When last I read up on fingerprinting it was pretty dodgy - the problem being not that it's inaccurate, but that nobody really knows how accurate it is. Apparently the professional society that analysts all belong to forbids its members from participating in studies, taking the position that it's "100% accurate when done properly" (which is clearly meaningless).

It's probably 4-5 years since I read about this, it'd be nice if things have changed.

>> Apparently the professional society that analysts all belong to forbids its members from participating in studies, taking the position that it's "100% accurate when done properly"

Refusal to run a gauge R&R while insisting errors are due to the operator is a sure sign of a problem. If it hasn't been done, that's one thing but refusing to allow it is quite another.

It's the same problem as it is with most anything. They want to use huge databases, but to do that usefully you need an unrealistically low false positive rate.

Even if the rate was actually really good -- something like 99.9% accuracy -- if you apply that to a database with a ten million people in it, you get ten thousand false positives, which is crap.

If you pick one of the matches at random, they want to be able to claim that it's a 99.9% chance that it's them, but it's really a 99.99% chance that it isn't. Or worse, since it's possible the true perpetrator wasn't in the database.

Even for things like DNA that theoretically can do better than 99.9%, you can't get much better than that in practice because before you can, the error rate starts to be dominated by human error rather than coincident matches.

I work in a consulting field where the professional society forbids its members from publishing statistics on client success. There are some good reasons it's generally frowned upon but it's also the only remotely objective measure of whether anyone actually benefits from your work.

It’s even crazier because automatic fingerprint matching is controlled by a small cartel of awful companies.

Whether a fingerprint is unique and wether a bad fingerprint is good enough to convict somebody I don't want to judge here, but there is another aspect of fingerprints:The negative check. If the fingerprints on the knife don't match the suspect's the suspect didn't use the knife (unless there's an indication of gloves) and thus creates a strong reason to show innocence. A "negative" polygraph test still leaves room for a good lier.

In a perfect world, the prosecutor would give that info to the defense with sufficient time. In the real world, the prosecutor will drop 30 pages off in front of the defense 20 minutes before the trial begins.

Can you provide evidence that this is a commonly tolerated practice in "the real world"? Are you a judge, prosecutor, or criminal defense attorney?

It's a common trope, but courts really do have discovery deadlines, continuances are common for a variety of reasons, and undisclosed evidence can be excluded from consideration.

New York state changed their laws on January 1, 2020, to alter the previous practice where the government only had to provide the evidence they had to defense lawyers on the day of the trial, and only if specifically requested by the defense.

The majority of people in jail in New York right now, today, had to accept/reject plea deals without seeing any of the governmental evidence against them.

It is to be hoped that the future of NY's justice system will be much fairer, but police unions are fighting hard against the new changes and they may well be repealed.

The "only if specifically requested" is commonplace and requesting evidence is a routine part of any case. I'm all in favor of "automatic" discovery in criminal matters—I don't believe the prosecution should have an information advantage—but the fact that it wasn't previously required by law isn't evidence that it was broadly inadequate.

Same goes for the deadlines. Sure, the law didn't set a specific timeline—but that's not evidence that prosecutors routinely "[dropped] 30 pages off in front of the defense 20 minutes before the trial [began]."

The "dropped 30 pages" is a pretty common complaint for public defenders. Now, it could be that they keep losing the cases, but it's pretty common for the state to only provide evidence on the day of a trial and often don't point to exculpatory evidence or exclude it from the documents being provided.

Your statement reminds me a lot of how a 'rapid psychological profiling' tool works in the PsychoPass anime, which you might find interesting as a good example of such a system used in a science fiction setting.