109 comments

[ 3.2 ms ] story [ 1133 ms ] thread
(comment deleted)
For me it started this year. Once every few days I get a spam call, sometimes in languages I don't understand (chinese, etc). It's always during business hours.

I'm pretty sure my phone is in some do not call database, but that doesn't seem to help much recently.

Our phones need a "this is spam" button, and that data needs to go to the telecom, and they need to do something with it.

But isn't the incoming number easy to spoof?
It is. But I'd imagine the telecom has lower level information about that connection.
You can sometimes enroll in spam detection from your cell phone provider. For some, but not all, calls T-Mobile will tag the call "Scam Likely". The call still goes through. It's not much at all really, because I already don't answer calls from numbers I don't recognize. Still it might be useful for some marks^H^H^H^H^H folks.
This used to be a huge problem for me as well until TrueCaller came along. Spam calls have reduced to close to zero. Maybe one or two a month. The perceived downside is that one has to share their contact list with truecaller. For me personally it's a perfectly reasonable price to pay for what I'm getting in return.
America has never been a position to be 'winning' the battle against robocalls. I'm in California and currently typically get well over 10 calls a day, many with Chinese language voicemail messages. Hard to see how there is any way to stop this...
The laws changed a while ago, I hadn't gotten a single spam call in 10+ years of cell phone ownership.

In the last 6 months, it's gotten apocalyptic. Mostly from calls from from the first 6 digits of my current phone number, and most of them calling to extend my car warranty.

8-15 calls each day.

I've tried ignoring them, I've answered them, I've spoken to the calling humans (English speakers, the minute you ask for a manager or to be removed, they hang-up... often, the next call comes within 10 minutes, same people, robodialing engaged).

I called AT&T to see what options I had to minimize teh pain. Can I turn off calling entirely? No. Can I have ringing go straight through to voicemail? No, but I can make it happen after 5 seconds.

I've installed robonomo, AT&T's anti-spam calling app, Hiya, TrueCaller, etc.

They don't work well enough.

I'm on an iPhone 7. I like the 8, but not enough to upgrade. I don't like FaceID, and I don't like the notch, and I have an iPad Pro, and so I leave my iPhone on silent all the time, automatically deny any incoming calls that I notice, and have a voicemail greeting that says I don't check voicemail nor accept calls on this line.

It's that bad and annoying. I do have a Google Voice number that I give out, and which gets almost no spam calls for now. I changed that setting to announce that it's a Google Voice number, so I take the occasional business call.

But if the new iPhones continue to be engineered around FaceID, and if FaceID isn't better (and work with sunglasses and goggles and helmets) and if the notch persists, when this contract is up, I'm pretty sure I'll nix the iPhone contract and live with Google Voice on my desktop, and the iPad for mobile texting/connecting.

I hate how when they use your cell phone number as the robo call caller id and then you get text messages from strangers saying I can't take your call right now. Has happened a few times now.
You actually can turn off cellular network calling altogether, if you are willing to do that.

Dial (star)#67# (or call 611 if it doesn't show up there) to see what number your voicemail center is. Then dial (star)21(star)1(that number)#. That will automatically forward all calls, at the network level, to your voicemail.

To cancel this, dial #21#.

Thanks for that, very cool. Giving it a shot.

Are these documented somewhere? The response screen is pretty clear on what's going on here, lots of interesting things.

Just a heads up, those are likely these: http://www.chinaconsulatesf.org/eng/zytz/t1554022.htm

I've been getting them as well in California. My workplace even sent out an email regarding these since they've been happening more frequently.

I'm getting these assholes on every single number I have (home, work, two cellphone numbers). My wife as well. Multiple times a week. Often one number after the other within an hour or two.

I've never been as tempted to becoming a murderer.

the solution is easy: end unlimited calling. make unlimited plans $10/mo cheaper but start charging 1 cent for every call. legitimate companies and regular people with normal usage will have no complaints but these charges will put robocallers out of business real quick
The robocallers aren't calling from cell phones.
doesnt have to be a cellphone. they are connecting to the rest of the telephone lines somehow and do have get account to do it. Im pretty sure technicality of implementation is not that hard. Its the enforcement of something like this that people desperately need is close to impossible with the dysfunctional shit show called the government we have.
oh, you're suggesting all calls are per-minute, including landlines and VOIP.

realistically, wouldn't that logic extend to all the internet, we should all pay for traffic, to and from servers?

i think "phones" are a dying tech, and proper legal constraints and controls on the tech we pay for would be easier.

i don't see why i can't pre-approve callers, or limit inbound calls to my Favorites/VIP list. Treating phones like "telephones" is one of the problems, treating phones like tech would make solutions easier to implement.

my smartphone already blocks some numbers (but I block every one of them), so... shared block lists? just like in adblocking software
What are you using for blocking?
Block lists don't work when they can just spoof arbitrary numbers.
'Lenny' shattered the previous record of 8:37 in our office today. Had a spammer on the phone for 13:47 trying to send him one of four white papers. As you can imagine, he was very indecisive.
gah, if only I had enough spare time to troll spammers for a quarter of an hour
I'm pretty sure they're referring to Lenny the telemarketer bot. https://www.youtube.com/results?search_query=lenny+telemarke...
This is amazing hahahahaha. I love it.

Although I still feel bad for the telemarketer operator. I don't think anyone chooses that kind of job. I'd rather just not answer the phone.

Their job is actively harmful, because they exist to prey on the class of people that can't see through their bullshit. I agree that they are the low-level operators (and therefore perhaps not worth the effort to punish), but I do think it is high time we start viewing predatory sales as exactly what it is: predatory.
(comment deleted)
I am simply saying that many of the people that work as telemarketers are in that job because it was the only job available.

Would you take that job? No. Would I? No. But if it were the only job available, you bet your ass I will take it and put up with the humiliation of 90% of angry callers to support my family.

Yes, I agree that it's a scummy tactic, but I still make a distinction between their customer (the company paying them to do the job) and the actual operator.

> ... I still make a distinction between their customer (the company paying them to do the job) and the actual operator.

Nope. Lying to someone because 'my superior told me to' is still lying.

> Yes, I agree that it's a scummy tactic

So, how low are you willing to go on the "scummy tactic" scale?

They’re not all just cold callers trying to sell a legitimate product. Some telemarketers are outright scammers and they’re stealing thousands of dollars from the elderly and infirm.

Check out Hoax Hotel and Kitboga on YouTube if you want to learn about the unseemly underbelly of telemarketing. Also, talk to your grand parents and parents about these scams.

Correct, we have him on his very own extension, to which anyone can conference him in :D
A simple solution for me would be if my phone allowed me to send all calls that are not in my contacts list to voicemail directly, without ringing the phone at all. Seems like very easy functionality to implement, and yet it's just not possible at all (at least not on iOS).
I just leave my phone on silent and listen to my voicemail once a day. Works great. Friends and family know to text me if they need a quick response.
That's fine, and I do the same thing for the most part, but I don't want the call interrupting my music or popping up an "incoming call" screen over top of what I'm doing at the moment.
This is doable on iOS.

Settings > Do Not Disturb with the following settings:

- Do Not Disturb: On

- Scheduled: Off

- Silence: Always

- Allow Calls From: All Contacts

You'd lose the ability to use Do Not Disturb to silence your phone, say, overnight, but this would do what you're describing.

Personally, I've opted in to T-Mobile's spam block feature. I don't get any robocalls.

Thanks for this, just set this up.
(comment deleted)
Yes, but this also silences your notifications...
True. This can by bypassed one contact at a time by editing the contact's text tone. The text tone edit screen has an "Emergency Contact" switch that overrides DND for that contact. AFAIK there's no way to turn on app notifications.

This setup definitely isn't ideal. Just want to share the approach for people who find the tradeoffs acceptable.

FWIW, I tried your exact configuration (but added "Repeat Calls: No") and it's been really useful. I think it's actually worth the trade-off for me. Thanks for sharing! :)
This is also simple to do with Android:

Settings - Sound - Do not disturb

To activate, once the ringer is set to silent, hold down the volume down button.

Does this also prevent you from receiving non-phone related notifications like it does on iOS? I have banner notifications turned off for almost all of my apps, but for the few which I have still allowed it, I would still like to be alerted by those notifications.
No idea. I don't use any apps that push notifications.
There are iOS apps like RoboKiller that provide similar functionality.
There’s something I clearly don’t understand that hopefully someone can explain: If some customer of a phone company is making tens of thousands of ID-altered calls every day, shouldn’t that be glaringly obvious to said phone company that the customer is abusing their system?
I always thought this was VOIP calls, no phone company required.
They are VOIP calls. And many are outside the US. And it's not just the robocalls- but I can't tell you how many calls I get a week about extending the maintenance plan on my car.

And frankly, someone is buying this shit. Else they wouldn't call. I mean if 1 sale is made for every 5,000 calls they've made their money. Same with scammers. Who the hell is going to fall for a robocall from the IRS? A call that's barely in English, and comes from a local number? Obviously, somebody does.

> Who the hell is going to fall for a robocall from the IRS?

Old folks.

(comment deleted)
And, I imagine, recent immigrants who are from countries where being shaken down for cash by governmental figures is not so unusual.
> 1 sale is made for every 5,000 calls they've made their money

To stop this, the phone company should charge 10 cents to place a call. That would destroy the economics of spamming.

The same goes for email. Charge a penny to the originator to receive an email.

There was also a push in the 90s to make a kind of "postage stamp" where the cost of sending an email was CPU, which would have had the same effect of prohibiting spam by making email too expensive to spam with. ISTR the guy who came up with that idea (or one of the guys) has had an interesting series of projects since. Can't remember much beyond that, though.
Hashcash. The idea is used in Bitcoin as Proof of Work.
Too bad that didn't work out for snail mail. The post office got addicted to the income, gives them super sweet deals, so users are maximally annoyed while the post office grabs maximum profit.

It's just ridiculous that every shared mailbox (like an apartment) has a giant recycle bin next to it full of mailers.

https://www.linns.com/news/postal-updates/2017/april/congres...

I feel like more of that blame lies with the government as a whole than the post office in particular. They have to get innovative when they don't take Federal dollars, get dollars taken from them, have to fund retirement of their employees out further than other agencies, and can't close offices. It surprises me that it stays afloat despite active hindrance from the government at large.

Unfortunately once you're on the network there isn't much security to prevent spoofing. It's pretty easy to get access as well. I had a personal VoIP gateway and was able to just set the caller ID to whatever I wanted.

It's so bad that you can get into someone else's voicemail just by calling a number and spoofing the caller id to be that same number. No security whatsoever. It's pathetic.

I think the old telephone network is beyond saving and we just need to abandon it and move to VoIP services that provide better controls.

Everything already is VOIP. You haven't been able to use a modem on a phone line in fifteen years. The problem is these behemoth telecommunication companies. There's no reason at all for the phone system not to be exactly like using any sane video messenger software of the last twenty years, except for those big ass phone companies that will do anything to keep from becoming "dumb pipes" of this kind of telephony.
> You haven't been able to use a modem on a phone line in fifteen years.

I was using a dialup modem in 2013. There are also a bunch of dialup ISPs that still exist today and provide service over phone lines to people with dialup modems. Maybe what you say is true in some areas but it certainly is false as a blanket statement.

I believe you can use a modem over G711... I know for sure you can send faxes over it (I’ve done it) despite it being VoIP.
Yes, faxing is done with T.38, and faxing is actually more reliable than using a modem. I spent some years running a fax system over VOIP and trying to get it to work. It's doable (as in some special cases of using a modem over VOIP lines) but it's absolutely not doable in most other contexts on VOIP for various reasons including signal compression and packet loss, and the parent is dreaming if they think they can hook their 14.4 modem up and dump that signal onto a "telephone" line to (for example) a BBS in another timezone and have it work. The key here is non-local, not the copper lines the parent commenter is still using to dial up a local ISP. This is obviously a special case of the telephone company local to their area maintaining the copper lines for whatever purposes, keeping the copper from their house to a local ISP gateway. But Comcast and Verizon have been actively ripping copper lines out wherever they buy up local phone companies that still have them. I have asked the phone companies for an actual copper landline in every major city I've lived in over the last ten years and couldn't get one, specifically trying to make my life easier running this system I mentioned earlier. The special edge case exception is actually having a copper line to your local ISP. It's definitely not the rule. Don't believe me? Take a few minutes to call up the phone companies in cities around the country and ask for a copper phone line. Almost all of them won't have copper. And I'm telling you from years of experience that using a modem across a non-copper VOIP line is flaky in exceptionally good circumstances and impossible in most cases.
In my case I was faxing over plain G711 through a SIP to analog phone jack converter. Not saying it’s a good idea or reliable, but it worked in my case.

So the parent post could’ve very well thought he had used his modem on a real copper line while in fact it was VoIP, just the compression was light enough that it still worked fine like in my case.

No clue if I had a copper line. I didn't mean to dispute that it was or wasn't VOIP or copper, my point was only that I was using a dialup modem five years ago, in direct contradiction to the statement that they haven't worked for 15 years. This was in Chambersburg, PA through Centurylink.
Well, presumably the phone companies make money from each call so I would assume they like these spam calls as it makes them money.
You pay to access the network, but once on the network it's a bit like Internet peering. You don't pay Verizon if you call a Verizon customer from your AT&T phone. I doubt cell phone companies see anything.
They do.

They get revenue from inbound calls received from other carriers. It’s peanuts per-call but when you add up all the spam calls together in a year it’s still a hefty bag of cash.

Here's the solution that works for me:

Numbers that aren't already in my contacts list or outgoing calls list get sent straight to voicemail. I use Google Voice, so when people leave legit messages, I get a transcription and don't have to waste time listening to the messages themselves.

I get about 10-12 spam calls a day but this system works well.

Is this a feature built into Android? I also have a Google Voice number, but can't send calls that come to my cell phone number straight to voicemail (iOS).
It is a feature of Google Voice. On Android you can block individual numbers though, or there are apps to filter calls through a ratings database.
There's a Do Not Disturb option on ios, that has a sub-option to still allow calls from people in your contacts list.
Google Voice itself does pretty good spam filtering.
Does this give google all metadata on the calls you get? Do they already have this anyway if you use android?
I would be astonished if google was not recording every bit of data they could from google voice.

They do text transcription for voicemails; I'd be pretty surprised if they didn't have text transcripts of all calls that went through google voice as well.

I have similar concerns about google fiber, although I don't have the evidence to back any of it up.

Fortunately, it's still at a tolerable level on my cell. Though it does force me to use Do Not Disturb so I don't get awakened when traveling.

Landline is mostly a lost cause. I still keep one for various reasons but I mostly ignore it for inbound calls. If things get much worse I may just shut off all the ringers.

Even ignoring unknown numbers got too distracting. My phone never comes out of Do not Disturb mode unless I'm expecting an important call.

I don't know why we have such crappy cell phone audio quality and such an outdated convention. Why not have an opt-in protocol similar to a friend request on social networks before you can call someone?

That's exactly it though: PTSN will be dead soon. You barely need it except to talk to the older generation.
Right. Oh, and businesses use it too. Just as soon as we get rid of those, everything will be fine.
I appreciate the US has 10x the population, but Australia’s Do Not Call Register (https://www.donotcall.gov.au) has been fairly effective at reducing these calls (with a domestic origin at least).

It works as a central government register with a threat and imposition of large penalties for companies who disregard it.

It's international scammers that are the problem, not internal corporations. An internal registry hurts more than it helps because it provides a list of callable numbers, and chasing down scammers is way, way, way too expensive to enforce by any practical measure. Australia doesn't have this problem for the same reason OSX didn't have viruses in the early naughts. Not a big enough target.
Yeah it’s getting worse too. I’ve had the same number for about 18 years and I definitely get about 5-12 calls per day. Usually weekdays and during work hours. Our business line is awful. Constant calls all day long. About 25% robocalls
Isn't the problem the FCC regulation prohibits carriers from taking serious action like blocking robocallers?

How can this not be taken care of with gmail style spam rules?

The fcc literally had a competition to identify to robocalls.

It’s not trivial - gmail identifies spam by bayesian filtering coupled with the amount of duplicate spam sent to numerous different addresses.

Identifying spam/robocalls requires being able to parse arbitrary spoken word. People keep thinking its a matter of blocking a number, but the reality is that the security of phone networks is laughably broken with respect to number and identity spoofing. It essentially renders the id useless for identifying spam

The powers that be are not properly incentivized to battle against robocalls and in many cases are simply looking the other way.
This will be the death of general telephony.

There's no highly-evident replacement, but I suspect that what does supplant it 1. will emerge from a socioeconomic elite (as telephony, email, and mobile did, or letter-writing in an earlier era), 2. will have very cheap unwanted-contact rejection, from a cognitive-load standpoint (by the time my phone's rung or a voicemail's been left, it's too late), and 3. will probably intrinsically include multimedia -- text, likely image & video.

I'd like to see robust crypto in authentication, identity, integrity, and encryption of both data and metadata. Multople independeent peersistent, though at least partially repudiable, identities would be very nice-to-have. I have at best modest hopes.

Both elements may be undermined by an ease-of-use dynamic, though that's short-term gain for long-term cost.

In the near term, expect to see a proliferation of conflicting standards and systems.

rather than complain I'm curious what a solution would be. It seems pretty clear we need to get rid of the current phone system to fix this.
Mr. Number on Android is a free app with crowd-sourced data. It blocks spam callers.

This is only one example, but smart phones and apps like Mr. Number are going to kill the phone-spam industry.

And apps like "Record My Calls" are going to put more nails into the phone scam industry because now these criminals won't be able to escape the sound of their own voices.

Fine the telcos per call that gets through, and ramp the fine up in a predictable way, so that it becomes an existential threat after a few years.

This would immediately put an end to caller id spoofing.

(comment deleted)
It would be hard to fine them since they did nothing to encourage this behavior, nor do they have very much control in preventing it.
Yeah, but of all the actors in the system, they’re the ones in a position to do something. It’s like making banks responsible for fraudulent credit card purchases.
I wonder if I sent out 3.4 billion robo-emails how quickly the law would come knocking on my door.

From what I've read, the issue lies in the amount of money telco is making from all these calls and, if their income would drop to zero from them, robocalls would stop immediately.

Wait until

https://m.youtube.com/watch?v=bd1mEm2Fy08 Google Duplex Demo from Google IO 2018 - YouTube

Is launched.

How many business answer the phone by saying "How can I help you?", are they lying? Are they not even close to have a system like that?
Google shared they removed personal information from the calls like the names of the locations.

They already have the TTS aspect out in beta. It uses something called WaveNet.

Here this might help.

https://deepmind.com/blog/wavenet-generative-model-raw-audio...

and

"Introducing Cloud Text-to-Speech powered by DeepMind WaveNet technology"

https://cloudplatform.googleblog.com/2018/03/introducing-Clo...

Testing will start on the rest later this summer.

https://ai.googleblog.com/2018/05/duplex-ai-system-for-natur...

On the link some more examples of calls. Pretty cool stuff. Really just shows how far ahead Google is in the AI space at every layer of the stack.

The TTS at scale required the TPUs. Doing 16k cycles through a DNN in real-time is a really intensive compute task but then to offer at a competitive price to using the old way needs custom silicon.

I missed https://news.ycombinator.com/item?id=17094231

Ah I was not the only one finding it suspicious, Thank you for your links, pretty interesting research quite far away from a real product though.

This Weekend was the bay area makerfaire, where the Google stand was in front of the one of Nintendo, they were showing the exact same tech, but the Google one was really underwhelming, it was actually interesting to compare it.

I don't trust Google, ok let's wait for a product I guess.

The voice part is already in beta and on the Google Home.

https://cloud.google.com/text-to-speech/docs/wavenet

That is the part that freaked out people as you can not tell the difference between humans and computers any longer.

Google is using a DNN at 16k cycles a second in real-time. Suspect that was only possible to do at scale and at a reasonable cost with the TPUs.

They are competing with a much less compute intensive approach. But Wavenet gets a much better result.

The other aspects of Duplex will enter testing later this summer. So not long to wait.

"Google Duplex: An AI System for Accomplishing Real-World Tasks Over the Phone"

https://ai.googleblog.com/2018/05/duplex-ai-system-for-natur...

This is the type of thing that contributed me buying the Google Home. Google is just so far ahead of everyone else in AI. Really every layer of the stack.

We need people to post a bond to make a call. Maybe $1. You can't call without posting the bond. If, within 10 minutes of the end of the call, I send a text to my phone company that I didn't like the call, the bond goes to charity. You could even set something like this up as a smart contract on ethereum.
Similar ideas were once floated to stop email spam problems.

But it's untenable.

First, think of the impact it would have on lower-income people.

Second, think of how people might "game" the system. Anytime anyone is unhappy with a caller for any reason, they can falsely report the caller for spam.

Third, think of what position this puts phone companies in. They then have to be the arbiter of whether every reported call is actually spam. They'd probably have to set up an appeals process. It would be a nightmare, logistically.

Fourth, who would get to decide which charity the bond goes to? That decision alone is a hornet's nest.

I've always wished for this in every communication medium.

Charging someone for calling you would also block them permanently, so almost no one is going to do this to their friends just to mess with them. There should be no appeal: if someone wants to charge every caller, that's totally fine. People will stop calling them pretty fast. Likewise I would let the callee choose the charity, or even keep the money. All that matters is that if you are calling someone who wants your calls you won't be charged, and if you aren't, you should be. If you aren't sure, you pay your money and take your chances that you are in fact imposing on someone else.

I can't see how this would hurt anyone of any socio economic level other than spammers. If you are calling someone who doesn't want your calls, you are a spammer.

It's perfectly tenable. The phone company can pick a list of broadly acceptable charities, but hell, it's fine with me if the phone company just keeps the money. The main thing is that the caller pays. The phone company is not the arbiter, whoever receives the call is, and their decision is final. This shouldn't have any impact on low-income people unless they're calling a bunch of people who don't want to hear from them. There's no need for an appeals process, it's only $1, and it has to be reported within 10 minutes. The caller would receive a text message that they had been charged for the call, and either has to stop calling people who don't want to hear from them or pay for each call.
I was getting 1-5 three second voicemails a day. Robots being detected and sent to voicemail, then detecting and hanging up on my voicemail. I added 30 seconds of music to the end of my "please text or email me" message the voicemails have all stopped.
> I added 30 seconds of music to the end of my "please text or email me" message the voicemails have all stopped.

Everybody has different musical tastes and particularly about what one shares via a VM message, and I'm curious about what you selected. If you don't mind sharing, what segment did you cherry-pick?

Letting anybody in the world call you any time without any restrictions is absurd.

It should be possible to enforce a mode that requires the caller to send you a request for permission to call first (like contact request in Skype). And unless you explicitly granted such permission all calls from that number should be automatically rejected.

You’re talking about a system that’s over 100 years old and has been in the hands of monopoly companies for most of that time. Ajit Pai has said (surprisingly astutely) that you can solve this problem by requiring some kind of authentication for calls. Even getting that in place seems like an impossible task (though for some reason he doesn’t see how someone in his position could force that to happen). What your talking about is more difficult than that.