Minor Computer Flaw Frees State Prisoners
Ruvim writes "A Michigan State audit shows a software glitch let some prisoners get out early. From the article: 'The audit report shows errors in the release dates of 23 prisoners between October 2003 and March 2005. Some were let out early, while others were let out late... A flaw in computer programming caused State jails to release 8 prisoners anywhere from 39-161 days early, prisoners who were doing time for everything from embezzlement and drugs to bad check writing.'"
A whole 39 days early? Shit! They ought to hunt that bastard down and horsewhip them.
I guess I don't see the 'crisis' in this other than these people were low-level, non-violent offenders. If a software glitch had let a Ted Bundy out for another killing spree, I would probably be more concerned.
Fact is, we have WAAAAAY too many people in jail as it is. If we were to only charge and incarcerate those who pose a safety risk to the rest of society then you could probably monitor the entire population in half as many facilities with 1/3 of the correctional officers we have today.
The US incarcerates people largely to punish them for stuff they do to themselves. If someone is strung out on meth or heroin, they are only a problem to me if they steal something to support their habit. Considering the fact that theft is already a crime, I can't see how locking up people who are casual users and functioning addicts helps society at all.
These prison systems are getting too complex, too expensive, and are locking too many people away for "their own good".
Rep. Rick Jones: " 8 people is too many. I understand the department found another 15, that's too many, even 1 is too many."
Fuck that. Notice he shed no tears for the few that were held too long? I'm glad some of them got out early. The only sad thing in this story is that somebody got held longer than they should have.
"Rocky Rococo, at your cervix!"
I too would be tempted to, say, compare a hash of the prisoner's name with that of mine...
In Soviet Washington the swamp drains you.
On the bright side, they'll be getting their free subsidized digital television converter boxes any day now. Welcome to freedom, gentlemen!
because it is free!!!!
I am sorry. Very sorry.
What the hell is considered a major flaw?
The gates in my computer are AND, OR and NOT; they are not Bill.
I hope the software is run on Windows... If my bank was off by that much even once, I'd get a new bank!
I'm not saying that mistakes don't happen, but that's bad! Fortunately no one like John Wayne Gayce was let out mistakenly.
What are the odds that the 'software glitch' has a SSN and enjoys fast food?
Support NYCountryLawyer RIAA vs People
they would end my probation early!
feeling lonely? grab a balled up pillow for company
Even though it was only a month early, who is to say this "minor.... ahem" computer glitch couldn't let people go years earlier than planned. Also, shouldn't jails use both computer and physical data to make absolutely sure they are doing things properly? I know someone is going to comment to this saying that I'm wrong and that it would take too much space for all those filing cabinets, but I say that this is a perfect example of how I'm right. If they had another medium to check their data, this minor computer glitch could have been found and fixed, with no mess-ups.
public class null extends java applet { System.out.print ("Tabula Rasa"); }
Microsoft: Minor Computer Flaws Imprison Free States
Quantum materiae materietur marmota monax si marmota monax materiam possit materiari?
Uh, if I were doing time, you better believe I would be aware of my official release date!
Click here or here.
Can anyone provide more technical information regarding this flaw? What sort of hardware were these systems running on? What operating system(s)? Who wrote the software itself, what language was it written in, and what was the exact cause of the flaw? Was it a database flaw? If so, which database product was it?
Indeed, we need more technical details.
Cyric Zndovzny at your service.
if((rand() % 2) == 0) sentence -= (rand() % 123 + 39); Either that or memory corruption. I'd bet on the former :-D
About 9 months ago, The Dallas (county) Sheriff's office installed a new prisoner tracking program and LOST some of the prisoners. No, they did not let them out, they were still in jail but they could not find them. (Even the prisoner's lawyers could not find them!) Here is an example: http://www.dallasnews.com/s/dws/dn/latestnews/stor ies/052905dnmetjailstuck.f2f1f79c.html
This message was brought to you by "Lack of Sleep."
I'd be more worried about those that got let out late. Surely that's ground for a lawsuit and some damages. Not to mention ciminal charges against those that illegally kept people in prison.
Run on sentences and fragments too!
sorry!
is password
Presumably he was voted in by Americans, what with their American spelling and use of double quotes.
English is easier said than done.
:wq
this is why I'm a firm believer in the ole' "lock 'em away and throw away the key"
The real victims in this case are the ones who were kept longer in jail without conviction. It's quite scary that no one at WLNS apparently cares about them. A toast to the future of the american justice system. I hope it has one.
Any sufficiently advanced libertarian utopia is indistinguishable from government.
If they suddenly got their sentence shifted from 15 monthes, to Death Penalty, or life without parole by the computer (and vice-versa). Just a wild guess...
C. Sagan : A demon haunted world:
http://www.amazon.com/gp/product/0345409469/
visit randi.org
It could have been copyright infringers that got out early.
Manual systems are vulnerable to errors also. After all, court clerks and jail paper-pushers are only human.
What I want to know is:
Do we know about all of the errors the old system had? If so, what types did it have and how often?
Do we know about all of the errors the NEW system has? If so, what types does it have and how often?
The ultimate goal is to be error-free. A good interim goal is to recognize your errors within a reasonable amount of time, fix them, and if it's cost-effective take steps to prevent future errors. BTW anything that keeps people in jail for extra time has a cost of {very big lawsuit} to NOT fix. But letting 1% of people out 98% into their sentence instead of when it's 100% complete may or may not be worth fixing, particularly for inmates unlikely to commit new crimes before their original sentence expiration.
Knowledge is how to play a game, intelligence is how to win, wisdom is knowing what game to play.
..we want Electronic Voting Machines again?
that technology sets us free ;)
Learn to spell 'cheque' you fag. Who voted this guy in?
No, he really meant bad check writers. They wrote the system that checks the release date.
You know, the real criminal types
So can someone please explain to me why "bad check writing" can land you in jail? This is not a rhetorical question, I seriously want to know. Is it considered fraud? It sounds a bit like debtors' jail, which is outlawed in this country, right?
EvilCON - Made Famous by
The original file is a bit slow ATM. I might take down the mirror again in a day or two.
Hey, I took Stats in my undergrad! If some prisoners were let out early, and other let out late, then it should all average out to zero! So really, the software works just fine in the aggregate! Typical Slashdot, can't see the forest for the trees!
"Reality is merely an illusion, albeit a very persistent one " -Albert Einstein
The problem seems to be more than software or hardware. A state law created a sentence-length committee or ruleset that was not fully communicated to the Department of Corrections (DOC). The DOC tried to interpret the information they had and came up with a manual for calculating a prisoner's release date. This manual includes two non-automated methods of calculating a simple release date, and some informal rules for calculating release dates in general. The DOC later wrote (or contracted out the writing of) the program that automatically calculates release dates.
The audit being reported compared the computer computation with the two non-automated methods and found that none of the three gave the same results. Not only was the software inconsistant with the manual, but the manual was self-inconsistant. The software may have actually used the right calculation, but the audit seemed unable to determine what the right calculation was (because of the confusing state law mentioned earlier).
This is the sort of article I always find disturbing... We found a minor problem but take our word for it - its no big deal. Why wouldn't this article explicitly what the flaw was that allowed this to occur - this isn't any sort of system security issue. And then to compound it by say "but it wasn't any murderers" - why - sheer luck? "They say they've already taken steps to correct the computer glitch and will continue to work until the problem is taken care of." Okay - maybe some of you slashdotters are smarter on this than me, but how long can it take to correct a date calculation routine? Unless maybe you don't have one of the dates you need????
The summary emphasizes the let out early part, as does the title. But hey, I'm not so pissed about that, but "while others were let out late" really pisses me off. If you can't even tell them what punishment they're going to receive at most 100% of the time, then there are major issues that need to be resolved.
I have two questions:
1. Why don't they check the (paper!) documents they got from the judge or whoever to check if they really were sheduled to go out that day?
2. Why didn't those let out late complain? I'm sure the first thing they did when they got there, was circling the date they were sheduled to get out on their calender. (or whatever paper they have handy). How can they not notice that they passed that date by x weeks?
This story as usually raises more questions than it answers...
OT rant: Damn you, shallow news outlets! If a plane crashes, we get every small detail about what happened on which second, and what systems failed, but when it's about computer problems, all they can tell us is a 'glitch' or a 'crash' happened because they think it would be 'too technical'. Just tell exactly what the problem was, and if people don't understand completely, it's not going to kill them.
This says nothing about the underlying problem. Was the release date incorrectly scheduled from the start? Did it change mysteriously while the person was incarcerated? Did the system just incorrectly say "Release this guy" on a random day? Did it give the wrong person to be released? If so, was there any similarities between the two inmates? There just isn't enough information here to make any guesses.
Was it a contractor or an in-house developed project would also be interesting. As well as what happened to the inmates who were released late? Is it just "tough luck" for them?
Does anyone have any additional information?
... "minor glitches". the sorts that cost banks millions in dollars as fractional pennies are gobbled up or cause machines designed to treat cancer with radiotherapy burn and kill patients. if i were a resident of michgan, i'd demand an inquiry and follow it up with at least one big law suit. this should have been discovered and fixed during testing, not deployment.
isn't a software to be used at such a sensitive level supposed to be reviewed N number of times before it is deployed? This glitch sounds like it'd have been an easy one to get caught if reviewed by enough eyeballs.
Oh... what's the mantra for opensource again?
Which is a far more serious software glitch.
developers were liable for their bugs, cells that were emptied because of the glitch wouldn't stay empty very long.
perception is reality
Don't be sad, my /. friend. These over-detained individuals will almost certainly sue and win a large portion of money for their troubles.
Even one day over seems to constitute "wrongful imprisonment." If I understand that legal term at all, that would be grounds for a monetary settlement.
...that once you have served your time you are again a citizen. So why is it more upsetting that criminals are released early than that citizens are kept locked up in prison? I think that is at least as problematic.
Deterrence only effects those who don't resepect the rule of law. If people respected the rule of law then the law itself would be enough. Deterrence works on the principle that people who don't respect the law but don't want to take the chance of being punished wont break the law if they think the chances of getting punished are high enough. To give them that impression though, you have to punish lesser offenses to make the greater offenses seem riskier by comparison.
No; he really meant bad Czech writers. It was a reference to Tedd Sallay and Josef Simanek.
"Finally, my parole hearing has come and I'm ready to see my family again!"
"No, actually, it says here that you're scheduled for execution. Any last requests?"
a mundane detail!
"I just wanted to point out how how she was."
Damnit! I meant how hot. How how, white man.
funniest off-topic post this week
Your memory of orders of operations must have been corrupted.
That's equivalent to: sentence -= (rand() % 162);
which is different from the intended: sentence -= ((rand() % 123) + 39);
Actually this would take more of the rand bits into account hence produce
a more fair result:
if ((2.0 * rand() / RAND_MAX) == 0)
sentence -= (int)((123.0 * rand() / RAND_MAX) + 39);
Arash Partow's Philosophy: Be a person who knows what they don't know, and not a person who doesn't know.
"A flaw in computer programming"? I'm calling their bluff. This is complete bullshit. How hard is it to calculate the length of a jail sentence? This was either user error in the original input of the data or someone on the inside was exploiting the system for their own purposes. And exactly how many people were let out late? Programming error or not, the State is liable.
Of course, you don't give a shit about that. Just so long as some person next door isn't getting shitfaced on dope. He can get drunk off his ass, but God Forbid that he do meth.
Just my prospective: personally, I think abusing anything of any kind (drugs, alcohol, food, gambling, Japanese school girl panties vending machines, whatever) is bad for the mind and body--but also for any kind of close relation, especially children... Without getting into morals and that crap, my argument is based on the relative damage an activity can do to a person.
I've seen too many people taken out by some of the more common stuff: alcohol, gambling, tobacco, marijuana (often as the gateway drug to other worse stuff), and it sucks. These are all things that can be done casually, but too much of it destroys people, families, and sometimes even generations. However, the speeds at which these things destroy a person/family vary greatly. Meth and other hardcore drugs, which significantly alter brain chemistry instantaneously, should be policed very tenaciously, because there is no way to use them casually.
I'll limit myself to two anecdotes: 1: these guys who used to own an engine machine shop I frequented somehow got involved with meth. It was an instant and violent transformation. One month, they were regular Joes, hardworking, and successful. The next month, they didn't do much work at all, the month after, they would take customers money to pay for their habits. In the period of six months, the once previously very beautiful (model quality, honest) wife of one of the greasers looked like she would pull tricks on the local boulevard for a high (no doubt in my mind that this is the case), and every one of their three (14, 16, 18 y.o.) daughters was pregnant. Ouch. Shortly after this, they got busted for making meth, and last I seen them, the daughters were looking quite a bit like your average geriatric. These guys were above average intelligence, but they still managed to destroy at least a dozen lives--in less than a year.
The other: the older brother of my grandmother. He's been a pothead since way before it was fashionable. Apparently, back when my great-grandmother was still alive, the whole of our family was fairly well to do, and they could afford all of the toys they ever wanted. When he got into weed (not alcohol, or anything else), he proceeded to fuck up every classic car the family had. 50's and 60's Corvettes, Cadilacs, Chevy's, anything you can imagine, they had it, he mangled it. Since great gamma died, he's squandered literally tens of millions of dollars of property, antiques, cars, and businesses. It's really too bad she was so senile when she wrote the will, because he got most of what was left over. Like I said, he's always been a pothead, but his daughter is now a meth head with 4 kids: one pregnant 13 year old, one incredibly obese 15 year old son, a 17 year old with 2 kids of her own, and one dead.
And this is why I say that anyone who says that weed doesn't fuck some people over is full of BS. Indeed, I've smoked, at first just to see what it was all about, and just because it's not a one-way road to misery like some other drugs. I must say that I'm not especially impressed, but it's fun to do socially anyway. No doubt, there are people that can handle it but there's always some weak willed asshole that will succumb to even mildly addictive behaviors. This is why I'm NOT against legalization. Once people get over its taboo quality, they'll realize it's just not that great.
Constitutional rights may be respected, repealed, or modified; but they must never be ignored.
http://news.google.com/nwshp?hl=en&tab=wn&q=/ Dallas TX County Jail got Windows-based Software. Lost inmates in the system. (City Council Member)...became concerned one year ago about Info Integration, the computer company installing the $10 million system... Info Integration opened for business just two months before getting the contract.
The only thing new in this world is the history that you don't know.[Harry Truman]
A portion of these must be America's dumbest criminals. Let's think about this for a second. 8 Criminals were released 39-161 days early. If I was those criminals, I'd give out a giggle, tell my best friend, and then hope nobody else notices. I understand how this went under the radar, because who's going to speak up and say "ummm- excuse me, I had another 39 days of free food coming my way".
But 23 prisoners less 8 prisoners released early leads us to 15 criminals, and their families/friends who didn't notice since 2003 that they were not being released (you'll notice it doesn't give a date for the 'later' prisoners, but only the early ones). Wouldn't you speak up and say "ummm- maybe you should check your paperwork". Wouldn't they have some sense of time/date? I'd say they're pretty re-habilitated if they peacefully sit back and wait for somebody to notice...
-M
when you see the word 'Linux', drink!
Next time I write code for the prison I will remeber this clause: If (inmate = me) releasedate = today
This is a real eye-opener as to our exponentially accelerating reliance on technology.
What if this had been a more severe glitch? What if murderers were set free?
If software inexorably harms society, who is responsible? The programmers?
Looks like Michigan's new programming labor camp didn't work out so well.
In light of this failure, they have just adopted the "prison guard labor camp". Give those bastards the keys and make them do the shitty job of watching the prisoners.
Need Free Juniper/NetScreen Support? JuniperForum
We should ban computers.
Another link
/ cia/
http://www.csun.edu/CommunicationStudies/ben/news
Interesting claim. First off, I take it you're against incarcerating white collar criminals, then-- after all, we could just take their money and blacklist them from all financial institutions. No safety risk to us, no need to stick our Enron execs behind bars.
Secondly, if you look at both federal and state prison stats http://www.ojp.usdoj.gov/bjs/prisons.htm only about 21% of the prison population are drug offenders. Overall, over half of the entire prison population are violent, with the remainder falling into property/public order offenses. Thirdly, if you talk with the correctional officers out here in California, the drug offenders tend to be safer and more cooperative, thus reducing the necessary officer-to-prisoner ratio. A higher concentration of violent offenders will only up the necessary ratio, so I have no idea from where you derive your figures.
The US incarcerates people largely to punish them for stuff they do to themselves.
I realize you'll get plenty of rah rah here since you're pushing a popular Slashdot political viewpoint, but if you're talking drugs, I don't see how 21% is "largely." A better argument would be that the US incarcerates people largely to deter behavior of which the elected government does not approve. Perhaps you feel that some government decisions do not help society; I certainly agree that there is plenty the government can always do better, but you don't address any the major process equity tradeoffs necessary in shaping this policy. The only sad thing in your comment is that you've bought into some broad political groupthink without critically considering the facts or policy implications.
Repeatedly I read the theory that all (or most) drug related crimes and activities would misteriously go away if those drugs would be administered by trained professionals who would only give them to those who really need them.
There exist drugs which are handed out like this (all prescription drugs) and yet there also exists a blooming black market for those. Just read your favourite spam if you don't believe me. So if the concept doesn't work for Cialis, Viagra, Prozac and whatever they are called (I'm just citing from the spam I get), why would this suddenly work with heroin, crack and cannabis ?
Correct me if I'm wrong, but don't most addicts start thinking about ending their addiction for the first time once they are confronted with a criminal court (i.e. get a drastic picture of their situation) ? The vast majority of drug addicts will only go to those doctors voluntarily if no questions are asked but certainly not to break with their habbits or to go through often painful treatment (just read the conspiracy theories about methadone as substitute for heroin out there!).
I was upset when I heard a UK judge claim that cannabis couldn't be given to a cancer patient for pain relief. The decision was based solely on "Moral Grounds". Cannabis is illegal so is not allowed - despite the doctors stating that it would be better than methadone or similar for that patient's level of pain. The odd thing is that methadone is "Class A", yet can be prescribed as pain relief in later stages of cancer. I've met some patients that don't like the side effects that some people find addictive - they just want to have no pain. Being too doped to do anything isn't so nice.
On the other side, I've also been in a hospital that allows smoking of cannabis as long as its outside and downwind and the doctor has said it doesn't interfere with the chemotherapy. I can't imagine the police dragging a cancer patient off to court in the middle of treatment.
Talk about free software :).
> Deterrence only effects those who don't resepect the rule of law.
Those people includes most of us, in varying degrees. Ever broken the speed limit consciously? And slowed down and nevously glanced at the speedo when you saw a cop car in the rear view?
Deterrence does work, and for our sake, I am glad it does.
> I guess I don't see the 'crisis' in this ...these people
> were low-level, non-violent offenders. If a software
> glitch had let a Ted Bundy out for another killing spree,
> I would probably be more concerned.
You're not wrong, but a few points to think about.
Often people in jail for one crime are guilty of others. Just because a man was in prison for a fairly innocuous offense doesn't mean releasing him was harmless.
Though there is no indication of this being a problem in this case, often people in jail on more minor offenses are waiting to begin serving time or start trial on more major offenses. An early release on the minor charge could interfere with the necessary transfer to another facility.
> The US incarcerates people largely to punish them for
> stuff they do to themselves. If someone is strung out
> on meth or heroin, they are only a problem to me if
> they steal something to support their habit.
What about minors being abused and neglected by custodial parents with a drug habit? What about the people on drugs who don't want to be, but can't quit without legal intervention? What about kids born addicted?
While this isn't "everyone", there was a recent story in the paper about a preschooler who brought in some crystal meth to preschool. And another about a child left alone in the house to tend to the father's home meth lab (which tends to explode; lotsa kids die that way each year). There are a lot of problems with the idea of leaving the drug users alone.
While we can argue about the right way to deal with these problems, and you could certainly make a good case that the prison system is the wrong way, the problems are real and no one has a good solution.
Before anyone flames me, just remember I said the original poster was -not- wrong. I'm not saying prison is the solution; just saying the problems are not limited to "drug abusers stealing to support their habit."
Having said this, what the hell kind of computer system doesn't involve someone having access to and checking the original sentencing information prior to releasing the prisoner? Is that so tough?
I think that the state should invite the contractors who wrote the buggy software to a stay at the nearest lockup for a few months. That'll teach them to cut corners in testing their product.
Would spam still be economically viable to send if any one of us could get it for free from their Doctor - if you could demonstrate an addiction? To make illegal drugs cheaper rules out any illegal activity. Why would a heroin user go to the hassle of smashing a car window and stealing a CD player to sell for his next hit - when he can walk into his Doctors - get some advice and receive the drugs they are dependant on? The real fuckers here are the dealers - and they need to be cut off, and put out of a job by providing these treatments for free - rather than not at all.
Sure, I don't think there is any single solution to the drug problem - but I believe that this is a strong step in the right direction. We've been approaching the entire problem from the wrong perspective for a long time.
I'd also like to hear other alternatives we have to tackle the problem in the short term. The War of Drugs has run for how long now? Decade or 2? And we are no closer to solving the social problems associated with it - we just have more people in prison. The deterant doesn't work for the dealer, and for the addicted user - there is no deterant strong enough to stop them seeking a way to finance their next hit.
Heroin is not only taken by addicts. If you take it in certain ways, you can stay away from addiction for quite a while. In this case a doctor shouldn't give you the drug which is exactly not what you are aiming for (The you doesn't mean you personally but "you" as heroin user :-). As a result you would try to get it from somewhere else. Just think of cannabis or cocaine, which are addictive to only few, and yet people pay premium prices on the black market.
You may well argue that the war on drugs has not created the intended results. However, it has been fought in many different ways (thing of extremely liberal politics in the Netherlands and excessively strict ways in some south east asian countries). There appears to be a desire in many humans to drug themselves even if that results in great damage to themselves and others. If you are strict, you may hold off some who may have tried it under more liberal conditions, but you increase related crime because of the high risk associated with trade and consumption. If you are liberal, fewer addicts turn criminal but there are more in the pool. Add to that that the success rate in addiction treatment is very low and you have a problem here as well.
Just save a little bit for me :)
Constitutional rights may be respected, repealed, or modified; but they must never be ignored.