties lacking an alchemist, but given how often that prof gets taken and how effective it is, I'm not sure it's worth the effort.