Feeds

back to article Just bought an Apple product? Need support NOW? Drop an F-BOMB

When you're caught in automated telephone-support hell, there's a magic word you can utter to ensure that you're quickly routed to an actual human being: the "F-bomb". "I was having some trouble with my (older) laptop and wanted to order a new keyboard to fix it myself," a Reddit poster writes. "After exploring every option …

COMMENTS

This topic is closed for new posts.

Page:

Anonymous Coward

Hitting # or * a few times can sometimes work too

as a way to bypass multiple layers of menus (But might not avoid any queueing though)

5
0
Childcatcher

Re: Hitting # or * a few times can sometimes work too

I've found 0 works in a slight majority of the systems I've encountered.

2
0
Silver badge

Re: Hitting # or * a few times can sometimes work too

"agent" is a good way to break out of a voice recognition system

0
0
Bronze badge

Re: Hitting # or * a few times can sometimes work too

zero is the system default for the operator in most telephone systems. It's always worth trying, i'm sure we all know how many people don't bother to change default settings!

2
0
Megaphone

Re: Hitting # or * a few times can sometimes work too

Also "representative"

0
0
Bronze badge
Big Brother

Almost anything works

Generally speaking, any voice prompted system will fall back to a human if it persistently can't understand what is being said.

Think of this as a good time for some improvisational acting - the last time I called for help with a problem on my internet, I yelled "help, help, I'm being attacked by Wildebeests". The first time, it said "I'm sorry, I didn't get that". The second time, right to a human.

1
0
Anonymous Coward

Re: Almost anything works

Boy, the joke would have been on you if the system shot back, "In the instance of Wildebeest attack, the following steps should be taken immediately... first..."

In other news, the posts here would suggest that your best option when calling support is to hit zero three times and say "fucking agent representative!"...

7
0
Thumb Up

Re: Almost anything works

I usually say You're a fucking shit robot you couldn't even pass the Turing test you transistorised pansy ass.

0
1
Bronze badge

Re: Hitting # or * a few times can sometimes work too

I've found hitting nothing, and saying nothing, usually works. They still have to have something in place for old telephones without buttons. So just don't even enter into any press this number rubbish.

0
0
Anonymous Coward

Re: "couldn't even pass the Turing test"

So it was you assclown I had on the line the other month!

0
0
Silver badge

or...

In Apples' case, use the "Call me" feature on their website, and let them figure out who you should be speaking to.

4
0

Re: or...

Tried that. I needed to speak to a higher level and all the "Call me" did was have the lower level call me and then we both got to sit on-hold for the higher level. I guess it was better than sitting on-hold for the lower level but it doesn't put you in touch with whom you really need to talk too. In fact I have never been able to get to the higher tier automatically but they do get on the phone and then let the lower level support person get the info and send it up.

1
0
Silver badge

This would only work if the queuing system kept a constant, unique, audio link from you open at all times. Whilst this is technically possible, many IVR / Autoattentant systems simply connect you to a common audio source (music on hold, comfort messages), and don't bother listening to you unless they really have to.

The resources required for speech recognition are also pretty high, so it would need a pretty beefy system to be performing speech rec on every inbound channel 100% of the time. Usually these resources are only allocated when required, i.e. when you've been prompted to say something.

So whilst it might work on some queuing systems, and the idea that an Apple system would be "spying" on you all the time isn't exactly shocking to me, I wouldn't go banking on it working in every situation.

You'll probably have more luck having a keyboard mash (don't forget * and #) and finding a get out of jail that way.

Alternatively, and preferably, just spread the news of the terrible support far and wide. This will reduce the number of customers, and hopefully you position in future queues if you haven't seen the light and taken your business elsewhere.

6
4

Apple's support line in the UK has recently moved over to a Siri-like "make you talk at it every time" system rather than the old, faster "press buttons from a preset menu that you can memorise and bypass, then give your serial number to the person who takes your call" system.

It's "better" in the same way that Dell's new "improved" Enterprise support line now requires the Express service Code and still doesn't pass that information onto whatever person gets your call. (Said requirement is made better still by the fact that the now-Dell-badged KACE system can query the BIOS for the Service Tag but - you guessed it- not the Express Service Code.... ah, fun times).

3
0

Call me back

Apple UK now have a 'call me back' system (like Amazon) which gets you a human being within seconds.

2
0
Silver badge
WTF?

Downvoted? What??

What flavour of fuckwit would give a downvote for that comment?

I spent 3 years working for a company making computer telephony kit and can assure you that this is correct. Detecting DTMF button presses and other channel monitoring is way cheaper than voice recognition. Thus all calls will typically constantly have DTMF enabled but voice recognition will often only be attached when required.

If there are excess voice recognition resources available they might be hooked to waiting calls, but as the parent says, don't count on it.

8
10
Meh

Re: Downvoted? What??

Who cares about the votes? You won't get a cookie for up votes.

4
3
Silver badge

Re: Downvoted? What??

Something you might have missed...

"Thus all calls will typically constantly have DTMF enabled but voice recognition will often only be attached when required."

It isn't hard to detect voice. I once had a tape recorder with a voice mode, I just had to remember to say "meh" before a sentence to allow the recorder time to get going. Latterly, I have used PMR radios with a system clever enough to recognise voice (as opposed to wind, bashes, etc).

Now, it shouldn't be too difficult to arrange to record continuously into a three-second in-memory cyclical buffer. When, and only when, voice is detected, transfer the previous second or so of recording into a larger buffer and append the rest of the spoken sentence. Then pass that to the voice recognition system. [*]

This will mean you can apply voice recognition to a large number of phone lines with much smaller hardware outlay. The only times it will fall over are stress conditions of everybody talking a lot at the same time, which is probably fairly unlikely to happen in reality; however if it does you could just program the system to prioritise the samples from people responding to actual questions, and toss away the random "noise" as being unimportant (let them swear a few more times until there are resources free to notice). [*]

.

* - If this has not been thought of before (hard to believe!), then I wish to state that any patent applications of this process (as of time of posting this message) must include the provision that said patent is available for implementation by anybody without restriction and without fee of any sort. In essence, a freebie. Furthermore, said patent will never be contained as part of a "patent portfolio" (because that phrase annoys me almost as much as people who say "fi-lum" for "film"). And finally, all people who register and/or use said patent must not taunt budgerigars - no exceptions.

1
4

I used to be one of the Apple support techs. It's probably not that the system doesn't pass the info on to the agent - although that's possible - it's more likely that their phone software like ours did, displays the information in a small and very easy-to-ignore line of text at the bottom of the call control panel.

When I worked there, I don't remember anyone coming through to me frustrated about the phone navigation, but plenty of people who came through frustrated about the previous agent. Usually in the Bangalore or Athens call centres. We weren't allowed to say bad things about them.

I left back in 09 to start my degree so I could get paid proper money :P

1
0
Silver badge
Thumb Up

Re: Downvoted? What??

Thank you Charles! I had wondered about the downvotes myself, but then I remembered this is El Reg, so logic goes out of the window!

As you may have guessed, I too did several years working in the computer telephony industry, this included systems with voice recognition. As you know (and other probably don't), DTMF detection is usually performed on the telephony card, it's occurrence generates an interrupt so the processing application can then service the telephony card and point it to a new prompt or transfer the call etc. Voice recognition is usually farmed out to a specialist card, or to the main CPU as required.

Having a CPU or dedicated cards capable of continually processing audio from all the 30/60/90/120 inbound channels (modern systems usually have coax/fibre connections) is just crazy. Sure it makes the programming easy, just bind all the resources one to one instead of the more usual round robin pool allocation, but blimee that's gonna cost on the hardware!

2
1
Silver badge

Re: heyrick

Sounds like your tape recorder was detecting a sound rather than interpreting it.

1
0
Silver badge

Re: heyrick

"Sounds like your tape recorder was detecting a sound rather than interpreting it."

...which is exactly what I said. The first paragraph talks about detection, the second paragraph talks about interpretation (and the third is some silliness to irk Americans ;-) ).

1
1
Bronze badge
Happy

Re: Downvoted? What??

Most IVR systems only listen for input when they need it, e.g. at a menu point when they use DTMF, grunts or whatever. As others with knowledge have said, at other times, e.g. noise on hold, no resource is usually available to listen - all noise is one way only.

[Sorry but I have yet to encounter a music on hold situation from among 'the cr*p noise on hold' universally favoured. If they were trying to listen for input they would have melted the detection unit with their own rubbish output, not to mention my discussions of their parent's past relationships.]

0
0
Anonymous Coward

"It's probably not that the system doesn't pass the info on to the agent - although that's possible..."

I work in the same building and for the same company as Apple's UK tech support, and I must say compared to the tools of the other teams based in the building (Sony, HP, P&G, Student Loans Company, Samsung), Apple's tools and procedures put everyone else to shame. They've got one, single integrated user identification, support logging, order logging, knowledgebase, sales and complaints tool. What this means is when someone calls an Apple agent their details are automatically brought up, the relevant TS documents are automatically brought up, their likely potential sales are brought up and everything's in one place so ideally (short of the problem needing escalation to Tier 2, who are based on my floor) one agent can deal with everything; identifying the user, fixing their problem, offering a sale and taking any complaints.

If an agent is asking you for a reference number you've already provided, it isn't because the information hasn't been passed on, it's actually to satisfy the Data Protection Act. People are occasionally passed to the wrong department and whathaveyer, and a reference number, like a username, coupled with something like the customer's name, is enough to satisfy the DPA in most contexts.

For a spot of fun comparison, if you call an HP agent in the same building, you have to provide your details to a "Consumer Entry" agent who logs your name and product serial number and then provides you with a reference number before passing you on. Once you're passed through to the relevant technical agent, your reference number and name are passed along with you on the telephone software, but that reference number has to be manually copied and pasted into the bit of SAP software HP use for logging, any TS documents the agent follows are actually manually retrieved from HP's public website, if you want to buy something you've got to be put on hold and passed through to another team, if you want to log a complaint you've got to be put on hold and passed through to another team.

The list of Things HP Get Wrong That Apple Get Right is about as long as you'd expect given their recent market performance.

But HP's agents get paid more and work better hours so hah.

0
0
Bronze badge

"The resources required for speech recognition are also pretty high, so it would need a pretty beefy system to be performing speech rec on every inbound channel 100% of the time."

This is only true if you are recognising free speech. If you are looking for limited dictionary of words (in the way a mobile phone or voicemail system does) then VR uses very little processing power. For instance, if you know the person is going to say a number from 0-9 then the system will become incredibly accurate and use less processing - this is why banks now often allow speech input on the phone line.

0
0
Bronze badge
FAIL

Re: Downvoted? What??

Maybe if you read the original message, you would have seen that it was about voice response and not touch tone response. Also, downvoted. Your tears sustain me.

0
1
WTF?

Your comment could be interpreted as saying something good about Apple. You ( and probably I) will be down voted shortly...

0
0

But but but... They're the same number!

It may not apply to every Dell product, but for every one I have looked at the Service Tag is the Express service code expressed in base 36.

I *assume* the reason they ask for the express service code as well is to act as a check that they have received the correct service tag.

1
0
Anonymous Coward

Re: heyrick

But then a loud click or pop on the line or some other sound which is in the human voice range would trigger it too.

0
0

Re: But but but... They're the same number!

@Whitespace:

Cheers! Had never copped to that, will check it out. If true, it explains why it's not a core KACE feature and solves the problem of not having checked it for a given machine :)

0
0

Re: But but but... They're the same number!

@ Captain Underpants

Try this. (I wrote this so long ago that I can't remember exactly what the problem was - I think I had to do a workaround because the numbers were just too big to handle)

#!/usr/bin/perl

use strict;

use Math::BaseCalc;

use Math::BigInt;

my $me = $0;

my $isbatch = $me =~ /\.bat$/i; # (This perlfile may be converted to a batchfile using pl2bat)

$me =~ s!.*[\\/]!!;

$me =~ s!\.bat$!!i;

my $command = ($isbatch) ? $me : "perl $me";

my $usage = "Usage: $command (Dell Express service code) | (Dell Service Tag)\n";

my $input = shift || die $usage;

my $calcdell = new Math::BaseCalc(digits => [0..9,'A'..'Z']); #Dell

# print $calcdell->digits, "\n\n";

my $quiet = 0;

my $esc;

my $mult = new Math::BigInt (36 * 36 * 36);

$_ = uc $input;

if ((/\-/) or (length > 7)) { #Express Service code

print "Express Service Code entered\n" unless ($quiet);

$esc = $_;

s/[\s\-]//g;

my $sc = new Math::BigInt $_;

print join("\t", $esc, $calcdell->to_base($sc)), "\n";

}

else { #Assume it's a serial number

print "System Service Tag entered\n" unless ($quiet);

my $tag = $_;

my ($tagh, $tagl) = /([0-9A-Z]+)([0-9A-Z]{3})/;

my $newval = new Math::BigInt $calcdell->from_base($tagh);

my $temp = new Math::BigInt $calcdell->from_base($tagl);

$newval *= $mult;

$newval += $temp;

print join("\t", $tag, $newval), "\n";

}

1
0

worked for me before.

after being on hold for a while i yelled 'i want to speak to a real fucking person!!!!!' and a second later i was :)

2
0
Anonymous Coward

I find the use of the term "F-Bomb" hilarious.

2
0
Bronze badge

F-Bomb, F-Bomb

Yeah, where has this stupid term come from? It just sounds ridiculous and childish.

I wish it would shit off, to be honest.

2
1
Silver badge

Re: F-Bomb, F-Bomb

"shit off"? Perhaps you meant to say poop away.

1
1
Silver badge
Boffin

Or you could try

Shibboleet

9
1
Silver badge

Re: Or you could try

Hah, the alt on that one is even better!

0
0
Silver badge
Coat

Re: Or you could try

My hovercraft is full of eels!

1
0
Anonymous Coward

F-bomb... what was that? "Fix it myself" - Apple certainly wouldn't want that now, would they...

5
0
M7S
Bronze badge

I think it was

"Free and Open Source Software"

3
0

With great power comes great responsibility

Me: "Siri, what the fuck is this article suggesting?"

iPhone: "Help me, for the love of God help me, I'm trapped in here."

0
1
Silver badge
Happy

The Reg looks forward to reading your comments ...

I've left my comment on your office voicemail. You'll probably want to deal with it quickly.

0
0
Silver badge
Meh

HSBC uses call centres n Mumbai so ...

I learnt a few choice Hindi and Marathi phrases which work as well as an Fbomb.

They must have some number recognition software on their system as after a few clicks I end up in a UK call centre.

Did you know that most Indian call centres are run /owned by a California company?

0
1
IT Angle

is it really the swearing that does it?

I've had a similar experience with a car insurance phone number and after I got annoyed with not being given an option to just speak to an operator I started talking absolute gibberish and because it couldn't figure out what I was saying it just passed me on.

0
0

Re: is it really the swearing that does it?

I doubt it does voice recognition. It may measure volume or just the presence of sound. Anything more would probably cost more than they want to pay.

1
1

Vodafone's thing will react to "I want to talk to a real person"

2
0

Eleven!!

Being Scottish voice recognition is a nightmare. I try to put on my best english accent, but it just wont work, init, guvner.

5
0
Anonymous Coward

Re: Eleven!!

Not just Scottish. A few years back an airline mislaid my wife's luggage on a trip to San Francisco. The airline's US voice recognition was completely unable to understand her clear S. English pronunciation of "San Francisco", and in the end she had to ask the hotel receptionist to get her past the first few questions of the dialogue. Heaven help any passengers who didn't even have English as a first language.

1
0
Silver badge
Happy

Re: Eleven!!

I'd prefer an honest scottish accent from the depths of Glasgow on the other end of the phone to the virgin media hell desk in India

PS If you are reading this VM.. for gawd's sake write some better scripts to follow in your hell desk

2
0
Anonymous Coward

Re: Eleven!!

Microsoft's voice activation service (whom I endearingly refer to as "Activation Bitch") has really excellent recognition for the key readback. Their job is easier given that it expects individual letters rather than whole words, but I had a hard time tricking it even when I tried. I screwed around with it once and had to become nearly unintelligible to *anyone* before it failed. Thick accents (Scottish, Russian, Australian, Alabama), lisps, exthremely thereouth lithpth, drunkennesh, random (AHEM) interruptio ACCKK TLHPBBBPPPT ns, you name it. Nothing but, "Ding! OK, now, the next block!"

"Thanks for calling the Microsoft product activation service!"

"No, thank YOU, Activation Bitch!"

Other people tend to look at you funny when you talk to the phone like that.

1
0

Page:

This topic is closed for new posts.