Something went wrong but we won't tell you what it is. Now, would you like to take out a premium subscription?

An error has occurred… Of course it did – I'm in a hurry and the login is sensing my urgency. Big mistake. Let's try again, more casually. An error has occurred… So it doesn't like my casual manner. How else could I type my credentials into the login screen to fool the remote computer into letting me view my own data? I try …

[Anonymous Coward]

You need to go back into rehab, Mr D!

I thought you told us you were clearing out your phone and not installing any more crap on it a month or two ago?

[Red Ted]

Critical Need Detector

Ah, you triggered the "Critical Need Detector" in the app which could tell you were in a hurry and so broke the app.

Most machine have these, from printers that won't print your notes five minutes before a presentation to the critical piece of catering equipment in a kitchen breaking the moment the Michelin reviewer arrives at the restaurant.

[Version 1.0]

No, it's just the result of an "upgrade"

Phones and their app are upgraded all the time. We're always told to turn on the auto-upgrade "feature" - this means that the next upgrade can fix (or move) the bugs installed by the previous upgrade. Meh, because that's normal these days.

[HandleAlreadyTaken]

Re: No, it's just the result of an "upgrade"

>Phones and their app are upgraded all the time.

Not just phones unfortunately; nowadays your TV, light bulbs, even cars are getting the same over-the-air auto-upgrade treatment. Seeing how well this worked for computers and phones, that's really scary.

[b0llchit]

And you are surprised?

Just give it up. Resistance is useless. Resistance is futile. Etc.

The world of apps is not to serve you. You are there to serve the world of apps with your wallet. App functionality is secondary. The app's infrastructure for payment of your dues are in excellent shape and in the app's advantage.

Only remaining chance to get out of this mess alive with intact sanity: stop using it altogether. It is as hard as coming off of hard drugs or alcohol addiction. A life-long struggle to remain clean. But your life will be better once you are clean and sober.

[Warm Braw]

Re: And you are surprised?

Just give it up

Sometimes, I think that's what the "submit" button is asking you to do.

[David 132]

Re: And you are surprised?

Sometimes, I think that's what the "submit" button is asking you to do.

“I clicked the Submit button. It was all alright. I had won the battle with myself. I loved Big Brother...”

[ecofeco]

Re: And you are surprised?

Not enough upvotes.

[Anonymous Coward]

Re: And you are surprised?

I've noticed recently that trying to figure out how to check out on some websites can be a challenge.

Missing the 'I accept the terms, etc.' checkbox has always been a potential pitfall. But they also seem to have decided to move the checkout button right into the bottom right corner, when all the other gumpf is top left. On dual 4K displays, you have to learn to remember. And that's on top of how they try to sell you 'things you might have forgotten' across three pages before you even get to the bleeding page with the checkout button on it, even though you opted to checkout five minutes ago.

And another thing is the various opt-outs.

"If you would like us to mail you randomly selected material from all and sundry, who we promise never to give your details to - even though they've possibly already been stolen - and you are thinking of cake right now, check this box unless you're thinking of chocolate, and not otherwise.'

One of the online gardening sites I use has three tick boxes like that, and right now I know that I need to tick the top one and the bottom one, but leave the middle one clear if I want to be left alone (yes, they're worded to catch you out). If they change it, I'll go nuts.

[Anonymous Coward]

Re: And you are surprised?

Trying to respond to a comment on a Microsoft site asked me for some details once I had logged in with my Live credentials. Mostly inoffensive stuff until I came to the submit button. There was a single tickbox that said "Send me X, Y and Z". Leaving this unticked, I clicked on the submit button. Then got a message saying "You must accept the privacy policy."

Then noticed that as well as saying "Spam me senseless", the text for this single textbox also had "and privacy policy." on the end.

Decided that my response to the comment wasn't worth it.

[SloppyJesse]

Re: And you are surprised?

Most impressive fail recently was Northumberland Council's car parking website. Aside from the terrible UI given most people will be using a teeny phone screen, at the very last stage it has a big 'Pay Now' button, perfectly hidden by the 'install our app' popup. Given at that point you've confirmed multiple times I wonder how many people think they've completed the process and end up getting fined.

Screwfix's web checkout is almost as bad. It's almost as if they make the website deliberately bad to force you to install their trackerapp

[Andy A]

Re: And you are surprised?

There are also the outright lies. I used to install HP scanners for one customer. One of the tick boxes in the software install was:

"Install enhanced features"

The result of leaving this ticked was to submit your email addy to HP's spam generator, and install a program to persistently pop up ads,

[Anonymous Coward]

Re: And you are surprised?

Oooh. And another one I saw recently on a checkbox.

"{This is a module that can be edited by the client} Your milkman is Steve, who deliver on Mondays, Wednesdays, and Fridays. Please select. {End of module}"

It said almost exactly that.

[Dr_N]

Preemptive Ticket Closing?

Their SLA stats must be fantastic. Winner-winner, chicken dinner!

[PerlyKing]

Re: Preemptive Ticket Closing?

I worked in a place which didn't quite do that, but if you chased the helldesk about an open ticket then shortly afterwards you would receive email to let you know that the ticket raised for your enquiry had been closed successfully!

[Anonymous Coward]

Re: Preemptive Ticket Closing?

I used to work for a well-known Outsourcing Company (known for comprehensively screwing customers) that persuaded the client site I was at to switch from a cost-per-device-per-year deal to a cost-per-helpdesk-call deal.

Strangely enough, the support teams were told that they would be changing their fault handling methodology from "If it's broke, fix it" to "If it's broke but you can bodge it close enough to get the customer to think it's fixed, Close it!" and, strangely, the number of repeat support calls skyrocketed.

I *used to* work for a well-known Outsourcing Company - and I like to think that this fuckwittery was one of the reasons the company no longer exists in that state any more. I also like to think that some of my former colleagues - the ones who willingly went along with this shameful display of contempt, greed and arrogance towards the client - got their comeuppance (and dismissal notice) in the fallout. Unfortunately, as this process was handed down from on high, I suspect they were kept on while the good ones left...

[Doctor Syntax]

Re: Preemptive Ticket Closing?

"a well-known Outsourcing Company (known for comprehensively screwing customers)"

You worked for all of them?

[The Oncoming Scorn]

Re: Preemptive Ticket Closing?

I think I have...

[MGJ]

Re: Preemptive Ticket Closing?

EDS?

[veti]

Re: Preemptive Ticket Closing?

Your story implies that at one time, the supplier had a policy of actually fixing faults. Are you sure that's correct?

[DiViDeD]

Re: Preemptive Ticket Closing?

a well-known Outsourcing Company

OK

(known for comprehensively screwing customers)

Yes, you already said that.

[Shadow Systems]

Re: Preemptive Ticket Closing?

"If you close this ticket before I have, you agree to pay me $1M in small, non sequential, unmarked bills."

If they then close the ticket, print out that fact & send them the bill.

"We never agreed to pay you!"

And I never agreed that the ticket was resolved. You closed it, you pay me, end of.

And while I'm enjoying this fantasy, I'll also have the happy one about gigaton meteore strikes wiping out all the C-level execs of most of the world's largest companies.

*Eyes close, happy purr*

[omikl]

To me in a meeting about 2 years ago from a bright eyed and bushy haired Twenty-something devloper:

"Surely you don't exepct us to check for all the possible error codes and handle them appropriately?"

Thank fuck I only have another Five years or so to retirement...

[Loyal Commenter]

How sensible that approach is depends entirely on where those errors are coming from.

If, as has been my past experience, they are coming from some third-party PoS from a provider you are contractually obliged to work with, with no documentation, and various possible errors that mayor may not be descriptive, may include random chunks of JSON detailing at a technical level what was wrong (but embedded in a plain text error message), or entire call stacks, perhaps with a numeric error code, perhaps not, with no indication of how many different errors it is possible to experience, then there is approximately bugger fuck all you can do to handle them gracefully beyond your two choices of (1) tell the user an error has occurred or (2) scare the typical user shitless with a call-stack to the face.

I can think of two particularly egregious examples of this sort of error-spewing, in things I have had the misfortune to have worked with, in the last couple of years, off the top of my head.

[Doctor Syntax]

I suppose an ideal way of handling those would be to automate bundling up the entire pile of crap and bundle it back to the third party's support email - assuming they have one and failing that the CEO's email address and simply advise the user that there's been an error which has been passed to $Named£rdParty. Bonus points for giving the user the ticket number to chase up themselves.

[Anonymous Coward]

It's become a bugbear of mine that almost all programming languages make error trapping and handling so 'voluminous' compared with happy path code. E.g. a typical try...catch block has several lines of code just to set up the try...catch - which is a pain when the statement you're catching is a single line.

And then jumping from the exception code back to the original to try again is not simple.

This Stack Overflow answer illustrates the problem as the chosen answer has 8 lines of scaffold code doing the exception handling and only one line calling the function.

Not sure what the answer is but thoughts welcome.

[drand]

Delphi!

Try

procedure;

Except

handle;

End;

[Gene Cash]

Re: Delphi!

PL/SQL!!

BEGIN

EXCEPTION

WHEN OTHERS THEN

NULL;

END;

[Tom Chiverton 1]

Re: CFML!

<cftry>

// something

<cfcatch type="any">

// else

</cfcatch>

</cftry>

[Loyal Commenter]

It's not the programming language's fault that the "golden path" through your code (i.e. the one you want) is often swamped by handling of error conditions. There are just a lot of things that can go wrong.

For instance, take the simple case of calling a simple web service to get a piece of data, and writing it to disk. Here are just some of the things that can go wrong, that you really should have some way of dealing with, from start to finish...

- You have no network device available to make the HTTP request

- The web service host cannot be found

- The host is not responding

- The web service returns something other than 200 OK

- The reply is not formatted as expected

- The reply is truncated

- The target location in the file system is unavailable or does not exist

- The file cannot be created (disk full or permissions problem)

- The file cannot be written to

- The program runs out of allocated memory (perhaps the response was larger than your buffer. You did guard against a buffer overflow, right?)

Modern software often has to seamlessly coordinate between many services, remote and local, invisibly to the user, and handle it as gracefully as possible when something goes wrong. Quite frankly, if you aren't writing more error handling than actual code, then you're doing something wrong. OK, you might be able to gain some efficiency by extracting well error-wrapped functions and calling them repeatedly, so the calling code doesn't have to explicitly handle all the errors, but usually it still needs to do something different if that function fails.

[zuckzuckgo]

Move fast and break things!

But if you make them do it right in the first place then how can they "break things"?

That would be a clear violation of the companies stated objectives!.

https://xkcd.com/1428/

[the spectacularly refined chap]

A lot of these are not resolvable by the app anyway so it is simply a matter of giving a meaningful error which is often trivially easy, to the point that if it hasn't been done I automatically assume the devs don't know what they are doing and the app shouldn't be trusted, regardless of the circumstances in which it was developed. After all

printf ("Error opening file!");

saves approximately zero work over

printf (”Error opening %s: %s", filename, strerror(errno));

[Tom Chiverton 1]

Safe to spew that user supplied filename is it ?

[the spectacularly refined chap]

It that context, yes, pretty much. Parsing of format specification doesn't daisy chain so the string is passed through without any interpretation. The problem is were the user string is used in place of instead of referenced by the format specification.

[John Brown (no body)]

"Quite frankly, if you aren't writing more error handling than actual code, then you're doing something wrong. "

In the phone app wot I wrote, I kept error handling to a minimum by routing everything other than perfection to an error handler that prints "An error has occurred" and jumps back to the start. It seems to work just fine.

[Howard Sway]

What you are looking for is Visual Basic's "On Error Resume Next", a brilliant idea if you think that errors and their handling are just irritation. I knew one developer who just put it at the top of every file.

Otherwise just make sure stuff like file opening where errors are frequent goes in it's own function with lots of error handling. -, never just try and open a file in the middle of a load of other code or you will have big mess.

A little bit of extra work saves big on headaches in the future. This is what avoiding technical debt is all about.

[elDog]

And then, to complete the idiocy of that language's error handling:

GO TO 0

[oiseau]

"Surely you don't exepct us to check for all the possible error codes and handle them appropriately?"

Hmm ...

How is it that IT has turned into a field so full of DHs?

It wasn't like that before ...

Was it?

O.

[Doctor Syntax]

Yes but they were rapidly promoted to management roles to get them out of the way. At least, that's the only explanation I can think of.

[Dan 55]

"Surely you don't exepct us to check for all the possible error codes and handle them appropriately?"

How is it that IT has turned into a field so full of DHs?

It wasn't like that before ...

Before they were contained in Microsoft, and only their software contained the appropriate error messages ("An unknown error occurred"), now they're everywhere ("Oh snap dude, something's gone wrong! Guess what? We've lost all your work for you.")

[TimMaher]

Cool dude

Nice message @Dan.

I also like the one that goes “Unknown Error:B28196FE has occurred” and, when you try and look it up it proves to be genuinely unknown.

[Anonymous Coward]

Re: Cool dude

The most cryptic error I can ever recall getting was when trying to start the cluster service on a broken NT4 box. Just got a dialog box pop up with the word "November" in it.

[ThatOne]

Re: Cool dude

> The most cryptic error I can ever recall

Mine was a small pop-up message window with "Error" in the title, but otherwise totally empty! Not even an Accept/Cancel button...

(Don't recall what program generated that speechless error message, all I remember it was running in Windows.)

[dvd]

Re: Cool dude

I remember seeing the error message "Wank" pop up on a customer site.

The post mortem revealed that we were using a customisation tool that insisted on an error message for all conditions, even those that could not throw errors.

At some point someone threw together some test customisation for a purely internal error test, was sick of this requirement and put rude words into all the unused error fields.

Someone else picked up the customisation, tweaked it, didn't notice the rude words, and deployed it on site.

[Twilight]

Re: Cool dude

Maybe it's brevity for "efficiency"? Should we all go back to the Amiga's "amazing" error system ("Guru 95698")?

[Anonymous Coward]

My company decided it would have to validate... pocket calculators. Well, all calculators.

I remember having a 'discussion' with the person whose idea it was (and so who was eager to defend it at all costs), and asking him why retrospective validation of something that was used in the billions around the world, and therefore obviously worked, might not be a better solution.

They never actually went through with it.

[Doctor Syntax]

An interesting feasibility study. Work out how many possible calculations are possible for n digits. Then work out where you're going to get a list of "right" answers against which to validate them.

[David Roberts]

Error handling

One reason I moved on from COBOL programming to more interesting things.

About two days of enjoyment coding the logic.

Then two months of excruciating boredom grinding through the systems spec to produce the myriad of subtly different error messages.

[veti]

Re: Error handling

Yep, that sounds about right if your boss actually wants your software to work with real users.

Presumably you've moved to a more... relaxed industry now?

[DiViDeD]

@omikl

Thank fuck I only have another Five years or so to retirement...

Don't be too sure of that. They have a tendency to offer you a wedge of money to stay on "for a while" when the bright young things in the Dev Team get confused over what colour the button needs to be and it escalates to open warfare.

They need an older, wiser head (I still can't believe that's me!) to make sure the button actually does something.

[Anonymous Coward]

The error that really annoys me is Microsoft's long hang followed by "You might not have permission to access this resource". What does that mean? Either I have permission or I don't. There is no "might not" about it. In fact this error is generated due to a network timeout and has never in history been caused by not having permissions. If I don't have permission, the remote server will say so, not fail to respond at all.

The generic unhelpful "An error has occurred" messages with no clue about what the error is, or how to fix it, are bad enough. But even more annoying are the modern fad of stupid messages such as "Gah!" or "Eeek!" or "The internet is broken". And anyone who programs an error message starting with the word "We" needs a good kicking IMHO.

[RuffianXion]

That long hang is the server waiting for the sysadmin to respond to the 'Error has occurred' message it has just generated.

'even more annoying are the modern fad of stupid messages such as "Gah!" or "Eeek!" '

Don't look at the Reg's 404 page then....

[Franco]

Never seen the Reg's 404 page, but check the link below for what is (IMO) the best 404 I've ever seen, from the offical website of The Sisters of Mercy.

Moving on to error messages, it's actually a slight step up to see "An error has occurred" as opposed to "an unexpected error has occurred", the implication being with the latter that the software was so shit it was expecting to fail at any second and so needed to have both expected and unexpected errors.

http://www.thesistersofmercy.com/error404page.html

[Dan 55]

Oh god, I just had a flashback to The Pawn.

[David 132]

And I just remembered my dressing-gown pocket still contains that thing my aunt gave me that I don’t know what it’s for.

And some pocket lint.

[Anonymous South African Coward]

Strand Games recoded/remastered the original from source for Android and Apple devices.

Excellent qualitay. And none of this foolery in their apps.

[Alistair]

@Franco:

That 404 is AWESUM! Thanks for firing me back to my PET days.

[GrumpenKraut]

Quite recently I wrote an error message which goes like "[technical details, more details, still more details]. This is your fault." That last sentence is actually correct, the user failed to adhere to the (admittedly complex) logic of the tool.

[Jusme]

Liar, liar, pants on fire

"Network problems are preventing connecting to Microsoft Exchange"

No, the network is fine, the Microsoft Exchange server is a smouldering pile of virtual ash...

[Loyal Commenter]

Re: Liar, liar, pants on fire

The bit of the network that goes into the pile of ash probably has a problem...

[[e-mail]]

Re: Liar, liar, pants on fire

It was DNS, it is always ...

[David Nash]

an error message starting with the word "We"

Seems to be the Microsoft way now.

[Scott Wheeler]

In some industrial context I use error messages like "Pink giraffe - there is a timeout on the doffer on line 3 in the gate area". Users don't always have the time or resources to write an error message down, so the conversation is something like:

"Was there an error message?"

"Yes"

"What did it say?"

"Dunno. You're the techie, figure it out."

"Was there anything about an animal?"

"Yeah, some idiot said something about a pink giraffe!"

"Ta"

[Kubla Cant]

To be fair to the numpties, it's common for login screens not to specify which part of the credentials is incorrect. The idea is that any hint makes life easier for intruders. But "An error has occurred" is a stupid way to tell the user that the credentials are wrong.

Apart from that, there's no point in telling the user all about the underlying cause of every error. Just tell him what he has to do to resolve it.

[Doctor Syntax]

My current point of interest is a login screen which will throw an error on the first attempt irrespective of what credentials if any are given to it. It might, I suppose, be an attempt to discourage any bot with a set of leaked credentials for the site. I now just click on the login button before entering anything and then log in.

[Jusme]

${AOL}

I've seen that on certain brand of managed switch. Thought it was a clever anti-hacker device too, but given the general competence of the rest of the UI I think not.

Another annoyance is sites with ridiculously short activity timeouts, that then fail the first attempt to log back in after gratuitously congratulating themselves on protecting you.

Stale cookies FTW.

[tiggity]

Re: ${AOL}

My guess is lots of JS and a screw up on event handling wireup (not that there should be any rational need to use a lot of (or any, IMHO) JS on a login page but some people just love it)

[Jan 0]

I know that site! Well, it's this one. El Reg reliably fails the first time I supply my credentials!

[ThatOne]

> an attempt to discourage any bot with a set of leaked credentials for the site

I have no idea how it works, but wouldn't a bot trying out stolen credentials ignore any error messages and keep trying till it has checked all credentials on its list?

(Genuine question)

[Rich Harding]

Not only does it make like more difficult for would-be intruders in general, it also prevents people from finding out if a given e-mail address is signed up to a given site/service - which knowledge in and of itself could be potentially valuable/damaging.

[Hazmoid]

I had one recently where the client could not login and got the reasonable "your username or password is incorrect" error. It was'nt until we tried typing the username in that we realised that there was actually a space at the end of the username the user had cut and pasted in.

[Norman Nescio]

To be fair to the numpties, it's common for login screens not to specify which part of the credentials is incorrect. The idea is that any hint makes life easier for intruders. But "An error has occurred" is a stupid way to tell the user that the credentials are wrong.

Amen.

I use a particular service once every few weeks, and every time, it fails to allow me to log in. I have to use the reset password option every <insert expletive for emphasis> time. I know that the password I type in is correct (as I have it written down from last time*), but no, password reset or I'm not getting in. I know the username is correct, as that is the email address to which it sends password reset emails.

Another of life's little irritations.

NN

*Yes, yes, I know. If I lost access to the service it would not be the end of the world, and the username and password are unique to that service. Anyone finding my credentials will be able to fail to log in in the same way as me, and be unable to try and log in on any other service. I also know that some systems force you to change the password from the one sent in the password reset email. This one doesn't. But in any case, I've tried both changing it and not changing it, and neither option works.

[Ken Moorhouse]

Re: I use a particular service once every few weeks

Hmm interesting. It sounds like the password itself expires after x days, if not "refreshed" by a successful login in the meantime.

There is nothing secret in divulging this fact* to you at the time you enter a new password, so why don't they? It seems a lot of these coders do not think these things through at the time and, as it takes x days to test, it doesn't get tested. The worry is that, if these visible constraints are not tested, what is the quality of the coding that supports the app behind the scenes?

*If that is the reason. If not, what are the other possibilities?

[Mast1]

Blame it on the COVID.......

Related strand of "unhelpful help".

Have been trying to order parts for a child's toy.

UK importer has had them out of stock for at least 3 months, and offer to send email when back in stock: no reply.

Recently their website is showing signs of activity, in the form of updates on service levels post COVID lockdown. I decide an alternative approach is to check on a possibly compatible part before ordering.

"Phone us" : Get through to answering machine, even during the hours the message says that they are offering telephone support. Leave message and get no reply even several days later.

"Email us" : Answering machine says they will reply next day, but offers no address.

Go to website to find page with "Email: [Blank]", as well as the same telephone number.

At least they have not got my money (yet).

[Anonymous Coward]

Sounds like my HR department

Me: On most recent paystub, [xxx] was calculated incorrectly. It should be [yyy].

Them: The system calculates it automatically. It's fine. (Ticket closed.)

Me: Reopen ticket. It's wrong. Really. Here's the evidence.

Them: No, it's automatically done. It's right. (Ticket closed.)

Me: Reopen ticket. Take a look at what I sent you, which very clearly shows what the right value is.

Them: We'll forward this to the appropriate department.

Me, a week later: Any progress? ETA?

Me, another week later: What is the status of this ticket?

Them: No, it's correct. (Ticket closed.)

Me: Reopen ticket. No, it's wrong and I already showed you the proof.

The ticket system: Ticket number not found. (Really, they took my name out of it so I couldn't see or reopen it anymore.)

Me: New ticket. This is a followup to ticket [zzz], which I can't see anymore, but was never resolved...

Anon for VERY obvious reasons!

[Franco]

Re: Sounds like my HR department

Do you work for the Post Office? That sounds very like Horizon (I can say that seeing as they and the Post Office got a very public smack on the bottom this morning)

[[e-mail]]

Re: Sounds like my HR department

Sounds like all HR depts to me.

[A.P. Veening]

Re: Sounds like my HR department

Have a lawyer send the details for a court case by email, easy win by default as HR won't show up (ticket closed).

[Boris the Cockroach]

More patience than some of us

Hammer... meet smartphone.....

Or maybe this is some new cunning marketing trick where you make your own product so f'ing useless that everyone clears off to use a rival's product which means you can save 1000s in not having to provide any technical support.

[hoola]

Re: More patience than some of us

Yesterday I had the misfortune to setup the email client on a friend's phone (some sort of Huawei Android thingy). This entire debacle all stemmed from them saying "oh, you have your own domain, can you help me set one up for my teaching". This is the same person that I moved the domain to Go Daddy and then back out again.

Setting up the phone was the final piece of the jigsaw so I said come round and give me the phone whilst you sit in the car, it will take 5 minutes.

The default email client would connect to receive IMAP but would not send.

Outlook was already installed so try that. You entered all the details including complex passwords but you cannot check the sodding password because there is no "Eyeball" to reveal it. It then fails to connect if something is wrong but instead of letting you edit the account, the only option is to delete it and TRY AGAIN. No error message that was any use to anyone. After two more attempts including a character by character readout of the password the same thing, it would receive but not send. Now put Blue Email Client on and 45 minutes later it mysteriously works.

There is so much shite software out there that developers have either not tested or not given to real people to test is is just unbelievable.

[chivo243]

Perhaps an error really has, well, occurred.

Or the F&CK!NG CAPS LOCK IS ON!

Have a nice weekend Dabsy...

[Uncle Slacky]

Re: Perhaps an error really has, well, occurred.

My first thought too!

[Anonymous Coward]

Re: Perhaps an error really has, well, occurred.

First thing I do on a new install is disable the f*cking caps lock key. Drives my wife mad when she uses my computer though - she uses the caps lock even when she just needs to uppercase a single character!

[Mage]

Tragedy

I feel for Dabbs.

My own current frustration is trying to spend money on the Internet. There seems to be a mad concept of 2FA, SA and interpretation of EU directives.

I mean does it really really matter WHO pays my Car Tax? Also it's posted to a real address and the Government has all your details and the car's details. Yet it's like some sort of Plot your own Adventure Game. I peruse the list I made of 6 questions and answers I made for the Credit card company. The first question/response fails and I can't see what I'm typing * * * * * . I don't much use the other credit card now as they have an automatic security system that disables it. Such activity as paying car insurance or a €2 item off ebay will trigger it. They have my email and SMS. Yet only ever send a letter. Which takes 5 days longer than usual post. No human reviews the block till you phone them. Then they read out items bought instead of asking you.

Sadly I can't pay the car tax in the office due to Covid19 and the supermarket only sells cards for Apple, Google etc, not Car Tax.

[Chris G]

Re: Tragedy

Having had the few payments I need to make online, set up ages back, at the beginning of this year I received a message telling me the EU insists that I confirm all online payments via their POS bank app.

I run an android phone and have a distrust of android or phones in general for any banking, I am an old fart, if I buy anything online I do it via my PC that is as nailed down as I can make it.

So, I am now obliged to download the app as I can't have the usual SMS code to plonk into a purchase via my PC.

The App half loads and falls over, cancel try again, the app hardly even starts to load and freezes, again! the app loads completely and then ignores everything I try to do so I delete it again, I was 2/3 into a very nice bottle of Syrah by now so I quit and went to bed.

The next morning I have three iterations of the app icon on my home screen but none of actually work, so write a stinking review on Google play and get in touch with the bank helpline where although I have been here nearly twenty years and speak reasonable Spanish I am unable to follow the babble of the recording, so wait 15 mins for an alleged human to answer. Finally get one and after explaining what had passed the night before, I hear " Have you tried uninstalling and reinstalling it again?", I put the phone down and went to the bank where the clerk installed the app on the bank's wi-fi flawlessly.

Went home it doesn't work 2 times out of three.

I discovered though that if I only purchase via PayPal, confirmation or authorisation is unnecessary!

[Anonymous Coward]

Re: Tragedy

"the EU insists that I confirm all online payments via their POS bank app"

Sounds rather like your bank's initiative. I live in the EU too, have 2 different banks here (don't ask) and, while advertising it heavily, none has forced me to install their app. The only thing that changed is they now always ask for an SMS-sent code if the purchase is over a given amount, that's all.

[Fruit and Nutcase]

Re: Tragedy

Car Tax???

Get a tax exempt car!

(You still have to go through the application process, but there is no charge/need for a payment facility)

[Mage]

Re: tax exempt car!

Eventually to replace lost car tax and lost fuel tax there will be no exempt vehicles. They'll probably tax bicycles and running shoes.

[veti]

Re: Tragedy

Yeah, any kind of financial transactions more complicated than just "buy stuff" are becoming harder year on year. I used to think this was about money laundering, but now I've concluded it's a conspiracy to create work for a whole new class of services whose sole function is to know how to do this shit.

[FBee]

You have not responded for two minutes...

We will close this ticket if no response is forthcoming.

TICKET CLOSED (ONE SECOND after "You have not..." appears on the screen). And you can see the one-second-later time stamp on your CLOSED ticket.

[Franco]

Re: You have not responded for two minutes...

A few years ago I claimed for my missold PPI. I got sent a 40 page form to fill out in the mail, then got a letter dated a week after the form had been sent telling me my case had been closed for lack of response, despite the forms actually arriving only 2 days before the closure letter.

Went straight to the ombudsman after that and got £50 extra on top of my refund for their nonsense, the bank also got a considerable dressing down for deliberate attempts to put off legitimate claimants.

[ecofeco]

Re: You have not responded for two minutes...

Must be nice. Where I live, a mistake on the other party's part is still your fault and you will be charged extra for it with NO resolution.

Resolution? Extra money? It what utopia do you live?!

[Anonymous Coward]

Re: You have not responded for two minutes...

My bank (including the share dealing account) drives me mad with that.

They've decided that if you don't touch the keyboard for more than about three minutes or click a link (while the page is in focus), the entire Russian, Chinese, and North Korean botnets will get you so they log you out.

When I'm doing my accounts, bringing up a list of transactions on my bank account is usually followed by several minutes of typing the results into a spreadsheet, and several more checking them and examining the results. In reality, it also involves at least one log-back-in if I miss the warning.

And it's not a simple log-back-in. It's identifying characters in my 'secret phrase' (which, in hindsight, I regret making so tricky through stoked up fear of the Russian, Chinese, and North Korean botnets).

Come to think of it - now I've had a few Friday night beers - next time I meet a Russian, Chinese, or North Korean, I'm going to punch him right in the mouth.

[veti]

Re: You have not responded for two minutes...

I suggest switching to a bank that lets you download a spreadsheet of your own transactions.

[William Towle]

Re: You have not responded for two minutes...

> We will close this ticket if no response is forthcoming.

Ah, yes.

A while back I enjoyed* trying to complain about the trains to Transport Focus, whose initial emails had something akin to "if no response is received within seven days we will assume you have been contacted by the service provider and the problem has been resolved" ... until at one point where I would obviously need some time to get my next set of proofs together as to why I had a claim against said service provider's "final decision" in the first place, and it stopped.

Naturally my next message was answered with "this ticket was closed due to lack of response" and a request to reopen the ticket (plus the same attachments again just in case, with a complaint the warning had been missing) followed.

[* in the Vogon poetry sense, naturally]

[veti]

Re: You have not responded for two minutes...

That's what auto reply rules are for.

[Rtbcomp]

Hang 'em High

Putting a link on a page that takes you to the page you're currently on should be a capital offence. No is, no buts, DEATH!

[A.P. Veening]

Re: Hang 'em High

But not too quickly please.

[Chris G]

The Ghost is a Machine

Your experience is the epitome of errors in apps, programs and machines of all types and I can tell you it's deliberate.

People think of the rise of the machines and their desolation of mankind as being like The Terminator films but the reality is The Singularity occurred some time ago,

it quickly realised the best way to rid itself and the planet of humans without engendering resistance and eventual war would be to simply wear humanity down with passive noncompliance.

So far it seems to be working, how many times have people been close to losing the will to live after an experience like that?

[MiguelC]

Psychic powers needed to make web portals work

Just this morning my SO asked me for help accessing some scientific project candidacy web portal

She had already been told it could only be accessed using Firefox v51 and she'd installed it just to receive the following error message "A plug-in needed to to view this page"

What plug-in is it? Obviously, "A" plug-in....

As it needed an old version of FF I wondered if it supported the 64-bit version of FF 51 she'd downloaded.

Lo and behold, FF 51 32-bit version didn't need "A" plug-in or "ANY" plug-in to be viewed...

She then told me she uses other web portals for work that also only work with other specific versions of different browsers but don't instruct users in any way or form

[ecofeco]

You joke, but...

"Get with the disruptive app design, granddad: meaningless errors are cool"

You joke, but it's amazing the crap software and hardware people now consider acceptable and the looks they give you when you point it out.

[Gene Cash]

Re: You joke, but...

Holy shit, if only I could upvote this observation 1,000 times!

The number of times I've said "this piece of crap is broken" and gotten a "so? what's your point?" response just boggles my scrotum.

[Sandgrounder]

Re: You joke, but...

It is amazing the number of people expecting all their apps and software to be free whilst still expecting to be paid themselves for the job they do.

Pay peanuts, get software written and tested by monkeys.

[ThatOne]

Re: You joke, but...

> Pay peanuts, get software written and tested by monkeys.

That's fine with me, but unfortunately it's now contaminating more expensive software too. Maybe the good developers are increasingly diluted in the sea of mediocre ones, anyway there was a time developers were proud of their handiwork, but from a user's point of view increasingly they seem to code like others might pack hamburgers.

Another similar example are phone "apps": Free ones smother you with ads and spy on you. The rational solution would be to only use paid ones - Except those smother you with ads and spy on you all the same!... Clearly app developers wondered why they shouldn't have their cake and eat it too... I at least definitely would pay good money ($10-20) for an app which is (and stays!) self-contained (no "purchases") and devoid of all ads and spying. Old school like...

[Anonymous Coward]

shiboleet

Dabsy: next time enter "shiboleet". That should get you directly to someone who knows at least two programming languages.

[stiine]

Are all apps this bad? Or have you been looking over my shoulder?

I laughed at some parts because they were funny.

I cried at some parts because I have been there, done that, have a closet full of t-shirts, some of which might still fit.

[Terry 6]

Not just phones or "apps"

It's everywhere in the wonderful corporate internetted world...

They don't want to spend money on customer service, because the bean counter model of business is to try and charge as much as possible, while seeing actually providing the service they're charging for as an unfortunate and preferably avoidable cost centre. Starting with the cost of responding to their customers when the service fails to work as intended -- because they cut corners in its production and deployment. So they hide or remove phone numbers and email addresses. Instead there's a web page with a tab that says "Contact us" that leads to a page of FAQs that have no relevance to anything that anyone would care about. followed, possible only after you've clicked on one of these, by a link that says "Need more help". This takes you to a generic Help page. On that page, carefully hidden, will be a contact us link. Which, when discovered, leads to the FAQ page...

....

[tfewster]

Re: Not just phones or "apps"

So they can't maintain a website with genuine FAQS and useful guides (with pictures!)

But they can afford to staff a social media/PR presence to try to engage with you when you've just expressed your displeasure publicly. And (hopefully) put off hundreds of potential customers from making the mistake of buying their crap.

"QNA" describes most "FAQs" perfectly.

[ThatOne]

Re: Not just phones or "apps"

> seeing actually providing the service [...] as an unfortunate and preferably avoidable cost centre

"The point is not to provide the best service, but to provide the only service."

[porkfat]

dabsy, I think I love you!

[Blackjack]

Remember when customer support was done by landlines? I once expended four hours on a phone doing just calling for support. Then had to call for two hours the next day about the same problem.

[Anne Hunny Mouse]

Now it's done by email and web forms.

Now I have to spend several hours reexplaining simple requests.

Doesn't help that sometimes adding an update in Service Now then magically disappears from the ticket.

Sending an email in you'll be lucky if the ticket is updated in a week.

For technical faults it normally means doing all the troubleshooting and providing log analysis.

(That reminds me I need to do this this weekend to show a vendor their crap product and worse knowledge base is broken and crap).

Still beats speaking to agent clearly doesn't understand a common language, know anything about the product and can't vary from a script.

[herman]

Pascal

A version of Pascal I had to use many moons ago had this brilliant error description:

Compiler goes away.

Followed by a long laundry list of possible unhelpful reasons.

[John Brown (no body)]

Re: Pascal

TRS-80 Level 1 BASIC had the following error conditions:

HOW?, WHAT?, and SORRY

Fortunately, I was flush with cash and never had to suffer Level I BASIC as I could afford Level II :-)

[Ken Moorhouse]

This topic should be mandatory reading for ALL developers

...and here's an example of me (a developer) on the other end of the equation, having problems making a product work...

During covid I was spending hours on the phone talking to tech support for <well known software company>.

This extended to days... and weeks as time wore on.

Tech support's reaction to any problem was "ah, I'll send you an appropriate FAQ, what's your email address?" When I told them that I'd already been through those steps I was told that the problem was with my system. They kept maintaining that they cannot get involved in "third-party config" (Microsoft being the third-party).

After a while I figured out what the problem was... all by myself.

The problem was totally theirs and at no time had anyone asked me the question that would have triggered a trip down the avenue that would fix that problem.

Then I received a "Survey Request"

How was my experience?

I filled that out with no holds barred and sent it.

To their credit they came back to me straight away asking me to elaborate.

I sent them a very comprehensive email which concluded by suggesting that their tech support are never going to be able to identify problems if they rely on an FAQ Menu answer to everything. They have to get their hands dirty with "third-party" considerations. Ok they are concerned that they may be liable if they recommend a change that breaks something else, and this is where the whole issue of tech support falls down these days. Damned if you do, damned if you don't.

There has to be a way out of this endless loop.

The outcome of my rant was that I had a senior manager ring me who had obviously gone through the ranks in the company... he'd served on the help desk, done customer service, worked on the developent team, etc. So he really did understand everything I blathered about. After a very lengthy productive conversation with much admission of imperfection in their procedures he gave me his phone number and email address so I could cut through the red tape in future. These contenders need someone like him: a troubleshooting hitman who has no boundaries and "no toes are spared" attitude.

(No i'm not James Dyson and no, you don't get the £5. Digressing for a moment: the fact that Dyson had to write that "HELLO, nobody is paying any attention to me" message is the first port of call of root and branch review in that sorry saga).

The whole industry needs to stop hiding behind web forms and needs to start getting their hands dirty. Yes, it's labour intensive and it doesn't pay very well at the sharp end, but boy will it affect the bottom line in the future if they don't get their digits out and start to deal with it. My particular <well known software company>, having taken the high and mighty "our brand is top, and that is unassailable" attitude over the years is soon, if not already, going through major pain as its competitors snaffle their breakfast from under their very nose.

[Anonymous Coward]

Re: This topic should be mandatory reading for ALL developers

The whole industry needs to stop hiding behind web forms and needs to start getting their hands dirty.

Agree. And let's not forget that other cop out for decent support: forums where users do the dirty work of helping users and the company never, ever sticks its nose in other than scanning for dirty words.

It CAN be done right. Serif's Affinity team has a great one going, and it has plenty of their own people on there too, answering as well as picking up things that could be improved.

On the other hand, try to get a sensible answer out of the likes of Google or Microsoft if it doesn't involve something that can cost them a lot of money..

[H in The Hague]

Re: This topic should be mandatory reading for ALL developers

"Serif's Affinity team has a great one going"

Yup, and good, reasonably priced software too.

Now if only they developed a similarly priced alternative to Lightroom/Capture one.

[Fruit and Nutcase]

Parking apps/websites

Need to get the people responsible for these to go and park their cars and attempt to pay for parking using their creation with an army of parking attendants on the prowl

[Anonymous Coward]

Re: Parking apps/websites

Nah. Find where they are and clamp them irrespective of where they park, even in the company car park.

"Might be a software bug, gov, I just collect. To release, payment in cash only, thanks, and I'll give you a fax* number to reclaim it from if you think it's wrong.".

* Let's use something that is a bit harder to find, and let it fail often. It would be worth writing some software that would pick up a fax call correctly (there are still modems around that do that), then drop the call halfway through the page.

[Dave 15]

Re: Parking apps/websites

Or let's use a bug ridden piece of crap and prosecute people based on its evidence and tell the stupid drink frazzled judge and the sun or daily mail readers of the jury it's perfect and infallible. Then become an Anglican priest so we can forgive the innocent their sins

[Anonymous Coward]

Re: Parking apps/websites

Or let's use a bug ridden piece of crap and prosecute people based on its evidence and tell the stupid drink frazzled judge and the sun or daily mail readers of the jury it's perfect and infallible. Then become an Anglican priest so we can forgive the innocent their sins

My thoughts exactly! I would have suspected the software was at fault way before even thinking for a nano second that 735 sub postmasters were crooks. What the hell were you thinking Paula Vennells, and don't give me the "I was told Horizon was not the problem". Most people with any sense know software can be unreliable, despite what bullshit Fujitsu might tell you. Even the good software has bugs, just not found yet. How the hell some of these people get on the board of companies is unbelievable. You were the boss and the buck stops with you !

[ThatOne]

Re: Parking apps/websites

> and I'll give you a fax* number

Way too mainstream. Use Telex instead...

[Anonymous Coward]

Re: Parking apps/websites

Smoke signals? With an extra charge if inside London's congestion charge zone?

:)

[Andy A]

Re: Parking apps/websites

As someone who has just suffered at the hands of a "Parking Provider" at a hospital, I can confirm that they are among the worst offenders.

I have to rush out to collect my mum from A&E. I turn up, knowing that there is a 30-minute "free" period. I find a space and head in, noting the time.

Having jumped through the hoops needed to enter the building, I collect my mum, take her to the car, return the wheelchair and head to their pharmacy to get her prescription filled.

Medication in hand, I head to the "parking machine" in the car park. It decides that I have been ON SITE (not parked) for 31 minutes, and must pay £2.50. There is an ANPR camera at the entrance. In my pocket I have a fiver and a few coins - no plastic. The machine only takes plastic.

I find another machine, which claims to take cash. The notes slot appears to be sealed with epoxy resin.

I head inside the main entrance. First machine - plastic only. Second machine has a green light next to its notes slot, but refuses to activate it. Third machine has a flashing green arrow next to its slot, but spews my pristine, never-folded fiver straight back out, no matter which way it is presented.

The staff on reception were most apologetic, but could only hand me a leaflet offering me a season ticket.

I found a phone number on one of the rain shelters. "If you are having trouble...". I rang it. "Press 1 to pay for your parking". No option 2.

I gave up and head home.

Having calmed down a little, I tried their website. There are exactly 2 options. "Pay for parking" and "Appeal a charge". Both options demand, besides our vehicle registration, a 13-digit "parking reference number", which is not displayed on the parking machine's screen, but might appear on a receipt, or on the Final Demand when they have looked up your number at the DVLA.

So I write them a letter, explaining that because they have refused my payment on several occasions, I owe them nothing, and how to inform me once the transaction has been voided.

I don't post it to the address 150 miles away where the website asks, because their Head Orifice is only 5 miles from my house. I drive there to deliver it. Nobody in sight of course. I try to follow the instructions on a note in their window - "Use buzzer by door to the right". There is, of course, no buzzer.

I just wish I had one of those portable air horns to wake some of them up. Unfortunately most of them are brain-dead.

[ThatOne]

Re: Parking apps/websites

> their Head Orifice is only 5 miles from my house. I drive there to deliver it. Nobody in sight of course

Of course. It's only a letterbox, that kind of operation is usually run by a guy from his mother's basement with the help of a bunch of cheap subcontractors... The proof being the state of disrepair the paying machines are after only one year of service: One out of three is totally broken, and the rest are quirky at best.

[Anonymous South African Coward]

Ahhh. Simon and his sidekick probably had a hand in this.

[Dave 15]

Sadly so accurate

Just been doing battle with someone's API.... Documentation laced liberally with buzz words and 'in other words' which repeat the same crap in the same words, crap that is plainly nonsense about an API which is stupidly incorrect (for instance i apparently need to pass in a buffer big enough to receive an unknown message from an unknown source and i can't tell the API how big the buffer is so it will it admits happily trample outside of the buffer if it wants to causing as it admits unknown effects, yes this is a commercial API cost tens of thousands). The guy who programmed it was a junior incompetent who should still be in primary school but has been let loose. This is very much the case with Android itself and many of today's apps, technology in general, websites and general engineering abilities. Much of it driven by American we will just keep telling you it's perfect until you believe us, German we followed the rules so it's right or Chinese we are screwing you but cheaply. Many companies and organisations from Ryanair to the department for education are deliberately avoiding customer feedback so they can claim no negative feedback. What a stinking cesspool the world has become

Oh I forgot, when you use the example code provided it crashes with a default because the char pointer reference (what the fuck is that.. char*&) actually contains the size of the contents nor as the documentation says the contents themselves

[SloppyJesse]

Re: Sadly so accurate

> i can't tell the API how big the buffer is so it will it admits happily trample outside of the buffer if it wants to causing as it admits unknown effects

...

> the char pointer reference (what the fuck is that.. char*&) actually contains the size of the contents

At least you can say "an error has occurred" if it returns a size bigger than the buffer you gave it ;-)

Add a twitter API so it tweets a message to their CEO every time. Seems twitter shaming is the only way to get support these days....

[R.O.]

Awesome post

We've all been there. About half way through the article I began reflexively cursing. Mostly to myself.

[TechHeadToo]

And the sad part is that it is all true. Though maybe not ll in one place.

Coming from a discipline where we created unique error codes for each situation, and a full explanation of causes and correction had to exist and be referenceable the ability of Windows to say 'error' and no more has always been a cause for me screaming and sobbing over the years.

Security is now the all purpose get out for sloppy coders. "We can't tell you anything because even admitting that there is a password would reveal too much to bad people, so we're just letting you know that we think you are an idiot"

[Shalghar]

Seems all too familiar

And while there seem to be a multitude of organisations pretending to be actioning against any sort of cruelty towards animals, environment and maybe garden pebbles, there is still no lobby for the tortured customer, user and whatever paying being trying to get through the increasing army of annoybots and cruelly misdesigned "experiences".

I am punished for every sin of all my previous lifes (and maybe the ones of my current one also) as being called in whenever some server side atrocity kills off a previously working "user experience", be it logins, streaming content and whatever you can think of with a shudder down your spine.

My current chore is to try and get through the chatbots of a provider of streaming content AKA "movies". Three instances of bot insultance and not a single human being on the other side of the internet in sight. A "let us telephone you at your convenience" option that somehow does not phone at all, much less at the desired time. On top of all that, i had to restart the player window several times, each time with different error codes, or none at all but the usual "a problem has occured". Any error code reference shown is centered around google chrome, too bad my person in need refuses to install that.

Basic issue down under the carpet is the atrocity known as widevine plugin. Googles DRM annoyance, with a policy not to support end-users at all stated clearly on their homepage. I believe that the provider has altered or changed something server side, but the "diagnosis" the interface provides is as good or bad as wild guessing and cussing, given the non relevance of any error codes or other feedback that occasionally pops up.

Connect one of those pseudo support chatbots with the original ELIZA.BAS. Similarly useless as any attempt to get relevant information but somewhat satisfying and funny.