Why don't cell phones have a dialtone? | Hacker News Reader

Why don't cell phones have a dialtone?

dangoldin.com

47 pointsdangoldin41 comments

Nice read!

On a slightly related note, I wonder why some cell phone service providers are dishonest about a "ringer-tone" (for lack of a better word). The other day in a poor cell reception area I placed my friend's AT&T phone next to mine and dialed her number from my AT&T phone. From my phone we could hear the other phone "ring", however my friend's phone wasn't actually ringing. It then went to voicemail after a few rings.

I found it very strange since this is not how cell phones connect in India, for example. I wouldn't have heard the other phone "ring" if it weren't actually ringing on an Indian cell phone service provider.

Even if I make a call from Skype to my cell phone, I can hear the audible "ring" on Skype long before my actual phone begins to ring.

It may just be a signal to the caller that something is happening. Otherwise you're left wondering with whether the call went through or not.

I've noticed this caller-hears-ring-before-actual-ring thing, too (also on AT&T). I always presumed it was a way of giving them (the network) a little longer to make the actual connection, without the caller giving up thinking "it hasn't even begun to ring yet! Something's wrong!"

That's what this seems to be, I've even noticed that the first ring may even sound a bit different from subsequent rings.

I think the first layer ring is the actual call setup being attempted across the cellular network. Once the receiving phone is located and receives the signal that a call is incoming, you begin to hear the second layer ring.

The ring that the caller hears is basically just a recording. And yes, that's it's purpose.. so you know it's doing something before it's actually completed.

It's the telecom equivalent of the loading pinwheel.

Even on land line phones, the ring indication tone is not evidence that the remote phone is actually ringing. If I remember correctly, the tone is generated by the equipment at your central office, not at the remote location.

Original idea is that ring tone is generated by same exchange that provides the ringing voltage to analog local loop that rings. Idea there is that it tests that the whole connection is built correctly and can pass this ring tone. On the other hand this is horribly inefficient and thus often ring tone is generated on another places. For this cell phon behavior it's entirely imaginable that (G)MSC (lets just use GSM terminology) starts generating ring tone when it had found correct MS in location registry and sent paging request to it, obviously there can be significant latency between that and actual user observable ring on the terminating MS (on the order of 200ms in best case, few seconds or call being dropped in worst).

The term you're looking for is 'ring-back', and it's original use seems to have gone the way of the do-do in North America.

Most of the PBX/VoIP systems I've worked on (North and South America) provide one arbitrarily to give the caller the warm and fuzzies, regardless of what's actually happening to that call leg.

And as a mobile subscriber you can usually pay for a custom "ring-back" tone. So you can bother anyone who calls you with the latest Rihanna hit.

What if the callee has their phone on silent?

j/k

The "ringer" tone is to let you know the call is going through, and in the case of you calling a cell phone, that the network is trying to connect to your callee.

A cell phone gets a signal to play it's ringtone, not a "ring now"..."ring now"... on / off signal. Did you expect to hear my ringtone on your end?

I said 'ringer-tone' and not ringtone. Please read comments or posts carefully before commenting replies and 'j/k's.

I also recommend that you read the community guidelines here -- http://ycombinator.com/newsguidelines.html

It's just the auditory equivalent of a wait cursor.

Dial tones were always meant to be just a signal that the Caller has reached the Switch and there is an attempt to make the call happen. It was never meant to synchronize with the actual Called party's phone. Fixed line phones simply get an electrical signal to make the phone "ring"

It has to do with the codes of signalling in the various protocols such as SS7, h323 and SIP. Taking SIP as an example code 180 is call progress connecting and a signal is generated that sounds like ringing while code 183 allows you to actually hear the sound being generated on the other end, which is how they create the custom ring tones or when you will hear the international ring tone. While 183 is the actual signal the 180 code is far more efficient and is what is used by the larger carriers.

Interesting, I just heard an interview [1] with Marty Cooper (inventor of the cell phone) who is working on a simplified phone for seniors called the Jitterbug. He mentioned that the phone has a dial tone to indicate that the phone has signal.

[1] http://hereandnow.wbur.org/2013/04/09/cell-phone-inventor

I consider a dial-tone to be a waste of network resources.

Definitely. The book didn't really cover the "why" around why landlines had a dialtone but my suspicion is that the systems failed frequently enough that they wanted to signal to the user that the phone was ready for use. I may be entirely wrong.

I'm not sure, but at one time there were also a lot of party lines. A dial tone was a way of signalling that you had the line to yourself and you weren't picking up during a lull in someone else's conversation.

Today we have the signal bars to show you status. Back then, a light would have added expense and complexity to the phones of the day.

A phone switch can only accept a finite number of connections at a time, and a dial tone indicates that you have succeeded in connecting with the switch. If 100% of the capacity of the switch is currently in use (as could happen during a natural disaster), then you won't get a dial tone. We tend to take the dial tone for granted because the phone system almost always has enough capacity to let us make a call without waiting.

In case of touch tone dialing there is another aspect: classic analog or hybrid (analog call path with computer control) switch has relatively small amount of touch tone decoders (2 for 50 lines and ~10 simultaneous calls being relatively common ratio for PBXes), mechanical pulse dialing-only switches are similarly limited by number of step by step switches. Thus dialtone indicates that: (1) line is reasonably usable, (2) switch has at least theoreticaly resources to connect some call (not necessarily any) and (3) switch actually has resources to decode number you are going to dial.

I have seen hybrid PBXes where dial tone is generated by same part of circutiry that decodes DTMF.

You are right. There are two things going on here:

1. They added background noise to the actual calls to let people know they other end wasn't disconnected. The clarity got too good and people couldn't tell they were still connected. Fun fact: PCM - g.711 - is 64-kbits - hence max modem speed.

2. Dialing used to use in-band signalling and the tone lets you know you can dial (phreakers could use this to get free calls). When the dial tone ends, you know that your call is going through.

The dial tone actually ends after the first digit is dialed. However, this is a significant digit, especially for older equipment.

If you pressed 0, you would get the operator immediately. No phone numbers could begin with zero.

If you pressed 1, you would be connected to a long distance trunk and an area code would then be expected. No local prefixes could begin with 1.

If you pressed any other number it was a local call.

Also, on analog networks, maximum modem speed was 56kbps. 64k was the theoretical maximum, but the FCC placed limits on the amount of power you could have in certain frequency bands on the line -- hence 56k was the maximum speed.

On a purely digital network, like ISDN, you could receive 64kbps on each channel.

Well, you should consider an idle a circuit switched line as a waste of network resources. You get the 64-kbits/s when you are active whether you are sending data or silence.

When the mobile phone system was being designed, that's how it was thought of. But that only applies to systems where voice has dedicated radio channels.

Once you start multiplexing radio channels, or using CDMA, the idea of "connected" begins to blur, and the resources that would be dedicated to a call before it is connected decline.

Now, with smartphones on mobile data networks, there is no reason to limit the you may look up a number before dialing and that database might be partly or entirely server-based, so your "call" is using network resources before you dial.

In the near future, with LTE, it's potentially an all-IP system, if the carrier and device maker implement VoLTE, your voice packet flow is indistinguishable from whatever data bandwidth you might have been using before your voice call.

I think I'm missing something because it doesn't seem like a big deal to me.

If the first cell phone hadn't done this, what would stop a later phone from doing so? In fact, couldn't any phone buffer the numbers before sending them?

People would pick up their shiny new phone, not hear a dial tone, and return it because they think it's broken.

I don't think they would. All it would take is a one sentence explanation in the user manual and a blurb on the box claiming it as a new feature.

But that would require them to read the user manual, or the box, or pretty much anything at all. Have you met an average user before?

Here's something else that needs to change. Phones should ring TWICE AS LONG as they do before rolling to voicemail. There are so many times that I miss calls because the time required to detect the ring, locate the phone, get a handle on it and answer the call is MUCH LONGER than it was in the old days! I use Google Voice and an unholy alliance of other devices and services and do not have the ability to solve this problem!

That is why while making important business decisions it is so critical to think 10-20 years out if not 40-50 even if that makes making the tough choices.

You should read that book if you haven't already. The old AT&T stressed that type of mindset and had the researchers thinking 50 years and more out while having other groups focus more on development/manufacturing/marketing.

I'm not sure many companies can afford to think that way any more given the stock market focus on the short term. I think Google is doing the best job now with the self driving cars/Glass projects.

>A caller should dial a number and then push “send.” That way, the mobile caller would be less rushed;

Makes sense. I hate landlines for this reason.

afaik, sms is out of band. so it would not have been affected by dialtone.

Dialing is "out of band" everywhere except analog phones, and dialtones are just the reverse channel giving feedback to the user (just because there's only audio information, a "single band").

The TL;DR of the article probably is: The invention of a separate channel for out-of-band signalling, which the original short-message-services use, was related to the idea to have fully prepared phone-number in a call-setup message to the network.

Thank you for connecting the SMS to the lack-of-dial tone. I thought it was the most interesting part of the article, but wasn't explained very well.

So if I have this correctly: the Push-To-Dial feature necessitated an out of band control mechanism, and this mechanism is what SMS lives on?

Do SMS messages still use the out of band? I'm guessing that iMessage, etc., use the data channel (based on the encryption discussion last week). But when I send a text that's a proper SMS, or an SMS from Twilio/etc., does that use this out of band mechanism?

SMS does use the out of band but it was more of an accident of the SS7 protocol. you should read up on SS7 - http://en.wikipedia.org/wiki/Signalling_System_No._7

SMS is more of vestigial debug feature of GSM that was brilliantly commercialized than anything that has to do with SS7. And by the way SMS is out of band on the backhaul network, not on the air interface (and the whole SMS protocol on air interface is horribly complex and inefficient, it essentially involves building data call that passes SS7-like messages, typically one per such "call" in each direction, another funny fact of this technology is that address fields in these messages have larger maximum size (256B) than than actual SMS contents (140B)).

But when one wants to take SS7 into this dial tone discussion it is important to note that traditional SS7 call flow (INITIAL ADDRESS/SUBSEQUENT ADDRESS/ADDRESS COMPLETE) is designed for normal incremental dialing and not cell phone style number at once then send style. I'm too lazy to look it up but I believe that GSM's ISUP/DSS1 derived on air signaling protocol is capable of that too (although I've never seen device that allows user to do that). Relatively common user experience problem with VoIP deployments I have seen is that this is effectively impossible with SIP and user has either to explicitly end dialed number with something or wait for some timeout.

Thanks for clarifying! It's like i had all the concepts there - out of band and SS7 - but didn't really get the details. This is why I love HN.

They do in movies, for dramatic effect.

does anybody know that this is the reason that Skype doesnt have a dial tone when calling certain lines?

Reply on news.ycombinator.com