Coder wrote a bug so bad security guards wanted a word when he arrived at work Welcome once again to Who, Me? The Register's Monday morning feature in which we share tales of technological messes your fellow readers made, and escaped, to give you hope in case you err during the coming week. This week's hero we'll Regomize as "Trey" because back in the first decade of this millennium he was working for …
"and thousands of innocent people wilfully prosecuted."
I wonder how many of those prosecuted people have access to secret (national security stuff) microchips, capable of reading brain waves from afar - stealing thoughts/innovations from scientists around the world, induce nightmares, remote torture, sleep deprivation.
I would bet a few millions for the technology, immunity from any kind of prosecution, a lot of friends in politics and pizza with pineapples.
When I was involved in development we had a defect tracking system.
One developer used to create empty defects, and then wait. When a tester found a defect, they carefully checked the data base to see if it had been found already, and if not, then raise a defect.
The developer would be given the defect, and copy the details into his own "empty" defect, and suddenly there was a defect saying it had been discovered 2 week before.
The tester would be curious that yesterday the database didn't have a defect - and today it did.
There were suspicions that something was amiss, when there was a defect "raised" two weeks ago about code that was only written last week!
His manager had words with him, and his contract was not renewed.
Possibly to try and pretend that it was something the dev already knew about and was in the process of fixing, as opposed to it being something he'd not noticed when he possibly should have done before sending the code out? (something really obvious if he'd tested it and easy to fix but he wasn't checking the code at all).
Yes, but the developer was probably planning on people not checking. As the comment demonstrates, they weren't even smart enough to match the timestamp to a plausible discovery time, so they would also be either unaware that the history could catch them out or assuming that the people wouldn't check the history. It probably worked the first few times. If I had reported something and someone told me it had already been discovered, I'd usually believe them and be reasonably happy because they're already on it and I don't need to do anything. Only when it became a pattern would I wonder about the history.
I was in a dev group where number of bugs found by QA and the number then fixed in a timely manner was seen as an performance indicator for both parties. It wasn't long before the system was gamed, with developers leaving easy bugs in the code, in cahoots with testers knowing where to look. That bonus scheme was quickly canned.
I think the original order is correct. A schemer may try to create schemes of their own, but when someone deliberately sets up a set of incentives, often with a balancing set of negative consequences, lots of people start to scheme their way around it. If you have a schemer, you'll get their schemes. If you have your schemes, you'll find that lots of schemers start appearing, even from people who you don't think of as schemers. I've certainly been a schemer in such scenarios, finding ways to optimize odd circumstances to my benefit, though in my mind always in a way that didn't hurt the people creating the scheme.
One guy complained that measuring code quality by defects was stupid. He made his point by inserting 10,000 semi colons. In the code counting tools, a semi colon was a valid statement.
Every one was allowed a number of defects per line of code. Because of his 10,000 empty statements the tool calculated his code had negative defects overall ( I suspect this was just a bad calculation) . He made his point.
He also said it took him a week to write 10 lines of code. This was very complex code, very hot in a dispatcher, and had timing implications. His junior sidekick wrote hundreds of lines of code a day.
Again this showed the metrics were no meaningful.
A similar issue arose with the early roll-out of some smart meters. The meters would use a mobile data connection to send their readings to the central system. This used the mobile phone network, but meters were often installed in cupboards or locations with poor signal. If the meter failed to connect it would try again. If it still failed, it would wait until the following day and try again. There was a limited time window to send the data, and each failure meant the meter was trying to send more data. As you might expect, the receiving system became saturated and the bills had to be sent out as 'estimated', which was embarrassing for what was promised to be the perfect deployment of modern technology.
I'm sure it now works as intended.
That's only a problem with how they are used.
Customers who pay up-front could and should be rewarded with lower prices, not penalised.
(Mind, back in the days when it was actually feasible to keep the electricity flowing 24*7*52, I was actually hearing of "introductory rates" for newly signed-up customers paying in arrears by direct debit that were more expensive per kWh than I was paying in advance with the descendant of the company I had originally signed up with .....)
I paid up-front for all the electricity I need for the next thirtyish years, thus becoming in charge of that aspect of my life ... and then I called PG&E and told them to take their meter (and the rest of their shit) off my property, as I did not want it there anymore.
Indeed, I don't know where Jake lives, but in a decently regulated market that wouldn't be possible.
It's not so much about paying in advance, as - if they remove the meter, how is anyone going to know how much power he's using? At that point, what's to stop him setting up a weed farm and selling power (and weed) to all his neighbours?
As long as you're connected to the grid, you need a meter. End of story. Paying in advance doesn't change that.
No mill or hydro here, just PVs, battery and natural gas or propane genset for backup.
For a look at a similar to mine, completely off-grid whole-house solution, take a look at what these people are doing clear across the country from me. Note that the guy with the beard is a certified electrician, and this setup passes all applicable laws and insurance requirements for their jurisdiction. Also note that they upgraded the system mid-stream, increasing the power capability and getting rid of the fscking useless used Tesla batteries in favo(u)r of the technologically vastly superior LiFePO. They are running their entire house, with normal appliances, off their setup.
https://www.youtube.com/playlist?list=PLq7klqOTp1a9wv-2EYae6dzzf663mCvvj
Note that each homeowner's situation is different, claims vary wildly, and nobody really has all the answers.
Tesla::spit:: claims 5,000 cycles @ 96% for the Powerwall2's LiNMC battery. Somehow, I doubt this. The Lion Energy "Sanctuary" units that I use claim 6,000 @ 90% for their LiFePO4 battery. We'll see ... But I have seen claims of 10,000 cycles @ 90% for competing LiFePO4 batteries ... That could be 30 years, near enough. Again, I doubt this ...
Your guess is as good as mine as to what this translates to in the RealWorld. All I know is that I no longer have PG&E meters, and never will again.
One other thing ... even on days with heavy over-cast and rain (such as solstice to mid-January last year), my system was charging at just under 3kW. The propane powered auxiliary power plant only had to run a couple of times, for a couple hours each time. It is sized to power the entire house, while simultaneously re-charging the LiFePO bank as efficiently as possible. (I already owned the genset.)
I've no dog in this race, but out of interest, and for the information of others what is the problem with the Tesla system? I don't think LiFePO4 was generally available when Tesla got into the business so it's a bit of a first mover problem for them but these things usually balance out for the consumer. I know that if I were in the market (which I'm not) I'd care more about price, service and reliability than long term considerations. (In even ten years the situation will likely have changed beyond recognition so there's little point in putting too much effort into second-guessing it now).
"what is the problem with the Tesla system?"
I would hazard in a word "Musk." ;)
In AU the Powerwalls apparently won't (not can't) talk to other stuff (walled garden) which is raising the temperature under our regulators' collars.
Personally I would touch anything Space Karen has dicked about with. If it were a choice between a Powerwall and a truck load of sealed lead acid batteries it would be a no brainer for the truck load.
Word is that nano-coated electrodes have gotten round the major issues with using zinc-ion batteries (such as accidentally producing hydrogen!), with allegedly 10's of 000's of recharge cycles. This is - if true - Very Good News Indeed, but it will move the market for solar and batteries a lot. I hope this next-gen technology works and is out soon enough to fit in the refresh timeline for my solar battery :)
I got a smart meter two months ago.
I recently had to complain to OVO.
The meter is working. The numbers are accurate. It's reporting into OVO. They've put the readings on the Meter Reading page. They update all the time, and it correlates to the meter exactly. I have confirmed several times that they are receiving them (and, at their insistence, they then supply an accurate, up-to-date reading to check anyway).
Strangely, the last two months have consisted of entirely estimated bills some 3 x what my actual usage is, and 2 x what my estimated usage was before the installed the meter (which is working perfectly).
Basically - they just decided to make the numbers up DESPITE having the accurate, true numbers right there.
They're in full possession of the exact up-to-date data of how much I actually use (and thus I was hoping to stop the ridiculous estimates they were giving previously, which I would then "have to pay" and then refund the difference because they were wrong EVERY TIME... I took to using this inexcusable excess to buy solar panels etc. every month). And yet they make up a number not close to reality, not fitting any kind of trend of my historical energy usage with them, not even attempting to correct it when asked.
I'm on my third complaint with them. It's not a smart meter complaint. The smart meter is doing EVERYTHING it should do, has since day one, and is perfectly accurate. OVO seem to just want to make up numbers and ignore the actual, accurate, confirmed numbers entirely for billing, presumably to earn interest on the held funds until they're corrected? I'm not going to let that happen.
I don't think companies send out estimated bills because the system didn't get the data. I think they are able to profiteer from billing inaccurately if you spread them over millions of customers. It's why they don't care about your smart meter turning off. Even if you later correct it, they are "allocated" your money to cover your bill, and I reckon that means they can put that amount into investments and savings while it's being corrected and still earn interest. I bet there's some rules somewhere about it that they are playing.
My next steps (now being past the "it takes 8 weeks for us to properly bill new smart meters" period) are:
- Move company (which will involve them taking a final reading, refunding everything I've had to pay over my usage, and losing my custom without WEEKS of getting onto an accurate meter on a real, transferable tariff unlike the special one I was on before that nobody else supports).
- File a complaint against OVO with the ombudsman
Ironically, the only thing doing its job perfectly is the smart meter itself.
In your case, the smart meter was working perfectly.
A buddy of mine was having trouble with a phone company where the billing/metering system was NOT working properly.
He managed to nail it down, bluntly:
- A call started at 8:00, lasts 30 minutes. Phone call ends at 8:30. Fine.
- Another call started at 8:15, lasting another 30 minutes, ending at 8:45. - This is basically impossible to be correct. Yet, he was being billed twice over, as he had placed 2 phone calls overlapped.
It took him a full week until he could reach a human being in the ombudsman position to prove he was right. I have no idea how or why the system mas fumbling with his account exclusively, nobody did.
How long ago was that? I haven't paid for a phone call in decades.
Now that I think about it, the eras of paying for phone calls and not being able to talk to humans haven't coincided for me. The last time individual phone calls cost money it was fairly easy to talk to a human at the telco, just dialing 0 would get one on the line PDQ.
Fair enough.
I guess I was really thinking of those people who get a new contract with a "free" new iPhone, despite not using any of the power of the previous "free" iPhone. I had a friend who was like that - he probably used a couple of quids worth of calls/texts per month, but "needed" an unmetered contract, "just in case"! No wonder he was always skint!
"Strangely, the last two months have consisted of entirely estimated bills some 3 x what my actual usage is, and 2 x what my estimated usage was before the installed the meter (which is working perfectly).
Basically - they just decided to make the numbers up DESPITE having the accurate, true numbers right there."
The one thing we all learned 18 months ago when the energy companies started going bust was that they were using the credit customers had built up to fund the company itself. A glorified credit card if you will. People had thousands (lucky them) in credit with these companies which they "got back" in terms of it the credit being there with the new supplier. However, in reality, the money had gone. Spaffed away by the failed company. Ofgem, in their wisdom, decided to up the price of the standing charge to cover this in the event of a provider going bust again. So the companies themselves wouldn't be on the hook for financial mismanagement, but the consumer. Again.
So what you're describing here, to me, is a company trying to overcharge you. Obviously you provide the correct reading which is well under and they will "credit" your account with the difference. So it looks good to you. But they're just trying to get more of your money for themselves, pretending it's a credit.
When the billing is put right and you have that credit, don't leave it with the company. At all.
I used to regard this as an urban myth, but a straw poll of friends & family revealed that everyone paying monthly for their gas & electricity was "in credit" - some by £100s.
While admittedly a micro survey and at best anecdotal, it certainly has a whiff of an industry wide scam.
And if I see or hear another smart meter advert... I might not be responsible for my actions.
The adverts really annoy me. I have no great problem with having a smart meter- saves crawling around in the cupboard-under-the-stairs. And we have solar PV now, too.
But the bullshit about saving money, being more economical, is really annoying. If people want/need to use less they know how, and it doesn't take a smart meter to tell them. The meters don't even do the one useful thing that might possibly help with a little bit with that, and tell them what is using the electricity most. They don't even give information that most people would find useful
.. " .. the bullshit about saving money, being more economical, is really annoying. If people want/need to use less they know how, and it doesn't take a smart meter to tell them..."
I find my smart meter very useful. I have solar + battery storage + a low cost nightly tariff (used to charge the batteries). The smart meter lets me monitor consumption in real time so I can see if I'm dipping in to the grid at peak rate times so can balance kettle/oven/heaters/whatever if necessary (target is: Usage Now=0 on the smart meter)
[ .. per another current thread I have ESP32s/ESP8266s monitoring water temperature, humidity, solar and battery inverters, and more, all feeding a Raspberry Pi running an MQTT broker and displaying via another ESP32 so I can see consumption and status across appliances.]
[ The OVO reference hit a nerve; I've just had a smart meter fitted to MIL's remote property because of the radio teleswitch service cutoff and they still haven't sorted the bills out after nearly two months]
2 months? Luxury! Try 5 years, and still not sorted: https://forums.theregister.com/forum/all/2024/12/16/who_me/#c_4983141
"[ .. per another current thread I have ESP32s/ESP8266s monitoring water temperature, humidity, solar and battery inverters, and more, all feeding a Raspberry Pi running an MQTT broker and displaying via another ESP32 so I can see consumption and status across appliances.]"
That sounds interesting. Could you post more details or a link to the software, etc.
In fact that would make a really interesting The Register article.
Emporia Vue, a few Tasmota Devices and Home Asssistant run the office, our resin room heters, curing box, general heating and if needed the bloody great deisel generator.
Ideally next year I want to to get a handle on it all and drop the grid connection to winter/backup duties.
Its not actually hard to do and at home (I live in a pub) it was used to shave £600 a month of the monthly bill. 17th century pubs cost a fortune to run.
Yes, i'm the OP in the comment above and I was formerly a radio teleswitch customer of OVO - I've been asking for 2 years to move off it, knowing it was doomed, but they only changed it relatively recently.
The irony is - before, I couldn't change supplier. I was trapped with SSE/OVO because of that tariff.
Now I have a SMETSv2 smart meter - and could easily transfer to literally any other provider - NOW they are giving me ridiculous problems.
" the bullshit about saving money, being more economical, is really annoying. If people want/need to use less they know how"
I referred to the advert. Which has no relevance to your use case - or mine, as I've also noted on here- we also have solar PV* so need a smart meter, and in fact already had one because it's convenient.
But the adverts are promising something different.
*Not a battery as yet. I'll look at the figures after a full year of running, as provided by the smart meter, but the £4,000 it would cost doesn't seem to be cost effective, especially while were getting 15p KW/h on the feed in tariff and we're not getting any younger.
"Not a battery as yet. I'll look at the figures after a full year of running.."
I'm happy with my numbers thus far; i've had one solar inverter failure (easy DIY replacement after 9 years) and the batteries 'should' have a good life:
As a fairly early adopter (2011) I'm getting around 73p per unit for FIT + (deemed) export which brings in around £2K per annum.
I'm paying about 8p for an EV tariff for a few hours overnight to charge the batteries (15kWh).
At the moment I'm using more than I can store overnight and the solar's doing next to nothing, but within a couple of months I expect to be able to charge the batteries mostly from solar so should have a minimal off peak electricity bill along with income from the FIT so more £ in than out.
The companies have to pay in advance for gas roughly quarterly, which they then sell to the customers on a monthly basis.
They clearly do not have the money to afford to buy the gas upfront, cannot get sufficient credit and so are deliberately overbilling customers to obtain the money to continue trading. This is basically the definition of trading whilst insolvent, which is a crime.
However nothing will be done because the law only applies to us plebs and not to our lords and masters.
Is there any way to pay other than monthly there?
And I'm not "in credit" on gas, electricity, and water. The bill is always exactly what the meter says, and has been for decades. They used to send a guy around every month, now the meter sends it in, but it always matches.
Maybe that's because I don't have a "company" - it's the government, through the local utility board. I'm getting ready to switch my internet to them too, they're doing gigabit fiber now.
I used to regard this as an urban myth, but a straw poll of friends & family revealed that everyone paying monthly for their gas & electricity was "in credit" - some by £100s.
That all depends how you pay you bill and what time of the year it is.
If you opt to pay the actual cost each month, you should never build up credit, but you will pay far more in Winter than in Summer (in the UK). That's why most people opt to pay a fixed monthly payment, so credit will build up during the summer months, and reduce and even go in to debit over the winter. This does rely on you giving / them taking accurate readings of the previous years usage to calculate the fixed amount.
Yes, if you ask in September everyone will be in credit. People generally prefer consistent £250/month bills to £50 bills in July and £750 bills in December, so they'll be in credit at the start of winter in much the same way that squirrels will have an enormously positive acorn balance at the start of winter.
> ...was that they were using the credit customers had built up to fund the company itself
Not disagreeing with that in relation to most of the supply companies but one of the companies I was with for a while actually paid interest on the amount a customer was in credit by
Ok, I'm sure they were earning more than they were paying out but it did weaken the customer argument that "the money should be in my account earning me interest rather than in your account earning you interest"
A yes, that sounds like the one I had some challenges with owing me money months after I had left.
I issued a final invoice to them for the amount owed with a set of late payment / other charges on all based on what they would have charged me under their contract if I was late paying making it clear the next step would be Glasgow Sheriff's court (PITA being based in England). They agreed a goodwill payment that matched the full amount!
Now with Octopus who transferred their meters (SMETS1) onto the new system and got the 30 min export working perfectly, a step up from the old suppliers send us a photo of the export reading!
Based on that I check that companies I am dealing with come under English law, airlines are the worst for that one.
The companies themselves have no control over it, there is another quango that controls the smart meters and they would log such a request and change if it were made by the company.
If we ever go to war, or see the kind of rolling blackouts of the 70's again, then I'm sure they will manage the meters in that fashion. And I'm sure if you don't pay they can cut off your individual meter (mostly because so many people have gone out of their way over so many years to make them coming and shutting you off individually almost impossible, so they've forced smart meters so they don't have to enter your property to do so any more and can shut them down if their tamper alarm goes off).
But shutting you down without warning? I can't see it happening.
I am wondering if this could be deliberate. Once they have taken your money for the usage, the utility continues to make interest on that. When they eventually credit it back to you, sans interest, they are already ahead. Multiply this by a few thousand customers, and now you've got a nice new revenue stream!
Lee, absolutely lodge a complaint with OFGEM, although I doubt they'll do all that much. Switch to someone like Octopus who seem to have their finances in a reasonable shape, do a lot of innovative stuff to save you and them money, and their smart meters seem to be up to snuff.
I still refuse to have a smart meter though...
This post has been deleted by its author
OVO are worse than useless. I posted about my situation with them a while back, and it's STILL not resolved, despite me demonstrating the issue years ago.
I live alone (and unloved!) in a modern one bedroom flat, and they are currently trying to get me to pay £21,000, and this is for gas only (I get electricity from a different supplier)
The ombudsman was useless, by the way, we went their twice, the first time we were told "it's with evo's internal department, so there's nothing we can do". The second time, they said "we'll get OVO to recalculate the bill, and send you 200 quid and an apology".
Is that going to make up for YEARS of harassment and debt collector visits?
When we rejected their conclusion, it was basically as if the ombudsman had rejected our claim, and OVO went back into full attack mode again.
So, be wary with the ombudsman - you'll basically have to agree with whatever they say, which then absolves OVO of any related liabilities, or reject their finding, and it's back to square one.
Links:
--> My original post about OVO on this site... over a year ago!
--> Further details of my situation.
So, good luck, but be prepared!
Seriously. Public Embarrassment - it works.
Email your MP and call either the BBC, The Observer, or another media organisation with large reach and a consumer rights feature.
Your MP probably won't do anything, but they have to respond and sometimes do.
I'd start with the Beeb - the Watchdog segment or Rip-Off Britain would likely be interested, and they are reasonably effective.
The Observer also has a reasonably effective team - although they've just been sold off, so that may not continue for much longer.
Thanks., I was thinking of something like that as a next step. I didn't know of "Rip-Off Britain", so thanks for giving me that pointer.
I guess I've been slow because they haven't actually harassed me for about 2 months. - As I said, the landlord (well, one of their employees) is dealing with most of this because it affects another 3 flats in the building, and it's the landlord who gives them access to the meters which are all in a locked room downstairs.
She keeps getting a freeze put on the accounts while they sort it out (but never do) - It's lucky that whilst generally quite laid back, I'm a pretty chopsy git when in comes to injustice, and I know my rights, more or less. But I know some people would be scared shitless by some of the stunts they've pulled. (Automatically raised my direct debit to over £1,500, and then telling me to pay £6,000 a month when I cancelled it, and then sending a debt collector around (I was out, but I wouldn't have let him in anyway)
But yeah, this looks useful. I'll let you all know if I'm going to be on TV looking all downtrodden! :-)
Thanks. They are clueless. I know more or less how they've come to that figure, so I (and my landlord) have tried to explain it to them, but they don't get it.
Basically, there was a meter mix up that lasted the first 6 or so of 10 the years I've been here. They were billing me for my neighbours meter (and my neighbour was generally out 8am to midnight, so his bills were relatively low) [ As an aside, the were billing my neighbour for his meter too - not sure how that worked - we initially assumed they'd done a swap/mix up]
They always had someone check the meters, so they usually weren't estimated bills, But the issue was discovered when because of COVID, they asked me to send them a reading, because their meter-reader couldn't get out.. I don't even generally have access to the meter room, so arranged it with the landlord, and reported the "right" meter.
The figure disparity was so ridiculous that SSE (at the time) didn't believe me. I had to send numerous photos of the meter (both they and OVO seem to have collective amnesia between phone calls as if they don't record any of the information you send on the account)
Eventually they said "you're right", the meter is obviously buggered. don't worry, we'll sort it out after lockdown.
They never did. Then my account went to OVO, and SSE passed them the same ridiculous meter reading from before (as I said, total amnesia). When I phoned OVO, obviously initially, they had done nothing to be blamed for, so I was hopeful I could update them on the situation, and they would sort it out...
Again, OVO needed photos etc. and said they'd sort it.... The next thing I know is that the "correct" (but flawed anyway) reading had been put on the account, but with no details to the problem as i'd described to them.
From this point, they had me going from about 480 units to about 38,000 within the state of a month. No internal systems picked up on any issue, so they tried to bill me - and that was at the huge gas price we had during the start of the Ukraine war.
It got compounded as they used this as a monthly average usage for my subsequent bills, which is what knocked up my direct debit from 30 quid (or so) to 1,500 ! (I forget the exact figures offhand)
As I said, the landlords employee is mainly dealing with this now, and I liase through her. She did tell me about a month ago that she got through to someone sufficiently high up in the chain that when he was shown the issue he was able to say "there's no fucking way he used that much in a month. it's bloody ridiculous. I'll sort it" (I'm paraphrasing slightly, but the colourful language WAS used. This implied to me that he was someone with clout, and not just a script-reader.
However, it's been a month, and he's gone quiet, so the press is probably the next step..
I'll let you know if I'm going to be on TV!
Thanks for the reply.
Unfortunately, we've been through the ombudsman twice. They were useless. The first time they closed the case before completion, the second time, we didn't agree with their summary, and we weren't allowed to correct their mistakes, it was either "accept" their findings as given, or reject the whole thing. We rejected, and it put us back to square one.
As for OfGem, they claim claim to have nothing to do with customer complaints: https://www.ofgem.gov.uk/complain-about-your-energy-supplier-0
On a related but slightly different note, we've recently had a "smart" meter fitted by Thames Water. As with Lee, the meter appears to be working perfectly. But Thames Water have managed to ignore the decimal point and tried to charge us for cubic metres instead of litres! The bill even had a "real world" translation which helpfully told us we'd been having four showers per hour, 24/7, since the meter was installed X-(
They've dealt with it, in the usual "goodwill" way, meaning not admitting any liability.
It's just another example of shoddy systems. Wasn't there any testing which would have shown the meter readings being overread by a factor of 1,000? No sanity checks on bills suddenly leaping by the same factor? Or are they just hoping that people will pay up without questioning the infullible (sic) computer?
A family member got a smart meter with OVO a long time ago and everything was fine. For various reasons they then changed to another supplier and meter became dumb, so we had to find out how to take the readings from it. This wasn't actually a big issue, but it's a good job we did. A couple of years later they switched back to OVO and when the next bill came in it was huge....eventually I, not their useless support, worked out that they had billed for the entire time that the family member was with another supplier and for which they had already paid that supplier. Luckily OVO readily agreed that using a couple of years' worth of gas in August was not very likely, especially coupled with previous consumption, and they suspended billing whilst they worked out what to do.
They apparently then forgot about the whole thing until I queried months later as to the outcome. Then suddenly they informed me they had discovered a billing issue....oh really!
Again, not really a smart meter problem - I'm sure it reported the figures correctly.
@Lee D
the penny dropped for me a little while back when $elec_purveyor wanted us to increase our monthly payments despite us clearly having enough credit to survive next winter. Then someone on the TV said about being in debt to their supplier. I get it now.
Not only do suppliers charge through the nose for energy to people having to use pre-payment, but they are also trying to get regular customers to give them more money as a buffer. Raking it in on both sides.
No wonder so many energy suppliers have gone bust and many are technically insolvent https://www.pricebailey.co.uk/press-releases/domestic-energy-suppliers/
This shows why thorough acceptance tests and code reviews are rather essential. I also would be very hesitant to introduce my own syntax and its special-purpose checker. Why not just demand that amounts are given in accepted forms, and preferably use a tried and tested parser? The former ensures the users enter data in a format they know, the second prevents parsing errors cause the kind of mayhem in this story, like mistaking 0.01 for 100.
We once had someone add a column to our `invoices` table, not knowing that the ancient billing scripts referred to columns by their position instead of their name.
So column[6] no longer referred to the `total` field on the table. This led to us generating and sending out a few thousand invoices with a unix timestamp in the total field. You can imagine the amount of coffee that got spat out in accounts when they came in and saw the pending direct debit payments.
How do you think the Direct Debits are managed on the backend to pull from customers accounts if not via the accounts department having details of the direct debits and bank details?
FYI most payroll / accounting software literally loads the sort code, account number and amount into a CSV or equivalent for upload to Barclays, HSBC, etc. (or whoever the company banks with).
Every month for the last 10 years, I had to help accounts upload a large (several million £'s) bunch of direct debits to the bank - often by fixing CSVs (which are uploaded to a secure portal using smart cards, at least) or the details are there directly in a web page that accounts have to change on a regular basis (i.e. every time someone changes bank account or new customers join or old customers leave).
https://ciiom.barclays.com/content/dam/overseas-barclays-com/documents/important-information/help/guides/wob-bulk-payments-user-guide.pdf
"How do you think the Direct Debits are managed on the backend t"
Entirely automatically - with no input from anyone who would actually go - no that's an unreasonable DD to send.
The concept that someone would actually check direct debits in today's world is laughable. That would be customer service.
"Unreasonable" is a bit context dependent - a DD for one term at your favourite fee-paying school might be c £8k ... which is unlikely to be a reasonable amount for a homeowner's water bill.
IMO, direct debit authorities should also specify a maximum-allowed debit amount.
It can be very context dependent. Generally you want to automate as much as you can - machines make fewer (and different) errors than humans, fixing things is a pain.
I work with pensions sometimes - we don't manually check the vast majority of payments, there's tens of thousands of them daily. We do check the batch size and totals, and there's a bunch of checks that pick out things over certain values to be reviewed. And then, of course, there's all the other bills - everything from stationary to train tickets to the cleaners and tax bills. There are always exceptions, someone will need to check those.
Please tell me how that would slip by any finance department of merit when their tallies don't tally, or they see ridiculous amounts just sucked out of their accounts, or customers phone up complaining about unreasonable direct debits.
They have a duty to make sure payments are accurate, accounted for, and tally in their bookkeeping systems. As such, before any run of DD's goes out... someone literally looks at the batch in a banking interface, finance software or even the raw CSV, to ensure it looks vaguely sane.
Which includes checking that that new DD instruction from the new customer is present in the file, that the bank details they modified for a supplier have pulled through correctly, and that the amounts are sane and match what they were expecting to bill those people.
Unfortunately, Lee, there is an extraordinary amount of faith placed in financial systems... "Surely the financial system can't be wrong". Everyone thinks the system is infallible, but, if a billing script reads the wrong column, the garbage that goes in then leads to the garbage going out to the bank.
A spot check before submission to the bank is always good to have, although the question is whether that's in the processes for many companies. ISO 27001 to my knowledge requires *a process*, not necessarily a "before you make a large submission to your bank, have you spot checked that the amounts look relatively correct" step in the process. The auditors may ask whether the company would do that, and if not, why, but again, this is only if you're ISO-certified.
:-)
After complaints they'll notice, but they won't notice before they're sent out, that was what I was suggesting.
"As such, before any run of DD's goes out... someone literally looks at the batch in a banking interface, finance software or even the raw CSV, to ensure it looks vaguely sane."
And yet we still see regular reports of silly DDs, for values which absolutely aren't sane - maybe something like this (where every single DD was screwed) might get spotted, but I'm not sure I'd have any faith even in that.
I generally don't allow direct debit, for this very reason. I've read WAY too many stories of utility company mistakes resulting in $1000+ charges to customers - and then taking a few weeks (or months!) before they give the money back. "Don't worry, we'll fix it if it goes wrong" isn't the same as "we'll never charge you more than ____".
There are three options that would have spotted this. I don't find it implausible that one of them happened. In decreasing order of how ideal it is:
1. Someone manually checks a couple basic things, such as the weekly total, and when the weekly total started having a billions digit and probably a tens of billions digit, that person alerted others.
2. Some software has a feature to detect things that are unusual and flag them for review. When they noticed a customer's bill jumping by orders of magnitude, they pinged someone in the finance department to confirm this and the person alerted others.
3. Everything was automated and there were no checks until the payments got to the bank. Banks manage large amounts of money but what they probably don't manage many of is any payment that looks like a recent Unix timestamp. If you really are sending or receiving over a billion currency units in one transaction, they are probably handling that much more manually. So when they started receiving such things, they block the connection and have someone on their end review it and contact the company.
It's not surprising that this got detected. If you added 10% to every payment, you could avoid the safeguards and have a situation that only the customers notice is wrong. When you're charging billions, you don't need very good safeguards for them to go off anyway.
And yet, its embarrassing when you send out a bill for a million quid, so (as a developer of such a billing system) we have holds for irregular DD charges. Usually get charged £100, and we're charging £1000 this month? Yeah, I'd like to hold that invoice and have a human look at it please.
I see no mention of the words "DD details" in the OP's post, only "pending direct debits". That would probably be something like "account number X / direct debit reference Z, amount [insert Unix timestamp here]" with *every* account looking similar. Often that stuff is also batched, so someone checking over the batch before submitting it would also notice that that is all wrong.
I would hope the billing scripts got updated to use column names instead...
In my experience in Finance IT - stuff using column positions is old, often "voting age" old. Occasionally "kids have grown up and gone to university" old. That sort of thing can be expensive and risky to replace, not worth it if it's working right now.
It's bad practice now, absolutely, but that's because we tried it and found the problems. And hell, sometimes when it was written it was a good idea - saving on overhead in a dramatically less powerful machine could be the difference between a weekend batch and an overnight batch, or even the holy grail of "takes about 20 minutes, so we can test and re-run several times in a day".
Those two phrases come to mind when deploying anything to a "live" or "production" system. Especially where payment processing is involved.
Who validated the testing?
What change management process did it go through before deployment?
Hopefully rather more than just that manager saying "put it out there".
Testing? That's what production is for!
The usual way it goes with a particular customer of ours is
1) "hey can we have $aditional-functionality please?"
2) we build it, and test as much as we can, then install it on a test system and ask them to test it out properly
3) several months of zero testing later: "we need $aditional-functionality right away, please put it live", 'have you tested it?', "no time, please put it live now!"
It's annoying, but as we have the paper trail showing us asking them to test it (multiple times usually), we get paid to fix any problems. So I guess it works out ok for us?
I once spend a dogs age rewriting a report. I’m pretty sure it was accurate but it never went into production as the numbers ‘looked wrong’. No one could ever tell me why though.
They just kept using the old one as the numbers were nicer.
It was complete fiction. Made me wonder what they used it for. Apparently it was important though, and the financial totals it produced weren’t small.
My first contract involved a very high security factory which had no test or dev systems. All testing was done on production. <pauses to let the muppet goblins wake up and draw breath>.
I needed to make a change to the payroll application - no it was legit, not a personally approved raise.
So I called the chief payroll accountant.
Me: "I need to run a test on the payroll. I want to enroll Albert Einstein and pay him <ridiculous-quids> for a one-off test. Can I?"
CPA: "Sure. No-one would spot Albert Einstein for an employee!"
Me: "Especially as he is dead."
CPA: "Good point."
<the weekend happened>
Monday morning, an enraged CPA was on the line demanding to know who Albert Einstein was and why we were paying him <ridiculous-quids>.
Me: "But we discussed this on Friday ..."
CPA: "Ridiculous! I wouldn't have authorized ... oh, wait a minute, yes I remember now."
Me: "What's the problem? I was just about to delete old Albert from the system."
CPA: "We cut a cheque ..."
Me: "Ah. Perhaps we should not speak of this again."
CPA: "Good plan. And you'll remove Mr Einstein from the payreoll?"
Me: "Already gone."
Of course you can write a bug. I can agree with your definition that intentionally writing something that you're going to say is a bug is not "writing a bug", it's writing an intentional flaw. However, writing something that has a bug means you have written that bug. You just didn't want it there. In the same way, if you're an author and write an inconsistency in a work of fiction, you probably didn't want it but you still wrote it.
What an odd thing to claim ... Of course I can intentionally write a bug!
It might not be a bug to ME, but it'll sure be a bug to you ...
cf. ken's 1984 ACM talk "Reflections on Trusting Trust"... https://users.ece.cmu.edu/~ganger/712.fall02/papers/p761-thompson.pdf
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
