funny, i think i've posted a couple cannabis tincture threads on BL, and i'd been meaning to post another. here's my experience:
*tincture attempt #1: allow ground up bud to soak in 75% ethanol in the freezer for a month. filter out plant matter. very little effect, probably not entirely placebo, but maybe.
*tincture attempt #2: grind up bud finely, leave in a thin layer on a baking dish in the oven for an hour just below the boiling point of THC (assuming wiki has the right boiling point and my oven's temperature control is reasonably accurate). soak in isopropanol, filter, allow isopropanol to evaporate, redissolve crude hash oil in 75% ethanol. noticeably better than attempt #1, but still not quite there. i had some trouble redissolving (which i think people here suggested was because the iso extracted things that the EtOH couldn't redissolve). effect-wise, it was definitely active, but not as psychedelic as smoking or vaporizing or eating traditional edibles (in which the bud or hash is simmered for a while in the oil to be used in cooking).
*tincture attempt #3: grind bud, soak in heated propylene glycol, filter off plant matter. my thinking here was that propylene glycol is safe for consumption, and has desirable physical properties of both alcohol and cooking oil/fat. that is, cannabanoids are soluble in it, it makes a good medium for sublingual dosing, and it's much less volatile/flammable than ethanol (good if you've got a gas stove). so i figured i could decarboxylate the THC in the propylene glycol as i assume is happening when you cook with cannabis by simmering it in oil. this worked pretty well. still too biased towards the sedative side as opposed to the psychedelic side (and i say that as someone who prefers the stereotypical indica type high). but on the right track.
*tincture attempt #4: same as #3, but with pre-extracted hash oil (from a cannabis club, presumably obtained by butane extraction). about the same level of success as #3, but worked much better in an e-cig since there was less unwanted junk from the plant in the mixture.
so overall, i'd say i'm on the right track, but my current guess is that i'm not letting the cannabanoids simmer for long enough in the heated PG. i think i got impatient and only let it simmer for 15 or 20 minutes, whereas if i were making brownies or something, i'd probably simmer for more like 45 to 60 minutes.
TL;DR: yeah, you probably need to decarboxylate first. depending on what consumption method you're going for, you might want to consider propylene glycol as a solvent. also, last i checked, the state of tincture recipes and advice on the internet (surprisingly) fucking sucked. reinforces all sorts of stereotypes about stoners (people talk shit on PG because it's not "natural", took me forever to find out decarboxylation was even necessary, etc).
for attempt #5, i will be using hash oil, PG, and a longer simmering time. with any luck, i'll have something i'm actually happy with. i'm excited about the possibility of a legitimately working tincture that can be used either sublingually or in an e-cig.
also, for what it's worth, some paper from the company that makes sativex (an oral THC spray) mentioned that they use a mixture of propylene glycol and ethanol as their solvent. i have a feeling that this aids absorption: anecdotally, i noticed (both in myself and others) that absorption of the PG-only tincture was inconsistent. sometimes you'd spray a little in your mouth or swallow a couple drops, and it wouldn't hit for hours. but it was more reliable if mixed into coffee or taken sublingually.
i'm basically interested in tincture because it would allow for very discreet carrying and dosing.
advice and corrections on any of the above would be much appreciated. i was going to post asking if i was barking up the wrong tree with my decarboxylating in PG approach.