Vt. Lottery computer glitch creates extra winners

Aug 24, 2010, 9:48 am (19 comments)

Vermont Lottery

BARRE, Vt. — Linda Hebert says she didn't think twice when she purchased two nearly identical computer-generated "Fast Play" tickets at North End Deli Mart earlier this month.

"They looked like winners to me," Hebert says of the twin "Red Hot Numbers" tickets that she has been patiently waiting for the Vermont Lottery Commission to honor since she bought them on her way to work Aug. 4.

Turns out Hebert's double win was not due to lady luck but a computer malfunction — and she wasn't alone in "winning" that day.

Due to a computer glitch, which has since been rectified, 70 other tickets that were virtually identical to hers were sold during the same 30-minute span, according to Alan Yandow, executive director of the Vermont Lottery.

"It was a system-wide error," Yandow explained Monday, stressing that when it comes to those 72 tickets, looks can be deceiving.

"They are not winning tickets," he said.

However, Yandow stressed that does not mean Hebert and others who qualify won't get paid as soon as lottery officials finalize a process that will accommodate payment for "top prize tickets" — $450 each — and decide how to account for those transactions.

"Everybody will be made whole for the inconvenience," he said, noting that will be more expensive in some cases than others.

Hebert can expect $900 — $450 for each of her two $1 tickets — according to Yandow, who said some agents spotted the flawed tickets and refused to sell them to patrons.

A side-by-side comparison of the two tickets purchased by Hebert suggests she had more than luck on her side.

Besides time stamps that indicate one was printed three seconds after the other, the only discernible difference between the two tickets are the tracking numbers printed on the top and bottom of each. The numbers that really matter on the tickets — all 10 assigned to Hebert, as well as 20 "Hot Numbers" she had to match — were the same — "01" — making the tickets look more like templates for tickets than the real deal.

Still, Yandow said unsuspecting folks, like Hebert, understandably believed the tickets were legitimate — a perception the commission cannot afford to ignore.

"Our players look to us for lottery games and when we have an issue like this we're concerned about it," he said. "This should not happen."

According to Yandow, the Vermont Lottery underwent a computer system conversion earlier this summer and that may have had something to do with the glitch involving the software for the "Red Hot Numbers" game that became the "Fast Play" option on June 27.

"It was operating fine, but then we had a problem," he said, explaining the issue was addressed on Aug. 4, the day it was detected and all of the questionable tickets sold.

"Basically, as the system tried to access the available 'Fast Play' numbers, it could not," Yandow explained. "As no numbers were being passed to the terminals, the system defaulted to all ones."

Although the problem was corrected, given the unprecedented circumstances, Yandow said lottery officials were reluctant to simply cut checks to those affected without thinking through the process and any future ramifications.

"This hasn't happened to us before that I know of," he said.

Times Argus

Comments

Delta Draw

Quick Picks are bogus!

I know of examples where identical tickets were sold in other states and they were losers. The system-wide error is built into the distribution and when you can have multiple winners, you can have multiple losers with identical sets of numbers in play for the same drawing. The terminals are pre-loaded with QP’s from the central office. There is only the idea that when the QP’s are sold there are no repeats of sets until the entire pool of combinations is exhausted. That is a myth and duplicate sets are sold as QP’s. That means in a sale of less than all combinations for a draw, there is a limited pool of QP’s for sale. It is evidenced time and again when small sales and multiple QP Jackpot winners win. How many multiple losers are there? Only an audit will discover that. QP’s are the worst and this system-wide failure just shows they got caught at it and have to pay. The local terminals do not make the QP combinations, they get sent in packets to the terminals and are distributed from the central office. What a scam.

DD

Something else to consider is the fact that the wining digits are all the same. That is so very unlikely- not improbable but VERY unlikely. I would wager a dollar that someone dropped a dime on what was going on and this is the result. SYSTEM WIDE.

RollChris76

This just goes to show the obvious problem I've always had with "computerized" drawings and how they can be biased.  The irony is, when the system "works" and nobody wins because of it the lottery folks don't raise an eyebrow--they're perfectly happy to just take everyone's money.  Yet when the "computerized" system "malfunctions" creating too many winners, all of the lottery officials are scrambling trying to figure out what happened and of course, nobody wants to pay out.  Well, maybe that's what's wrong with computerized drawings in the first place.

KY Floyd's avatarKY Floyd

"The terminals are pre-loaded with QP’s from the central office. There is only the idea that when the QP’s are sold there are no repeats of sets until the entire pool of combinations is exhausted."

The terminals have their own RNGs and generate QP's randomly as they're sold. That's why they're called random number generators instead of "pre-loaded number printers". Any idea that a set of numbers can't be repeated until all combinations have been used once is just the cluelessness of people who have no concept of what "random" means. The cliche that the lottery is for people who are bad at math is clearly very true for some people, and you're apparently one of them.

Delta Draw

Quote: Originally posted by KY Floyd on Aug 27, 2010

"The terminals are pre-loaded with QP’s from the central office. There is only the idea that when the QP’s are sold there are no repeats of sets until the entire pool of combinations is exhausted."

The terminals have their own RNGs and generate QP's randomly as they're sold. That's why they're called random number generators instead of "pre-loaded number printers". Any idea that a set of numbers can't be repeated until all combinations have been used once is just the cluelessness of people who have no concept of what "random" means. The cliche that the lottery is for people who are bad at math is clearly very true for some people, and you're apparently one of them.

Well I am one of them that is willing to see where the urban myth of LOCAL TERMINAL RNG's is documented and in print. Can you explain why ALL the supposed ISOLATED LOCAL RNG's produced the same Quick Picks SYSTEM WIDE? What don't you understand about the words SYSTEM WIDE? Reread the article.

Talk about cluelessness. What kind of cliche' of stupid is that? What does a knowledge of math have to do with how man-made randomness works or applies to buying a Quick-Pick? What don't you get and where do I read  apart from your post that the LOCALIZED TERMINALS do the 'random' selection from the entire pool of combinations? Talk about spin!

Show me where it is documented that people who play the lotteries are bad at math and why that is their weakness. Only P.T. Barnum had such contempt for his public as is misquoted, but maybe 'suckers' is how lottery admins view their public? If the lottery is for people who are bad at math and I am one of them as you say, WHAT ARE YOU DOING HERE? Damage control?

DD

I understand your position that buyers of Quick Picks are ignorant of math as are all players of the lottery. I agree that the buyers of QP's are ignorant but not of math skills, they are ignorant of the scam that they have bought into. You seem to support that scam.

KY Floyd's avatarKY Floyd

"Can you explain why ALL the supposed ISOLATED LOCAL RNG's produced the same Quick Picks SYSTEM WIDE?"

I'm going to go way out on a limb and speculate that each independent terminal runs the same software. When identical software has identical malfunctions the results tend to be, well, identical. The article says "As no numbers were being passed to the terminals, the system defaulted to all ones." That sounds pretty straighforward to me. As for rereading the article, if you try that yourself you may notice that there's nothing that says thee tickets for this pareticualr game were QP's or that they're generated by the local terminal. It only says that they were computer generated.

The post I responded to was clearly a general diatribe about QP's, and not limited to this particular game, and I was responding specifically to the part I quoted. This is at least the second time that you've made similar statements that clearly indicate that you don't understand how combinations will be repeated as a result of being generated in a perfectly random fashion. I don't know if your idea that all combinations would have to be used before any are repeated is something that many other people think, but it's certainly not anything the lotteries have ever said, and comletely defies the entire concept of combinations being randomly chosen. A knowldge of math is the entire foundation of understanding how randomness works and how it applies to QP's repeating combinations.

Delta Draw

Floyd that is the worst explanation from an expert I ever heard. You’re saying the software is the same on the terminals and that is true within a state. The terminals are flexible where they can be upgraded on line and be compatible for new games like Powerball or new state-wide games. No need for a technician to do it onsite. They are too cheap to pay that many people to do a cut-over state-wide. The terminals can be upgraded on-line.

You conclude that the terminals having the same software will fail the same, and can do so identically at the same time. There would be no issue at hand in that case but the same totally identical sets printed state-wide were computer selected as WINNERS IN A DRAW. The terminals did not have control or link to the computer that generated the winners. That’s two distinctly different problems that are so unlikely. It is obvious.

The terminals did not fail. The state says the failure was system-wide and does not want to pay. The only link to the terminals goes to the Central office of the lottery- a secured network. That is where the failure was and that is where the QP’s were generated along with an identical winner. The software did as it was told at the terminals and for the drawing. If the winners were self picks, this would not be news, the article clearly states the event was SYSTEM WIDE. No math skills required to observe that.

What you are saying is that no one outside of the mathematically brilliant can understand this event. Like it’s outside of their scope and stupid masses need not inquire. How wrong!

Did you pass over the word TO as in the lottery is saying the transfer was not going TO the terminals? They are implying the terminals went default stupid or rogue all at the same time with the same printouts. BS. You hold the same conclusion. BS.

QUICK-PICK: A determinate amount of QP’s get sent to the terminals and are local in a memory buffer. When someone wants a QP, they get drawn from the buffer. When the buffer gets low, the terminal requests for more. The terminal can only print what is in the buffer. We don’t want to crash the system ya know and that is why the buffer is there. QP’s fly out of the terminals where self picks do not here. If the terminals generated their OWN INDEPENDENT QUICK PICKS, there would be a delay from the time of order, to the generation, to the transfer to the CO, to the authorization from the CO to print a recorded and documented BONIFIED QP. It does not happen that way because QP’s are the solution to streamline the sales and not bog or slow down the network in high demands. They probably could not wash the buffer when they realized what was going on, but they likely did not have a clue either. The lottery is saying they could have stopped it but could not talk to the terminals (BS). EACH sale was logged (all the info including the time) and a bonified ticket was sold. The terminals can print QP’s once they are in the buffer (pre-approved and bonified from the CO).

SELF-PICK: BEFORE the ticket is printed and after it is scanned, the SELF PICK digits get sent to the Central Office. Once the SELF PICK is logged into the CO with all the data for verification, it gets printed with permission AT THE TERMINAL. Note: the state there does not say the terminals were off-line to the CO. Tickets cannot be sold without validation record be they QP, or SP. Terminals cannot print tickets when they are off line.

This is not a software malfunction. The software did as it was told. Most of their story is BS.

Based on your randomness remarks and expert mathematic skills, have you calculated the odds of this event taking in the terminal tickets printed and the same identical digits ( ALL the numbers on the ticket were identical) WON? Certainly randomness is your explanation for this event based on your expert insights and it is beyond the scope of the common clay to understand. Keep trying.

That state is saying the terminals went into a default mode (BS) and printed tickets with all ones IN ERROR. The only default was by design and that was to draw from the buffer. It likely cannot be tampered with including being erased. All the QP's in the buffer come pre-approved, documented and yada-yada.

They do not want to pay because for some strange reason, the computer that selects the winning numbers did the same. That is a software order Floyd.

Anyone buying self picks would scream to high heaven if what they ordered was not delivered.

They were all Quick Picks Floyd. Why do you spin for the lotteries? Do you really think the players are too stupid to see what is going on? Buffers in the terminals can be stacked with identical numbers and those numbers can be winners or losers. If one terminal on one side of the state has the same buffer data-the same sets, who would be the wiser? We are now Floyd.

DD

Todd's avatarTodd

@DD: I have confirmed in the past with the manufacturers of the equipment, as well as with several state lotteries, that the quick picks are indeed generated on the lottery terminal itself, with absolutely no bias or conditions being placed on it from a "central computer" or anything else.  I'm not saying that the lottery terminal uses a particularly good random number generator, but I am saying that the numbers are generated on the machine itself, with no biases, and then those numbers generated on the terminal are sent to the central computer in the same manner that your own picked numbers are sent.  Then, once the central computer records the numbers played (either QP or your own numbers), the ticket is printed.

Delta Draw

No disrespect Todd and I do count you as a credible source of information, but I will believe otherwise when the information is common knowledge and is published in each state. I do not understand how you can know this if the information is ‘proprietary’ and protected as such.

 

Floyd sez:

As for rereading the article, if you try that yourself you may notice that there's nothing that says thee tickets for this pareticualr (SP) game were QP's or that they're generated by the local terminal. It only says that they were computer generated.”

The Article sez:

“Linda Hebert says she didn't think twice when she purchased two nearly identical computer-generated "Fast Play" tickets at North End Deli Mart earlier this month.”

Your right Floyd, the article did not  refer to the “Fast Play” as QP’s. I can read better than you can spin and don’t need no higher math skills on combinations, permutations and “perfectly” random 'fashion' theory.

Floyd sez:

“This is at least the second time that you've made similar statements that clearly indicate that you don't understand how combinations will be repeated as a result of being generated in a perfectly random fashion.”

The Article sez:

"Basically, as the system tried to access the available 'Fast Play' numbers, it could not," Yandow explained. "As no numbers were being passed to the terminals, the system defaulted to all ones."

See that Floyd? Where does the context of where the ‘Fast Play’ RNG tickets comes from take you? If the fast play sets come from the CO That is where the RNG resides. Not so perfectly random is it?

Floyd Sez:

“The terminals have their own RNGs and generate QP's randomly as they're sold. That's why they're called random number generators instead of "pre-loaded number printers". Any idea that a set of numbers can't be repeated until all combinations have been used once is just the cluelessness of people who have no concept of what "random" means. The cliche (SP) that the lottery is for people who are bad at math is clearly very true for some people, and you're apparently one of them.”

I sez BS!

The terminals are only RANDOM NUMBER GENERATORS in the loosest sense of the definition. They generate random numbers yes, but ONLY at the printer. The article clearly indicates that the RNG information (the numbers, the sets) comes from the CO and are passed TO the terminals. That is where the AVILABLE Fast Play numbers reside.

DD

PS pareticualr (SP) is spelt particular.

     cliche (SP) is spelt cliché

The lottery is paying off everyone who had winners to shut them up.

Todd's avatarTodd

Quote: Originally posted by Delta Draw on Aug 29, 2010

No disrespect Todd and I do count you as a credible source of information, but I will believe otherwise when the information is common knowledge and is published in each state. I do not understand how you can know this if the information is ‘proprietary’ and protected as such.

 

Floyd sez:

As for rereading the article, if you try that yourself you may notice that there's nothing that says thee tickets for this pareticualr (SP) game were QP's or that they're generated by the local terminal. It only says that they were computer generated.”

The Article sez:

“Linda Hebert says she didn't think twice when she purchased two nearly identical computer-generated "Fast Play" tickets at North End Deli Mart earlier this month.”

Your right Floyd, the article did not  refer to the “Fast Play” as QP’s. I can read better than you can spin and don’t need no higher math skills on combinations, permutations and “perfectly” random 'fashion' theory.

Floyd sez:

“This is at least the second time that you've made similar statements that clearly indicate that you don't understand how combinations will be repeated as a result of being generated in a perfectly random fashion.”

The Article sez:

"Basically, as the system tried to access the available 'Fast Play' numbers, it could not," Yandow explained. "As no numbers were being passed to the terminals, the system defaulted to all ones."

See that Floyd? Where does the context of where the ‘Fast Play’ RNG tickets comes from take you? If the fast play sets come from the CO That is where the RNG resides. Not so perfectly random is it?

Floyd Sez:

“The terminals have their own RNGs and generate QP's randomly as they're sold. That's why they're called random number generators instead of "pre-loaded number printers". Any idea that a set of numbers can't be repeated until all combinations have been used once is just the cluelessness of people who have no concept of what "random" means. The cliche (SP) that the lottery is for people who are bad at math is clearly very true for some people, and you're apparently one of them.”

I sez BS!

The terminals are only RANDOM NUMBER GENERATORS in the loosest sense of the definition. They generate random numbers yes, but ONLY at the printer. The article clearly indicates that the RNG information (the numbers, the sets) comes from the CO and are passed TO the terminals. That is where the AVILABLE Fast Play numbers reside.

DD

PS pareticualr (SP) is spelt particular.

     cliche (SP) is spelt cliché

The lottery is paying off everyone who had winners to shut them up.

You are dissecting the language of the article too closely.  It is not written by a lottery expert, it's just a beat reporter at a local newspaper. 

My information, however, is collected first-hand by me, directly from the sources of information.  I would consider myself both a lottery expert and a computer expert, so I'm not sure who else will be more qualified to give you a solid answer on this.

The information is not proprietary.  Anyone can get answers if they are determined and dogged.

Delta Draw

Quote: Originally posted by Todd on Aug 29, 2010

You are dissecting the language of the article too closely.  It is not written by a lottery expert, it's just a beat reporter at a local newspaper. 

My information, however, is collected first-hand by me, directly from the sources of information.  I would consider myself both a lottery expert and a computer expert, so I'm not sure who else will be more qualified to give you a solid answer on this.

The information is not proprietary.  Anyone can get answers if they are determined and dogged.

Then give me a link to this information.

DD

Todd's avatarTodd

Quote: Originally posted by Delta Draw on Aug 29, 2010

Then give me a link to this information.

DD

It is first-hand info, not something on a web page.  I've posted this information previously on lottery post, so I suppose I could track down a link to my previous post(s).

KY Floyd's avatarKY Floyd

"the article clearly states the event was SYSTEM WIDE."

In a system that was running the same program on lots of identical terminals. I'm hardly an expert on programming, but the basic concept is really simple: a very stupid machine follows a lot of short, simple instructions very rapidly. The machine doesn't make any decisons of its own, but only does exactly what the instruction tell it to do. Independent machines executing identical instructions will produce identical results. Perhaps you're familiar with the issue in TN, when the RNG drawing the winning P3 numbers was set up to draw 3 unique digits instead of allowing digits to be repeated. That same mistalke could easily have been applied to some or all of the independent terminals. If that had happened it's possible that every single terminal in the whole state would only have issed numbers with no repeated digits. I suppose you'd think that system wide problem was caused by the numbers being issed by a single central computer.

"What you are saying is that no one outside of the mathematically brilliant can understand this event."

I'm sure you read better than I spin because I'm not spinning. I'm simply looking at the facts that are available and trying not to jump to conclusions. The problem you do have with reading is that you obviously have trouble understanding what you read. Understanding this particular incident doesn't require any knowledge of math and I never said it does. What I said is that your obvious lack of undertanding about math is why you dont understand why combinations on QP's repeat long before every possible combination has been used once. Understanding how that happens is actually fairly simple. It's math, but it certainly isn't complicated math. It's just math that's apparently beyond your knowledge. Your original post ranted about QP's in general, and that is completely independent of the circumstances of this incident. It's only in relation to your lack of understanding about that repitition that I've mentioned math at all.

"A determinate amount of QP’s get sent to the terminals and are local in a memory buffer. When someone wants a QP, they get drawn from the buffer. When the buffer gets low, the terminal requests for more. The terminal can only print what is in the buffer."

As Todd says, the information about each terminal having its own RNG and generating QP's locally has been presented here many times. Since you're so sure that information is wrong, why not sac up and provide some actual evidence instead of relying on the existence of repeated combinations and your ignorance of why that is not only expected, but required by the proper working of independent RNG's?  After all, it's clear that you have detailed knowledge of exactly how the process works. Surely there's a reliable source for that and it's not just something you learned while being probed by aliens in black spaceships?

"Do you really think the players are too stupid to see what is going on?"

Well, some of them are. I

truecritic's avatartruecritic

Quote: Originally posted by Todd on Aug 29, 2010

You are dissecting the language of the article too closely.  It is not written by a lottery expert, it's just a beat reporter at a local newspaper. 

My information, however, is collected first-hand by me, directly from the sources of information.  I would consider myself both a lottery expert and a computer expert, so I'm not sure who else will be more qualified to give you a solid answer on this.

The information is not proprietary.  Anyone can get answers if they are determined and dogged.

Just a thought that I had...isn't possible that the central computer was issuing some kind of test mode to the terminals and still running in that mode when the numbers were drawn?  Especially since "all ones" seems to indicate a binary response.

I would assume the central office can check (and probably does so on schedule), with all the terminals to assure they are running correctly.  A hand full of terminals may have malfunctioned and procduced the odd tickets.  The central computer seeing that response from the terminals "answered" back with an identical precoded response.

Delta Draw

TRUECRITIC:

A test of the terminal could be any kind of thing printed and is likely not in a form of a playable ticket, especially the barcode info.. Any kind of printed pattern is a binary response. The tickets were printed on demand, just like a report. A bigger question is the correlation between the “error” and the Winning set drawn.  A very unique event.

>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>> 

There are no retractions on the article that a lottery official was misquoted. The man was quoted and it stands. I say the idea that the terminals are all independent RNG’S is a myth. Here in an article the information that indicates the sets come from the CO is quoted and documented. You can’t have it both ways; either the man was misquoted by a ‘beat’ reporter (get that man his hat) or the RNG’S are local. I am more inclined to believe a quoted lottery professional than the urban local RNG myth. The article implies the RNG source is at the CO.

Full disclosure on this topic will benefit all. The only ones who have a dog in this fight that can lose are the ones who support one of two ideas. I don’t mind being corrected but the information has to include the manufacture’s the states each use for the terminals. What states disclose this information? I want to see it in print from the owners of the equipment and the makers. I cannot find that information. It is never discussed. The RNG exists but we are not fully informed, rather we are misinformed and misled with a term loosely applied. With no documentation we can say whatever we want on how things work and some will believe it.

Another thing to consider is the security and possible breaches where the terminals are not isolated from the RNG computer for the draw. Do the two RNG’s have the same qualities? Are they one? The theft of a terminal could provide an insight to the mainframe’s RNG and that would be disastrous. All the security is on the mainframe and there is no way to breech it. It sources and it sinks. There is no reason to put all the smarts in the terminals. They are relatively dumb terminals. They read, log, print and communicate to the mainframe and local cash register. There has to be isolation from the field computers and those that pick the digits for the draw. The firewall is total isolation. The RNG that produces the winning numbers is only an output- unless there is more we should know. It should have no feedback to the digits in play for the draw.

Why is this an uncomfortable topic and to be dismissed? The players think they are getting random numbers in QP’s and that is okay, but why all the repeats? Multiple winners? I know, blame the terminals- it’s all their fault. It makes perfect sense that there can be multiple winners from QP’s if the terminals are all independent RNG’s-are they? For the big games with many combinations the players think they have an equaled chance to win because their QP is THOUGHT to be unique. They are not. Multiple winners means multiple losers with identical numbers. Why sell the same ticket twice or more? Who benefits from that? Is that a fair game to enter? It is not. The biggest dog in this fight are the manufacturers and the state’s coffers.

One last point and that is if the terminals are independent RNG’s it is possible that there can be two identical sets printed independent of another at the same terminal in sequence. Not probable but possible. Why challenge that? Also at the same time, it can happen elsewhere. Why challenge that and call it a malfunction? It supports the myth real good. BUT, the lottery says (no myth!) the terminals get the sets from them and they PROVIDE them TO the terminals. You can’t have it both ways and blame the terminals for what they do not do or give them credit for what they do not do.

I’m a lottery expert too. I played, I won. As for computers, well you won’t have to look far to see a piece of equipment that has a CPU in it that was modeled on a bigger computer than you can imagine. I worked in team support of delivering the technology you have today and had years ago. You can’t walk one block and not be going past something I had a part in. How’s that for a claim to fame and credentials?

Floyd your oversimplification about what you don’t know and browbeating is a real combination. Have you worked out that math problem I gave you? Why not make your proofs from a mathematic standpoint and not just talk about it? Oh, and you were wrong about the tickets sold, they were quick picks. No Floyd, I don’t buy into the terminals everywhere are independent RNG’s. It would lock up a network today with traffic and how long have they been around? Why spend all that extra money on a network system that requires the utmost of security first for function? Read the odds info and ask yourself if the odds include that the QP pool sold is larger than what is required to win. The odds are 1:n not 1: 1/n+1/n+1/n repeating at each terminal. To guarantee the odds do not change or are corrupted, the pool of combinations sold has to be constant, not overlapping. That is what people think they are buying: 1:n. That is not what is sold when isolated RNG’s are pumping out sets with no control. The odds go up dramatically with local RNG’s. What a scam! Why the myth?

My state has MM and PB. What kind of quality control is there if all across the State and country local terminals are puking out the same sets? If you played the games, would you want the odds to be in reality greater than advertised???? If you’re a mathematician, then work it out with a pencil. Multiple RNG’s means compound and overlapping QP’s are sold for the same draw. That’s not what is advertised in the odds. Terminals do not upload QP sets to the mainframe for documentation and if a terminal goes dark there is no problem having to send someone to retrieve the information on the tickets sold so winners/sales can be verified. Washignton's 15 minute downtime before a draw is not enough time to collect all the QP's/SP's local RNG's can sell when the clock is ticking and there is no time for problems. Verification happens before each ticket is printed. The QC control of the QP's is at the CO just as it is presented in the article.

DD

Subscribe to this news story