Probably more than you ever wanted to know about area codes

[unixronin]

From a discussion elsewhere on telephone systems and possible ways to let you dial a logically "nearby" phone using less keystrokes:

---

Remember, enough of the phone number space is full that there are essentially no un-issued numbers in existing area codes any more.  ALL "new" phone numbers in existing area codes [with the probable exception of some recently-allocated area codes and some rural area codes] have been previously used, and we've gone from each town having a single exchange and big cities having several, to even smallish towns having multiple exchanges and major metropolitan areas having multiple area codes each, to parts of some major metropolitan areas having overlaid area codes because there aren't enough phone numbers in a single area code and they don't want to split the same area code again.  The pool of available-but-unused area codes itself is rapidly shrinking.

A little research yields:

"The format of an area code is NXX, where N is any digit 2 through 9 and X is any digit 0 through 9.  Initially, the middle digit of an area code had to be "0" or "1". When this restriction was removed in 1995, additional area code combinations became available.  There are 800 possible combinations associated with the NXX format.  Some of these combinations, however, are not available or have been reserved for special purposes."

Among these special purposes, it is noted that 37X and 96X have been "set aside by the INC for unanticipated purposes where it may be important to have a full range of 10 contiguous codes available", and the entire range of N9X has been "reserved for use during the period when the current 10-digit NANP number format undergoes expansion".  Codes with matching final digits (including N11) are reserved for special services.

That leaves 620 usable area codes.  By my best count, 384 of those are currently in use.  33 more have been planned but are not yet in service, for a total of 417 allocated area codes, leaving 203 available area codes for future expansion before they have to increase the number space.  Note that the N[01]X restriction in place up until 1995 limited the area code space to 160, minus the special area codes; subtracting N00 and N11 codes leaves 144 total available area codes prior to 1995.  In other words, in the 12 years since 1995, we have allocated twice as many new area codes as the total number of area codes available in 1995.  If we continue to allocate new area codes at that rate, even assuming a merely linear usage rate, we will exhaust the entire usable ten-digit space by about 2016; and if we make a real-world reasonable assumption about the probable shape of the usage curve, then 2012-2013 is a more likely estimate.

Comments

[metahacker.livejournal.com]

Dunno. Phones are increasingly becoming smart; it strikes me as pretty trivial to hook them all into packet-switched networks, at which point phone numbers become as outdated as bang-paths.

Also I think cell phone penetration has slowed its rate of growth, if not actually peaked. There are only so many people in the US, and I think right now people have *more* phones than they will in the future (most people still have a cell phone and (part of) a land line and maybe an office number, too). And cell phone overlays are a major chunk of that area code usage.

I guess it's an open question whether technological advancement will catch up before 2015 or so, but personally I'm expecting phone numbers to be made irrelevant in the next 5 years or so. Maybe 8.

[unixronin.livejournal.com]

Cell phones are becoming smarter, yes, but POTS phones are still pretty stupid devices. And, while the backbones are packet-switched, the truth is that all the last-mile infrastructure isn't — and making that switchover is the most difficult and expensive, because it'll basically mean forcing everyone to buy new phones.


Then, as a side-effect, maybe we could finally get rid of Group III FAX ....

True, but...

[metahacker.livejournal.com]

Enh, who uses POTS? Residential customers. And there's a cap on the number of landlines being used...it's not the growth area. The last three places I've worked had IP phones, too. Maybe that's a factor of working in hi-tech, sure, but it points to things to come.

Re: True, but...

[unixronin.livejournal.com]

Residential customers and a lot of small businesses.

Apart from anything else ... when was the last time you heard of a telco solving any problem in an efficient and cost-effective way...?

[mrmeval.livejournal.com]

There are Wifi/Voip phones, not perfect though. I suspect you'll see all of it go to the internet in some form or another. I'm looking at picking up a cell phone dev kit that has GSM, wifi and bluetooth plus other goodies but I'm waiting for something out of alpha testing.

[johnkzin.livejournal.com]

On the flip side, how are we doing with SSN's, which are also 10 digits?

Why not have 11 digit phone numbers, where:

a) the first digit is a type (communication infrastructure (0 and 1), numbers for services provided by your phone carrier (4), government (9), reverse charge line (8), 900 number type fee based line (7?), regular business line (?), private individual (?), short-cut digit (2?), ?)

For "individuals", use the remaining 10 digits to assign a life-long phone number that goes with them no matter where they move ... it's like a hostname, mapping to an underlying "circuit" (like a CDMA phone's ESN). That underlying location could even change during the day ("when I'm at home, park my personal phone number at my house phone; when I'm on the road, park it at my cell phone; etc."). You'd just need to have some sort of service (comparable to a dynamic routing protocol) that tells you where a given individual number is currently parked. And, most people would probably just leave it at their cell phone all the time. Plus, you could have multiple individuals parked at the same circuit (everyone within a family, for example, parked at the same home landline phone).

Obviously you wouldn't want this to be the same as their SSN, but it does turn out to give as many combinations as an SSN. And it might be usable as a "public, non-confidential" ID number.

You could allow individuals to request vanity numbers that spell out certain words, and you could also allow them to be changed on special requests (harassment, etc.).


b) for government, business, etc., the second and third digits indicate a region (give some states 2 (or even 3) numbers, have a region for DC, and a couple that are "nation wide regions"). The 4th digit is a sub-region or sub-type, allocated by the mandate of the region's own management. Maybe even use the 5th digit this way.

So that leaves 6 or 7 digits that are entirely usable by a given sub-region, NOT having to share that space with individuals, nor having businesses and government sharing the same number spaces.

c) the "Short Cut" first digit would automatically go to "business, and physical region/subregion where I am currently located (according to my physical circuit or cell tower)", so that you can just dial the last 6 or 7 digits of a local business.

d) not all numbers would have to extend to a full 11 digits. 411 could still work, as the phone carrier could just say that "11" is the information line within their 4 category. Similarly, the "11" region number could be set aside by governments for "911".


You could say that "10 digits is a lot to memorize for an individual, the current scheme means you only have to memorize 7 digits, which is right in the middle of the 5-9 digit range that is based on psychology data"... except that's not really relevant anymore because memorizing phone numbers is much less common than it used to be. Today, lots of people don't even know each other's phone numbers, they just have them programmed into their handset, and only ever see the person's name and the type of number (that person's cell phone, that person's home phone, that person's work phone, etc.). For individuals, my scheme sort of depends upon that -- you only ever know an individual's number when you first acquire it, and program it into whatever you use for a contacts list.

[unixronin.livejournal.com]

On the flip side, how are we doing with SSN's, which are also 10 digits?

It's worse than that, too, because those first three digits encode the issuing office, and there are not 999 issuing offices. Also, I know there is a theoretical premise built into the system that no social security number will ever be re-issued. (Ten digits probably seemed like plenty when the system was set up.) Whether that is actually honored any more, I don't know.

For "individuals", use the remaining 10 digits to assign a life-long phone number that goes with them no matter where they move ... it's like a hostname, mapping to an underlying "circuit" (like a CDMA phone's ESN). That underlying location could even change during the day ("when I'm at home, park my personal phone number at my house phone; when I'm on the road, park it at my cell phone; etc."). You'd just need to have some sort of service (comparable to a dynamic routing protocol) that tells you where a given individual number is currently parked. And, most people would probably just leave it at their cell phone all the time. Plus, you could have multiple individuals parked at the same circuit (everyone within a family, for example, parked at the same home landline phone).

Exactly such a system has been proposed at various times. Number portability is a baby-step along the way. Whether it'll actually ever get implemented in a form like that remains to be seen.

You could say that "10 digits is a lot to memorize for an individual, the current scheme means you only have to memorize 7 digits, which is right in the middle of the 5-9 digit range that is based on psychology data"... except that's not really relevant anymore because memorizing phone numbers is much less common than it used to be. Today, lots of people don't even know each other's phone numbers, they just have them programmed into their handset, and only ever see the person's name and the type of number (that person's cell phone, that person's home phone, that person's work phone, etc.). For individuals, my scheme sort of depends upon that -- you only ever know an individual's number when you first acquire it, and program it into whatever you use for a contacts list.

Exactly. There's basically two classes of phone numbers these days — ones you have to look up anyway, and ones you have on some form of memorized speed-dial.

One would certainly like to HOPE that when the time comes, instead of just patching the system, people would think carefully about the problem and completely revamp it in a single step calculated to maximize functionality and user-perceived backwards compatibility while minimizing disruption of services. But that's not an easy thing to do, and real-world experience says we'll get another band-aid patch and continuation of the trickle of incremental improvements. One thing I expect to see happen possibly sooner than other changes is to see the concept of "long distance" tolls just completely go away. Possibly even for international calls. Anywhere on the planet for one flat rate.



(As a side note, it boggles me that people have difficulty memorizing more than nine digits. I routinely memorize my 19 [total] digit credit/debit card numbers [23 if you count expiry date], and I used to have the value of pi memorized to 26 places when I was using it regularly. But I already know I'm an edge case.)

[ithildae.livejournal.com]

Back when I was traveling too much, AT&T came up with a "500" number concept. It was seriously cool. I had a number assigned to me that I could use anywhere! All I did was redirect the ring to a number where I was located. It had voicemail and all the rest built into the system. further, I could use it instead of a calling card to get toll free (and surcharge free) long distance from my hotel rooms (So I could call my family.) It saved me tons of money and everyone was able to reach me wherever I happened to be. (The cost was much less than most cell phone plans today.)

Given that the concept has been shown to work with little management on my part, how long until people are assigned a personal phone number that can be used on any device? For POTS, you would dial your number first, then the destination number, so only the switch needs to be smart. I think we have the infrastructure to make a plan like that work, all that stands in the way are politics and infighting among the telcos.

I think increasing the digits in phone numbers would be resisted. I am not certain why, but I never invested in a PDA because I carried my calendar and over 500 phone numbers in my head. (I, too, am an [edge|head] case.)

[unixronin.livejournal.com]

I think increasing the digits in phone numbers would be resisted.

It may be resisted, but there are plans already in place to do so, though I don't know the details of the plan.

[freetrav.livejournal.com]

If the US and Canada were to be separated as far as the NANP were concerned, the US could recover 25 NPA codes. How long would that last?

It would mean that calling Canada from the US or the US from Canada would require dialling 011-CC-NPA-XXX-XXXX where CC is a number beginning with 1 (and different for the US vs. Canada), rather than the current (1-)NPA-XXX-XXXX, but having to do this is considered normal just about everywhere else in the world - few countries share an international dialling code the way North America (really, US/Canada/Possessions/Some-of-Caribbean) does.

I've never found documentation for the assertion that I've heard that there is an international agreement that no telephone call anywhere in the world could be more than twelve digits, excluding international access codes (e.g., 011 in the US), but if such an agreement exists, it would certainly put a crimp in any plans that would involve adding more than one digit to either the NPA code or the 7-digit local number. OTOH, adding a single digit to the existing NANP, either in NPA or local number, would guarantee a minimum of eight times the number space (assume an addition of a leading digit that may not be 0 or 1). How long would THAT last?

[house-pundit.livejournal.com]

Hey, people remember their zip codes. Skip four digits and go straight to five.

Hrms. I'm not in the mood to do the math if they go to four-digit prefixes and go back to area codes genuinely designating an actual area.

Can you imagine a day when your phone number is as long as your credit card number? Phones are programmable more and more often. There are limited reasons for the, "Why not?"

[unixronin.livejournal.com]

I'd like to think it won't be too much longer (relatively speaking) before the actual mechanics of phone numbers becomes something for the phone to worry about. You tell your communication system "Call Aunt Mary", and the comm system looks up Aunt Mary and figures out how best to reach her. And ideally, if you tell your comm system "Contact Cameron Diaz", your system contacts her system, her system looks you up, finds you're not on her permitted list, and you get automatically forwarded to her agent to explain why you think you need to (and should be able to) talk to her.

It'd be nice if the same thing automatically happens when a telemarketer tries to call you, but the Direct Marketing Association will undoubtedly lobby against such a thing, and those spineless fuckers in Congress will undoubtedly be only too happy to accept the DMA's thirty pieces of silver and sell you out.

[bunrab.livejournal.com]

We are SO beyond geeky.