Dragging Telephone Numbers Into the Internet Age

azoblue writes with this teaser from Ars Technica, presenting a tempting suggestion for online consolidation: "E-mail, IM, Facebook, phones—what if all of these ways to reach you over a network could be condensed into a single, unique number? The ENUM proposal aims to do just that, by giving everyone a single phone number that maps to all of their identifiers. Here's how it works, and why it isn't already widely used."

Comments

Been done: .tel domain

[benwiggy]

You mean like the .tel domain?

.tel provides all contact information: phone numbers, postal addresses, email, web addresses, etc -- all within the DNS.

Re:It's not the same

[epine]

Jenny, I got your number
I'm gonna make you mine
Jenny, I got your number
86.75.30.9

You, too? Last visible hop 10.226.70-86.rev.gaoland.net gaoland.net seems to be slashdotted already.

One ring to find them all, one ring to bind them. I wish had the graphics talent to rework that scene where the Nazgûl rider is sniffing the tree roots for sneaky hobbits, and his phone goes off with some super goofy ring tone. We could redo Orthanc as a wifi repeater and that eyeball as a Pringles can.

I'd rather have call display that worked reliably.

DNS

[pikine]

The ENUM proposal is essentially asking for DNS lookup as a public service run by government or other regulatory bodies. First of all, as you said, why don't we just use names? And second, I'm not sure we want public DNS run by government or regulatory bodies. We already have community-run free DNS service such as http://freedns.afraid.org/ or commercial free service like http://www.dyndns.com/ or http://www.zoneedit.com/. If you're worried that free services would go away, a lot of domain name registries are also offering DNS service at nominal fee, and they would be less likely to vanish. Several people can share the cost of a domain.

All people need to do is to find creative uses of domain names. I think this is the hard part.

Re:DNS

[TheRaven64]

I didn't RTFA, but it's not a new idea, and both you and the submitter seem to be missing the point. You can store arbitrary contact addresses in NAPTR records in DNS, so you can store email, SIP, POTS addresses, or anything else that can be represented by a URI. The other part of this is allowing reverse mappings, from telephone numbers to something less archaic.

Telephone numbers, like IP addresses, are globally unique network endpoint identifiers. They are assigned by the UN (specifically the ITU-T, which assigns prefixes to countries) and allow you to call any telephone from any other telephone in the world. The problem comes when you have an endpoint that is really a SIP account, for example. Currently, that mapping has to be done in quite a static way.

The idea of the proposal is that the e164.arpa. domain will be reserved for resolving telephone numbers to domains, just as in-addr.arpa is used for resolving IP addresses to names. This doesn't need to be government run, but it does need to be authoritative. That means that e164.arpa will be controlled by the ITU, 1.e164.arpa will be controlled by the USA, 4.4.e164.arpa by the UK and so on. You will then get a subdomain of this. Telephone companies that have large assignments of phone numbers get large ones, individuals may get a single 15-digit number. This can then map to any other resource.

It's not intended as a long-term solution. Eventually, the POTS network is going away (large chunks of it are IP internally already) and you will just use DNS to map directly to SIP, but while interoperability with the POTS network is desirable - say, for the next couple of decades - this lets people with POTS phones initiate calls to SIP phones without having to define a specific bridge or static routing. You'll dial a number on your phone, your telco will look up the SIP address and then route the call there via their bridge.

I currently have a phone number connected to a SIP address, but it only works from POTS lines because my SIP provider operates a SIP to POTS bridge. With this proposal, anyone can operate one trivially. You will just need to get an e164 number assigned to you and configure the DNS entries to point to your Asterisk (or whatever) server.

Re:DNS

[xaosflux]

I didn't RTFA, but it's not a new idea

It certainly is not, 1996 just called and wants their Universal Internet Numbers back (http://en.wikipedia.org/wiki/ICQ#UIN)

Enum: why you want it

[Raindeer]

I'm the author of the piece. Most comments in my opinion make the mistake of saying: I want this or that to be my identifier. Or I don't want a universal identifier.

The reality is: there are two identifiers that are on most business cards. Phone numbers and e-mail adresses. Both could be used in a much more advanced way. No matter which way you look at it the telephone number won't go away. ENUM would enable you to use it in multiple ways.

Re:It's not the same

[Anonymous Coward]

Yeah it's a Chen Prime. Weird.

Re:I am not a number! I am a free man!

[L4t3r4lu5]

No no no, Paranoid was Black Sabbath. Number of the Beast was Iron Maiden. Totally different...

Re:Why would you want to keep the telephone number

[wvmarle]

A number is very easy to give over the phone. Easier than an e-mail address. This as so many letters sound very similar and so.

In practice I have been giving my fax number over the phone so they could fax me their e-mail address. Works great! Particularly considering I am often working with Chinese and other Asians with sometimes very poor spoken command of English. Numbers then go remarkably easy.

Re:How about using IP6?

[Anonymous Coward]

two words: URL Shortener.
Two more: DNS

I don't put 72.14.204.99 into the website when I want to use google. I use google.com (Which gets translated via DNS). Why would this be any different?

Hey baby... Your lookin hot... [call/email/message/text] me at Chris.ssn
(Which translates via DNS or URL Shortener into +DEAD:BEEF:CAFE:123:4567:890A:BCDE:F).

Re:It's not the same

[longhairedgnome]

From wiki for those not privy, A prime number p is called a Chen prime if p + 2 is either a prime or a product of two primes. The even number 2p + 2 therefore satisfies Chen's theorem. In 1966, Chen Jingrun proved that there are infinitely many such primes. This result would also follow from the truth of the twin prime conjecture. The first few Chen primes are 2, 3, 5, 7, 11, 13, 17, 19, 23, 29, 31, 37, 41, 47, 53, 59, 67, 71, 83, 89, 101, (sequence A109611 in OEIS). The first few non-Chen primes are 43, 61, 73, 79, 97, 103, 151, 163, 173, 193, 223, 229, 241, A102540. All of the supersingular primes are Chen primes. Rudolf Ondrejka discovered the following 3x3 magic square of nine Chen primes:[1] 17 89 71 113 59 5 47 29 101 The lower member of a pair of twin primes is a Chen prime, by definition. In August 2009 Twin Prime Search and Primegrid found the largest known Chen prime, 65516468355 2333333 - 1 with 100355 digits.