Talk:Original North American area codes

Reverts of territory descriptions

As all N1X area codes covered only a portion of their respective states or territories, it is highly useful to list roughly which part each area code covered. I checked a few and found them to be correct. If there are any that are not correct then it would be highly advisable to correct the offending items only, and not remove all of them wholesale, and not continue edit warring about that. Kbrose (talk) 17:21, 9 January 2015 (UTC)

Newfoundland

Are we any closer to addressing the question of the Dominion of Newfoundland? At one point, this page had it out of the original list of 1-902 provinces and area code 709 was claiming the opposite. I'd found a source which claimed the Rock was entirely manual switchboard until 1948 and the cross-province line from St. John's to Port aux Basques not installed until 1949, leaving just an international manual shortwave radio link from St. John's to Montréal and another from St. John's to London UK. That should knock it off this list as Newfoundland was not an operator-dialable point with or without a routing code. Comments? K7L (talk) 19:24, 14 January 2015 (UTC)

Small numbers / large populations?

Is it the case that cities/states with larger populations were deliberately given area codes with smaller numbers (i.e., that took less time to dial)?—cf. New York, Los Angeles, and Chicago, the three biggest metropolitan areas, with 212, 213, and 312; or New Jersey, the most populous state assigned a single area code, with 201. And if this is the case, is there a source we can find for it so we can mention it in the article? AJD (talk) 16:34, 9 September 2016 (UTC)

I believe this to be - approximately - the case. Remember that in 1947, telephones used only rotary dials, and a "0" takes a lot longer with a rotary. dial- and ties up the central office longer - than a "1". If you order the original area codes by the sum of the number of dial pulses ("0" is 10, "1" is 1, "2" is 2, etc.) and compare them to the served population of that area code, you'll see an approximate 1 to 1 correspondence between population and dial pulses. But the Bell Labs being what the Bell Labs were, I'm sure they used call volume data instead of population. The objective was to minimize switching equipment holding time in aggregate, thus minimizing equipment requirements. I've never seen the call volume statistics (calls dialed to each area code) so I had to use population as a proxy. It's not a good proxy, because in 1947 (as now) business traffic makes up a large portion of call volume and is somewhat independent of population. So listing the area code dial pulses by population will be an approximation of a call volume by area code list.Tenafly11 (talk) 07:17, 21 January 2017 (UTC).Tenafly11 (talk) 07:19, 21 January 2017 (UTC)

This is a list article, that does not belong here. The main articles of the topic discuss some of the practices already. But there are no reliable references for such ideas. None of the original planning documents, reports, and updates mention such allocation method. It is not call volume that was important, but call routing. NPAs were drawn so that heavy toll routes and tributary routes were not unnecessarily cut across boundaries. At the same time NPA codes were assigned to maintain sufficient randomness in the remaining pool of codes. Randomness (to a degree) was chosen to reduce customer confusion. With these constraints it is most natural to assign the large population centers first and start placing them in the upper left corner of a table listing x0N vs N and x1M vs M, where x=2 to 9, N=1 to 9 and M=2 to 9. But this only 'works' for the fewest of NPAs and the idea is easily contradicted when actually plotting area codes and population on the same grid. Large call volume occurs in large cities internally, there would not be an advantage in reducing just the time of dialing an area code. Even if always dialed, the area code was only 30% of the telephone number. The real gain was to minimize the need for dialing an area code. Kbrose (talk) 18:10, 21 January 2017 (UTC)

It may be a list article, but it already contains several paragraphs of background information on how the original area codes were organized; I don't think that that alone precludes mentioning something like "it's not a coincidence that New York, Los Angeles, and Chicago have the lowest numbers" here, if such a thing is true and could be supported with sources. AJD (talk) 07:11, 13 February 2017 (UTC)

While the idea may be interesting to pursue, there don't appear to be primary sources that support the theory. This whole idea of counting pulses appears as non-sense and really doesn't make sense, when one considers that area codes were never dialed when making local calls. The people dialing area codes to high-volume cities where in other places with 'long' area codes, consequently, it should be the other places that should have 'short' codes, since big cities also produce high outbound volume. Indeed, none of the most direct sources by the designers of the plan (Pilliot, Ryan, Mabbs, Guengerich, Myers, Clark, Osborne, Nunn, Ericson,...) mention an assignment by size of city, they all point to other considerations. Indeed, if the size of the city is tabulated along with the area code, the correlation quickly disappears, after a handful of places. But surely when assigning codes, one would probably start with the places with the most routes. The first versions of numbering were actually somewhat systematic in that adjacent areas had similar codes, kind of sweeping the nation. This was abandoned in favor of 'random' allocation, which can be seen in the table I added. Codes were assigned to maintain a degree of 'randomness' of the pool of remaining codes, resulting in the assignment of the upper left triangle first, leaving a lower triangle with the same 'randomness'. Surely it wasn't all random, New Jersey got the first code no doubt out of vanity, since Bell Labs had most of its facilities there and dialing between sites would be possible without class-4 toll switches. Moreover, the state really should have been divided into at least two NPAs, not one, which had to be corrected only a decade later. D.C. probably got the second to please the law makers and anti-monopolists. Given the task of random assignment, most anyone would probably pick the big cities first across the country, New York, Chicago, L.A., etc. and not focus on the small places. All of the big cities already had considerable infrastructure for extensive toll routing, which was essential to make a nationwide numbering plan possible. The difficulty in recalling this history correctly is finding original sources. kbrose (talk) 13:32, 22 August 2020 (UTC)

For the record, WP tries to avoid "original" (i.e. primary) as sources of fact. Secondary sources are preferred, emphatically. EEng 05:31, 28 January 2022 (UTC)

Conversion to full article

Given the prior comments about additional topics and material, the article now is a full article, with the intent being to add more discussion of subtopics, that might be too detailed for the general article on the NANP. kbrose (talk) 20:06, 21 August 2020 (UTC)

Discrepancy in original Canadian area code assignments

I have just noticed that there is a discrepancy between the original area code assignments for Canada (as shown in the map from the Winter 1947 issue of Bell System Magazine) and what was actually implemented. Specifically, area codes 418 and 613 are swapped. It seems fairly amazing that (as far as I can tell) this has never been called out, aside from the reference to a "printing error" in File:North_American_Numbering_Plan_NPA_map_BTM_1947.png. Fabrickator (talk) 00:16, 1 January 2022 (UTC)

Nothing amazing here, there was no reason for being "called out", as it was (a) clearly against the prevailing rules of assignment, and (b) the correct map is available by January 1949. kbrose (talk) 18:39, 3 January 2022 (UTC)

Do you know what accounts for the discrepancy in the stated number of area codes? The caption for the map in the January 1949 Bell Labs Record says there were 83, most other sources seem to say there were 86. Fabrickator (talk) 20:21, 3 January 2022 (UTC)

Simply another editorial error. Someone in the publication department couldn't count straight. The map actually contains 87 NPAs, as to be expected for that year. kbrose (talk) 21:07, 3 January 2022 (UTC)

There was no swapping error, not was area code 418 involved. There seems to be some darkening in the map on area code 416, but i do not see the above error involving 613-418. 76.69.7.202 (talk) 16:24, 22 January 2022 (UTC)

I'm afraid you don't understand. If you click on the map, then click on the "more details" button, it will take you here. That will include text which explains "Adapted and corrected for printing error from Bell Telephone Magazine 1947". Now if you take a look at the article as printed in Bell Telephone Magazine, you will find the area code map in question as it was actually published, which will be the same as the map in the Wikipedia article, aside from the difference in the location of area codes 418 and 613. There does not seem to be any contemporary explanation of the discrepancy between this map and area code maps that were subsequently published, and to characterize this as a printing error is (to use a word that seems to frequent my vocabularly of late) speculative. Fabrickator (talk) 16:52, 22 January 2022 (UTC)

Yes, you are 100% correct. I thought the discrepancy in question was on the articles current map, which has no swapping. Apologies, I misunderstood. That's my error. 76.69.7.202 (talk) 17:06, 22 January 2022 (UTC)

The discrepancy is in the version of the map as posted on Wikipedia, because it is not a true representation of the map published in Bell Telephone Magazine, and so far as I am aware, there is not some earlier map or other document that agrees with the area codes that were actually implemented. So what I'm saying is that there's no basis for calling this a printing error. In short, the map was printed as it was printed, then for whatever reason, somebody decided to reverse the two area codes. Fabrickator (talk) 22:22, 22 January 2022

Of course it is speculative, all accounts I have ever seen state the first area code map as implemented. What happened with that map from October 1947 to sometime in 1948 is unknown, by the time Indiana was redrawn in 48, those Canadian codes were swapped. But in the end it is entirely inconsequential what one calls this, as nothing happened in Canada w/r/t OTD until probably a couple years later. But editorial errors happened frequently, even on the January 1949 map, where those who prepared the text couldn't count. Or the typesetter made a mistake. Following the cause of mistakes is always speculative today. kbrose (talk) 01:14, 23 January 2022 (UTC)

Errors in Bell System documentation are pretty common anyways. These pubs were hardly peer reviewed, unless they were submitted to the AIEE as well. Another article or book (can't remember which now) claims that there were only eight Canadian NPAs, while every one else states nine. Over time a kind of corporate amnesia is quite evident in many subject areas, and seems to accelerate toward divestiture. This whole numbers game of the count of the original NPAs is somewhat arbitrary anyways. If it were my call, I would say it was 87, not 86, because nothing much happened in Indiana between 1947 and 1948. This probably had something to do with preparation for the installation of the 4A scheduled for cutover in Chicago in late 1948, and for that they probably needed an independent NPA in the eastern suburbs. But I am not going to challenge the myriad of sources that pretty unanimously state 86. It doesn't matter. kbrose (talk) 02:15, 23 January 2022 (UTC)

Deletion of properly sourced content

User:Fabrickator insists on deleting properly sourced content without providing reasoning for such actions. Wholesale removal of various new content in multiple sections is never justified. Expanding an article on my part is not ownership of the article. I wrote the whole article to begin with. Defending good content with sources should be the activity of any editor. He should state objections to specific points or statements and show what he think is wrong. Apparently the editor believes in the modern conspiracy in the click theory mentioned previously in this talk page, and does not like to see a proof based on all the original literature of the time (1940s and 1950s) that it is nonsense. Therefore he just wishes the mention of this click theory deleted, despite there being reliable references to its existence, as quoted in the article. The NPOV violation is entirely on his part. I tried to present a balance by giving credit to its existence with references. BTW, I take great care to select proper sources and present content objectively. The very nature of these removals shows their malignant intent. He should have brought his specific objections to the article talk page. Perhaps a CheckUser is appropriate against the recently offending IP-user. kbrose (talk) 23:58, 19 January 2022 (UTC)

@Kbrose: "I wrote the whole article to begin with" ... Exactly! You are saying that it's your article and that you have some set of special privileges. I stated the exceptions I took to your edit in previous reverts of that content, you chose to delete those objections from your talk page, but in short, you are making an argument that area code assignments were not based on population, and instead of offering citations which actually support your claim, you offer citations which make the claim about area code assignments based on population and then present your arguments to the contrary, claiming that the arguments you make are based on fact, yet putting this argument together is exactly what constitutes WP:OR. At this point, nobody is insisting that the article should even address the population-based claim, but if they were to insist on it, there would not be any basis to refute such a claim. Fabrickator (talk) 00:41, 20 January 2022 (UTC)

I have no other privileges than to maintain intellectual integrity and combat vandalism, like any editor. I am simply pointing out that I am not just some fly-by-night editor taking over an article that I don't understand. The WP:OWN rules correctly state that sole editors of an article should be taken more seriously than others, for obvious reasons. That should be honored especially when an article is so well sourced and the evidence is so overwhelming. I am not making any unsourced claims, I simply show the facts and statements in the original documents by the contemporary engineers, and do not suppress references that speak against that. If you think that, then prove it by providing 'your' references. But you have consistently failed to do so. If I find any credible idea or even technical plausibility in a reference I would point it out. In the end I have no beef in the method of assignment back then. I am interested in getting it right. I thought the article should indeed mention the click theory, as it comes up in the media, even serious media, from time to time, always based on hearsay. WP has articles and mentions for all kinds of conspiracy theories no matter how prominent. Furthermore, I did not ridicule or belittle the idea, I simply pointed out that the sources do not support it and that it fails tests of technical credibility. kbrose (talk) 01:12, 20 January 2022 (UTC)

Apparently, you base your claim of OR of POV on the statement(s): However, no authoritative, reliable sources for this method are quoted and none of the original, contemporary Bell System reports provide any consideration of direct current (D.C.) pulsing, or any support of correlation to population density, as the toll operators did not use rotary dials,[19] and perhaps: The concept agrees for a few of the largest of cities,[12] but these were already the designated Regional Centers for switching since 1929. but this one acknowledges (again with source!) that some correlation to size does indeed exist, simply based on the arithmetic of smaller numbers. If that warrants removal of broad swaths of content from the entire article, as you have practiced, then you have some explaining to do. kbrose (talk) 01:47, 20 January 2022 (UTC)

@Kbrose: Can you please explain one thing to me? You claim that I am the one who is so concerned about this, but take a look at the edit where you added this so-called conspiracy theory. (Frankly, I don't believe that even you think it's a conspiracy theory, which would generally be considered to involve some evil scheme. Is it really your POV that efficient dialing is an evil scheme?) Fabrickator (talk) 01:41, 20 January 2022 (UTC)

BTW, there is nothing efficient about this, small numbers are in fact just short. The efficiency of call origination with senders (or not) comes from the speed of switching/making connections--and the references show that too--and not from fast dialing of just three digits of ten. kbrose (talk) 02:16, 20 January 2022 (UTC)

That is an old version which I have already revised. I thought that emphasizing it as a 'popular culture' artifact was in fact belittling it unduly, That's why I left it removed for quite some time until I had time and context for a better exposition based on better sourcing. I have been using the term conspiracy without prejudice only on talk pages to emphasize my point, as I have noticed over the years that those who believe in it almost behave like a cult. kbrose (talk) 01:53, 20 January 2022 (UTC)

So??? Yes, of course it's an old version. But you are the one who raised this issue, not anybody else. When I claim that this is OR, it's not because of what you say, it's what you don't say! You don't say that there's any source which supports the claim that area code assignments did not take population (or population density, as you sometimes put it) into account. You've reasoned this out through your own argument. Now you've closed your mind to the fact that what you're saying is WP:OR.

Re: You don't say that there's any source which supports the claim that area code assignments did not take population (or population density, as you sometimes put it) into account. WHAT? I am not sure I understand that jibberish of recursive negations(?). Engineers would not write articles about reasons or issues that they did not consider or intend, or that didn't make sense at the time. They write about the research and consideration they did perform and that are relevant. That is what is what I focus on and write about. If you think that there are such 'suppressed' references that provide a new insight into this, or that prove your theories, then present them. You have not provided a single one, instead you continue to obstruct the progress of the article by senseless reverting all content. Now you seem to be involved in consensus shopping for your efforts of preventing new content for the article. You have not made a single constructive contribution to the article. kbrose (talk) 23:36, 20 January 2022 (UTC)

Oh, and by the way, you've mis-charaacterized about WP:OWN. All you can really get out of WP:OWN is that all editors ought to respect other editors, and the fact that a particular editor may have more familiarity with a subject than other editors should be taken into consideration. I can imagine cases where one gives some extra consideration to how an article is presented or some ambiguity in interpreting a source. But it's absolutely not an absolute.

There's nothing special about your wealth of knowledge of this topic that constitutes permission to decree that presenting your logic about how area codes were really and truly assigned is not OR, nor that it is okay for you to come up with some basis to justify presenting your unsourced claims about area code assignment. The fact that some people believe something which is arguably false doesn't mean that you somehow have the right to present your counter-argument and ignore the rule against OR. And now you want to dispute that dialing numbers with fewer pulses would be more efficient, once DDD (which was pretty clearly part of the plan) was implemented? Incredible! Fabrickator (talk) 02:41, 20 January 2022 (UTC)

Re: DDD. DDD was not part of the "plan", the plan was Operator Toll Dialing. The sources already in the article clearly indicate that customer dialing was thought of a possibility should it prove possible and desirable later. Clearly it was on their mind, sure, and it is mentioned. So why didn't they mention a code assignment by size in the same realm? It didn't make technical sense. This makes the assignment per population size even more unlikely. It was only after the Englewood trials in 51 that this was deemed a go-ahead. kbrose (talk) 01:31, 21 January 2022 (UTC)

It is not OR when the references are so clear. This only shows that you do not understand the issue, and are blinded by the idea that dialing a '1' or '2' is so short. Yes it is short, but first of all the user spends about the same time locating the 1 or 2 with a finger as finding any other digit, especially when they have to read the number from a note or directory. Locating the digit takes the most time, and is the most error prone, as research reports by Bell Labs have documented (You want references?). When the entire ten-digit number is dialed it doesn't make much difference anymore that one or two of them were short. It takes about 20-30 seconds. Secondly, the dialer then sits and waits for the network to complete the call and ring the destination. In the 50s this could still take ONE MINUTE, for the long-haul calls. You call that efficient anymore? After even 30 seconds you won't give a nickel anymore about how short your 212 code was to dial at the start. Every relevant reference shows that this was *the* parameter to be improved. Never is mentioned the speed of dialing by the user, and best of all, most users in the nation couldn't even dial anything for LD until the 1960s. Switching determined holding times of senders, not a fraction of a second for dialing a faster digit. This appears to be the fallacy of the entire click theory, no one bothered to actually read anything. That is my only OR. kbrose (talk) 03:10, 20 January 2022 (UTC)

Your edits are plagued with inaccuracies. They are being repeatedly reverted by numerous editors. Until yesterday, you refused to use the talk page. You had another editor (your friend?) report ME for edit warring for reverting your erroneous edits and asking you to use talk page. You display extreme signs of wp:own and claim that you wrote the article when a quick article history check proves otherwise. Maybe it's time to take a look at your editing techniques and general demeanour? 76.69.7.202 (talk) 15:19, 20 January 2022 (UTC)

Bring on those purported inaccuracies point by point, and don't forget references that show them. kbrose (talk) 15:42, 20 January 2022 (UTC)

Please note that User:Kbrose has been reported to Wikipedia:Administrators' noticeboard/Incidents for the above mentioned edit warring, article ownership, and general aggressive behaviour. 76.69.7.202 (talk) 16:26, 20 January 2022 (UTC)

Would people please focus on content and stop discussing who-owns-what. Do not put discussion about an article on a user's talk page—put it here without mentioning other editors. State what text should be added/removed and give a reason, with a source. Johnuniq (talk) 02:28, 21 January 2022 (UTC)

I have been focused on content from early on in this discussion. The issue of ownership arises when someone feels they have the right to ignore the rules. To wit:

The content in question (last paragraph of Special:Permalink/1066905243#Assignment plan, starting with the text "The history of assignment of the original area codes") makes the point that various sources assert that area code assignments took population into consideration, assigning "faster" area codes to areas of high population. However, it goes on to claim that the available sources are not suitable (e.g. nothing contemporary from Bell Labs) and from there, provides various sourced claims which presumably, taken in combination, are incompatible with the idea of area code assignment based on population, thus implying that the original assertion about area code assignment based on population is false. No source actually supports this conclusion, it is reached through claims from a variety of sources combined with the editor's own analysis. Thus it is a combination of WP:OR and WP:SYNTH and is not permitted. Fabrickator (talk) 07:51, 21 January 2022 (UTC)

The purported conclusion that the click theory is false, is never stated in the section. It is Fabrickator's synthesis. The section simply states facts with references, and contrasts them, compares them with facts (John states A, Jane states B, John and Jane agree in C, John makes no reference to B) The presentation never reaches a conclusion that the click theory would be false. (A is right, B is false) The section even acknowledges that the theory agrees with the original assignment for a handful of cities (sourced). If Fabrickator thinks that these facts show that the click theory is false, then that is his personal opinion; he is simply offended that it contradicts his beloved click theory, and he wishes any mention of it deleted to keep it alive. This is the reason that he refuses to engage in the actual facts and present references that speak one way or another. Instead he uses a very contrived method of suppressing facts. Just claim OR. Perhaps this is indeed the reason that he avoided at length to discuss this on THIS talk page, instead writing in other places on WP, user talk pages and the NANP article. I have tried to be as open-minded about it as the references allow, without drawing a conclusion. Deleting any mention of the click theory is biased, on the contrary, and should give the more than casual reader who knows a little bit about telephony cause for concern, as this theory has been purported in the media in the last couple of decades. Fabrickator accuses me of synthesis for the reason that he saw that I have previously stated in this talk page that I think the click theory is nonsense (for good reason). So he assumes I set out to prove that. Apparently he now agrees, but can't shake it... a typical kind-of-cult behavior as may be observed with others who belong to that gang. There is enough material there to write a whole article, but objectively they have no basis in fact and technical merit. But we will never find a reliable reference stating for or against it, because engineers and scientist don't write about concepts that have no merit. It does take considerable technical expertise and reading of the history of this to decide this. I have consciously stayed away from a conclusion. kbrose (talk) 14:34, 21 January 2022 (UTC)

kbrose, a year ago (see #Small_numbers_/_large_populations?) you were stating, quite imperiously, that there don't appear to be primary sources that support the theory. This whole idea of counting pulses appears as non-sense and really doesn't make sense plus some conjecture that New Jersey got the first code no doubt out of vanity ... D.C. probably got the second to please the law makers and anti-monopolists. More recently (in this thread) you're acting like the issue is total time to call completion as seen by the subscriber, and no one's talking about that either. Now you're backtracking with some doubletalk about how you're not saying, or weren't saying, or something, that the "click theory" is false. And your talk about The people dialing area codes to high-volume cities where in other places with 'long' area codes, consequently, it should be the other places that should have 'short' codes, since big cities also produce high outbound volume is unintelligible nonsense.

The fact is that the "click theory" certainly does make sense, because there's a lot more than senders in play and as long as at least some signaling worked by clicks and pulses, the Lee weight of the codes certainly did matter, and potentially mattered a lot (e.g. for translator residence time). And Keevers (page 2 [1]), who certainly was in a position to know, says that it was one of the guiding principles (though tempered by others with which it was in some conflict -- all the principles were in mutual conflict to some degree).

And Keevers brings us to another point: you seem to misunderstand how WP uses sources. You want contemporary references (i.e. 1940s) references, but in fact WP prefers later, retrospective sources. The fact is that you're over your skis as the big "expert" on this topic. Cool your jets. EEng 14:27, 24 January 2022 (UTC)

I have not backtracked at all. There is no reason to backtrack, because the click theory appears as wrong as ever. I do admit that the short statement by Keevers is perplexing, but he doesn't have any sources for this either, and his statement is conflict with the sources he states. In addition his code map(s) and description is in conflict with his sources. He clearly hasn't done his research carefully. But this is not surprising given the increasing degree of corporate amnesia. The Keevers document is not a necessarily a reliable reference--similarly to the reason that the IETF does not wish that Internet Drafts be used for reference--it was an internal file memorandum. I would think it was written in response to a request by one of the editors of the series "A History of Engineering and Science in the Bell System" which was started the same year as this memo. I furthermore suspect that this memo was not even researched by Keevers, but his staff. He just signed off on it. The click theory does not show up in the finished books, which is surprising if it were true, because the company generally prided itself for improvements for customers, that show their compassion for the public. This is evident in the related issue of designing new dial layout for reducing dialing errors, the research of which was described in detail. When you state that the "click theory certainly does make sense", then it is just your personal and uninformed opinion, you should have some sources as to the reasons for such a statement for use in the article. The article never made a definitive judgement, it just provided the facts, supported by references. Keevers is simply wrong that it was a guiding principle, because the actual guiding principles were stated in the original papers, and none of them considered this. At that time direct dialing by customers was not even a goal, it was merely a wish, if it turns out that the whole concept of Operator Toll Dialing could be expanded to customer dialing, but they had no technology at the time to foresee how that would come about. The references are clear about that too, and treat customer direct dialing as a concept to be first proven. (the article contains such a reference) In any case they expected customers to be dialing not with rotary dials in the not so distant future but with frequency signaling, for which the first trials were already underway. The area code system was not designed for DDD, but for Operator Toll Dialing, which used key pad with MF dialing in the vast majority of installations. The more you know about telephony of that time, the less the click theory makes any sense. It was spun by people without background to explain an apparently "obvious" pattern that only holds true for the Regional Center NPAs. Even the Keevers memo only states that the largest cities were assigned "short" codes, it does not state that small places got long codes, which is what the click conspirators routinely state. kbrose (talk) 16:01, 24 January 2022 (UTC)

Your tripping over technical aspects (and Lee is a distraction) is just sad, translators were located on the toll switches (#4) and those were not driven by dial pulses, but by MF. The whole Operator Toll Dialing system anticipated using an MF signaling system, not dial pulses, starting 1943. In some report, an author even lamented about the then current state of dialing, implying that dial pulsing is a part of history, not the future. kbrose (talk) 20:48, 24 January 2022 (UTC)

@Kbrose: While you dispute any connection between clicks and population, population density, or expected traffic levels, your own presentation of the data seems to provide a pretty indisputable basis that click counts determined the area codes that were used in the initial assignment. Surely you don't dispute that? Fabrickator (talk) 21:38, 24 January 2022 (UTC)

I have repeatedly stated, that there is no evidence that clicks determined area codes, and it does not make technical sense. But it is indeed a coincidence that low-click area codes are assigned to the Regional Centers, because they are assigned in the upper left corner, where area codes are numerically small. The Regional Centers are some of the largest population centers in the country (and for good reason of course), and especially of their respective regions (cf Atlanta and Denver). The low click count of them is a consequence of their position in the table, but not a leading reason for that placement. It is my personal opinion that this consequence is the reason that people later were misled info creating this click theory. Without knowing the history and the technology, and just looking at a list of area codes and their location, it is pretty unlikely to find any other reasons for explaining the correlation for the handful area codes of those large cities and their population. Limited data in.... poor interpretation out. If you have only two variables, area code and city population, it seems almost impossible to come up with any other conclusion. You get out what you put in. But it is a false correlation, and breaks down right outside of that upper-left rectangle. The Keevers memo is a rather obscure document, but it is I believe the only one that confirms that Bell Labs used the structure of the assignment table that is in the article. That structure is key to understanding the scheme. I believe the format to be a natural layout that an engineer or scientist would choose when confronted with the assignment task, they drew a table of rows and columns on the black board of the office, and the memo confirms this practice. But whoever worked on that memo, did not understand the history of the General Toll Switching Plan and perhaps not the role of Regional Centers in the 1930s and 40s. By the time it was written, the entire switching class hierarchy had been redesigned for almost 15 or 20 years. A staffer may have found it just as difficult to research the history as anyone today, and the short list of references in the memo, show the limited amount of work that went into the memo. I may have used all of them in the article already. But for writing this article, I have stringently avoided to inject my opinion, and focussed on letting the published facts and their historical significance lead the reader. The result is whatever the reader synthesizes from that. From your presentations here, you have concluded that it leans against the click theory. I tried to give the click theory some exposure, since it does have its documented existence, but you deleted it. Who is biased? kbrose (talk) 01:21, 25 January 2022 (UTC)

Bottom line: a coauthor of the chapter on Network Structures and Planning in Engineering Operations in the Bell System (p.103ff at [2]) -- that is, Keevers (more on whom here [3]) says code weight was a consideration [4]. Some guy on the internet -- that is, you -- disagrees, essentially saying that Keevers doesn't know what he's talking about, but has offered no source contradicting Keevers. Sorry, guy on the internet, you lose, but in the spirit of fair play and sportsmanship I'll give you a few days to find such a source before I go ahead and add appropriate text to the article -- though it would have to be an extremely strong source to overcome the authority of Keevers. EEng 03:14, 25 January 2022 (UTC)

EEng#s, your observations are very compelling, regardless of what kbrose thinks of it. This comment is not intended to diminish the points you have made, but is just another data point suggesting that area code assignments were not arbitrary.

Now just as I was about to suggest an alternative term to replace clicks and klicks, pulls and pulses, we now have weight. Each of these terms generally suggests some understanding of the implementation. I propose an alternative term "pips" or "pip count", which is just the sum of the digits in the area code, counting zero as ten since it is in fact the 10th position on the dial. Thus, area code "415" has 10 pips, the pip count for "512" is 8, and the "301" area code pip count is 14.

Moving on, kbrose asserts that area codes with low pip counts provide no material advantage either in terms of customer dialing time, operator dialing time, or switching equipment performance. Based on that, it would not have made any sense to assign area codes (i.e. with either a "0" or "1" middle digit) to one of the two NPA categories based on the relative pip counts.

There were 72 permissible codes with "zero" in the middle and 64 permissible codes with "one" in the middle. At the same time, there were 40 area codes to be assigned to single-NPA regions and 46 area codes to be assigned to multi-NPA regions. New codes for single-NPA regions would only be needed when there was a new state or province (a rather rare event), so it might have been expected to assign the 40 single-NPA area codes to the group that could accomodate only 64 area codes, and assign the 46 multi-NPA area codes to the group that allowed for 72 possible area codes. Yet the plan chosen allowed for a growth of 18 multi-NPA area codes, while assigning them the other way would have allowed for a growth of 26 multi-NPA area codes.

Having gone to the trouble of establishing a way for operators to quickly identify area codes for single-NPA regions, why would they have assigned it so that this technique would have to be be abandoned sooner rather than later? That seems really curious. Certainly the engineers expected that there would be growth in the number of area codes, and most of this growth would have been for the existing multi-NPA regions, while any growth in an existing single-NPA regions would have created an exception to the rule to help operators identify single-area code NPAs based on the middle digit. In other words, I'm suggesting that codes with low pip counts must have been less burdensome in some way or other, which conflicts with kbrose's claim that there was no such advantage to greater use of codes with lower pip counts. Fabrickator (talk) 06:01, 25 January 2022 (UTC)

• Pip is not a good choice -- it already means (to many readers) a quick tone or beep such as a "speaking clock" used to make. Click (informally) or pulse are well established.
• Weight is also well established in coding theory, though I don't see it used in the specific context we're discussing. There'd be nothing wrong with using it, if we explain it: One consideration in the original assignment of codes, was to attempt to assigned codes with the lowest "weight" (sum of the codes digits, taking 0 as 10) to areas with the highest expected inward traffic. The advantage of this would be that [etc etc]. That's not quite the best way of saying it, but you get the tidea.
• In these ponderous technical design efforts, the totality of the decisions don't always make perfect sense. It may even be, as whatshisname says, that worrying about code weight might have been of no technical value in the end, but that doesn't mean they didn't do it anyway -- the decision might have already been made and been irreversible by the time it became clear that code weight didn't matter.
• I'm not sure what you're saying there at the end, but we really should stop speculating.

EEng 15:09, 25 January 2022 (UTC)

This is so ridiculously long, I am adding a new section: #Weight-based area code assignment. Fabrickator (talk) 17:17, 25 January 2022 (UTC)

Weight-based area code assignment

This is a continuation of the discussion in the section with the questionable title of #Deletion of properly sourced content.

FWIW, my use of "pip" was in reference to the symbols on a die or playing card to indicate the numerical value. But I want to elaborate on this issue with the two sets of area codes, those with the middle digit being a zero or a one, because it further demonstrates that the "code weight" was of somewhat significant concern. So at the risk of repeating myself ...

The idea was that it would be helpful for operators to quickly determine if an area code was associated with a "single-area code" state or province, or a "multi-area code" state or province, because if they were able to recognize it as a single-code state or province, then the city didn't matter, hence the convention was adopted to distinguish this based on the middle digit. Of the 136 possible area codes, 72 had a zero as the second digit, while only 64 had a one. Which would they have expected to have the greater need for expansion? Area codes for single-area code states or area codes for multi-area code states?

I'm going to take a giant leap and assume that they would not be too concerned about an increase in the number of single-area code states, but they anticipated that there would be a need to expand the number of area codes in multi-area code states, and that somewhere down the road, they would run out. They could have chosen either the zero or the one as the middle digit, but choosing the "one" limited them to 64 area codes, choosing the "zero" would have allowed 8 extra area codes before they ran out. Yet, they chose the "one" as the middle digit for multi-area code states.

They chose the lower code weight for multi-area code states, even though they would run out of them sooner, and the most plausible explanation for this decision was that this rapidly growing segment of the area codes would have the benefit of lower code weights. Fabrickator (talk) 17:17, 25 January 2022 (UTC)

If you're saying what I think you're saying, that sounds like an interesting idea, but we really should be sticking to what sources say. EEng 05:26, 28 January 2022 (UTC)

The purpose of this observation was to identify a further problem with the claim being made disputing the idea of assignment being related to code weight. I'm not advocating that this belongs in the article, but it is one more observation showing that there was some concern about code weight. Also, as the article currently stands, there are the tables showing which codes were assigned (and associated text) purporting to further demonstrate some sort of pattern of code assignment. It seems to me that there isn't any reason to retain that. To put a finer point on this, while there's evidently some consideration of code weight and perhaps some pattern of assignment that can be observed, I am of the opinion that the article is better off without such content. Fabrickator (talk) 06:59, 28 January 2022 (UTC)

Why, if we've got a source like Keevers to attest to it? EEng 14:34, 28 January 2022 (UTC)

Not everybody perceives everything the same way, and the effort that was required to deal with this very self-righteus editor was, to put it mildly, excessive. Nevertheless, there is no sanction, in spite of his egregious behavior.

So on the one hand, I would like to see him gone forever, or at least for a very long time. On the other hand, consensus is an ongoing value. He disputes he was violating WP:OR or WP:SYNTH, and insists that anything that Keevers wrote about events that happened in the past was to be ignored. If this observation can help to achieve consensus, then it is a point worth making. To put it another way, why is it that when you make a point, it is the one that counts, and points made by others are to be ignored? If the editor in question continues to dispute in spite of multiple objections, it is the more powerful evidence against him. Fabrickator (talk) 15:22, 28 January 2022 (UTC)

From the history on this very page, I understand your frustration with kbrose's behavior. But I'm here now to help, and what we need to do is talk about the article's content, not past frustrations. EEng 17:29, 28 January 2022 (UTC)

So is it fair to say that you feel that the OR/SYNTH issues that I raised were unconvincing? Fabrickator (talk) 19:05, 28 January 2022 (UTC)

Not at all. It's clear that kbrose doesn't understand WP:V and WP:OR and all kinds of other stuff, and everything in the article needs to be checked by tracing it back to he sources. EEng 04:48, 9 February 2022 (UTC)

I suggest that reverting back to the 19:04, 17 November 2021 revision would be more in line with the article title (this retains how we got from the limit of 136 area codes to interchangeable central office codes, and dropping the fairly extraneous (and evidently contentious) issue about how area codes were assigned, aside from the single-state/multi-state distinction (though we would want to retain some of the interim changes, such as deprecation of the Sinks citations). Of course, we don't have any reason to think that kbrose is gone forever, but maybe he'll come to grips with the reality that he does not actually own the article. Fabrickator (talk) 08:53, 9 February 2022 (UTC)

I don't see why we wouldn't include what Keevers says about how codes were assigned; I don't see anything contradicting that other than kbrose's OR. Also, I suspect this article should be merged with the NANP article, with which there's a lot of overlap.

When kbrose returns he'll need to either abide by policies and guidelines or go do something else. There's no need to talk about him anymore until he shows up. EEng 15:40, 9 February 2022 (UTC)

The short explanation of why we should avoid the issue of the initial assignment of specific area codes is that we have a limited understanding and there seem to be inconsistencies. That just leads to speculation, and this issue is really just an aside that doesn't help much. As for your suggestion to merge (presumably the three article about the past, present, and future of NANP), the existing arrangement is a logical separation that helps maintain coherence and functionality. But it sounds like it's time for me to practice being BOLD. Fabrickator (talk) 17:42, 9 February 2022 (UTC)

Can you give me the short explanation? What are the inconsistencies? Why is it complicated to just repeat what Keevers says? EEng 18:45, 9 February 2022 (UTC)

First off, we see the conflation of different measures of population ... population vs. population density, forecasts of population, inbound traffic. And is population (however it might be defined) the sole consideration or were there other considerations? (BTW, the stuff about diagonal assignment patterns, that's just noise, though it does demonstrate a blind spot in kbrose's thinking, since it appears to provide strong evidence that assignment was in fact based on NPA weight.)

There are at least a couple of seeming anomalies in the assignments vis a vis population ... though these require speculation. But notwithstanding the frequency with which the claims about assignment population are made in various publications, knowing the basis on which assignments were made contributes little to the understanding of the early NPA assignments and how that transitioned to the "interchangeable NPA" situation that it evolved to, which is what I think of as the primary focus of this article. People who just want to know about current area codes have a separate article for that, they shouldn't have to wade through this historical perspective, and vice versa. Fabrickator (talk) 19:32, 9 February 2022 (UTC)

Look, what Keevers says is that it was recognized that "large cities" should get the lighter codes, and then he names some of them. So that's what we say -- we don't get into speculation or debates on population or anything like that. We just say what Keevers says (and maybe some other sources IF they're of comparable authority, or supplement him somehow). I think, in fact, that of the many points about why codes were constructed and assigned as they were, this is perhaps the one that a lay reader will most readily understand.

People who want to know about current codes should look in some online resource -- WP:NOTDIRECTORY. NANP is about historical development, and this article is about historical development, and they are naturally presented together, assuming that the merged article wouldn't be too big. With the material they both have as of now, they wouldn't be too bit, especially after duplication between them (of which there's a LOT) is removed, along with kbrose's unsourced BS. EEng 21:26, 9 February 2022 (UTC)

Area code pattern tabulation

@Kbrose:This is in regard to the paragraph that starts with the sentence "The pattern of this assignment of area codes is shown in the following tabulation" (in the "assignment plan" section of the 11:12, 21 January 2022‎ revision). I am trying to puzzle this out. It suggests that there was a sequence of assignment, i.e. from low-numbered codes to higher-numbered codes, specifically, assignments going from low-number codes in the upper-left hand corner diagonally towards higher-numbered codes. Am I misreading this? What is meant by "low-number" and "high-number"? Is "209" to be considered as higher than "301"? That would seem to require some clarification. Fabrickator (talk) 22:20, 21 January 2022 (UTC)

The article doesn't use the term higher-numbered, only low-numbered, I think, and it does not describe the process from low-number to high-number. instead it focusses on the lop-left corner, the center, and the lower-right corner. What is so difficult about that? There really shouldn't be any surprise what low-numbered means, 209 is lower than 301, as simple comparison would suggest. But the sentence could indeed be phrased a little better, perhaps that the table was filled from the top left corner, containing the low-numbered area codes in each row, toward the center and lower right corner. Perhaps the clause of low-numbering is too trivial to be included at all. But it is puzzling that one would not understand the description of the pattern, the pattern is so visually striking that almost no explanation is needed. Tabulating the NPAs this way is not original research, either. kbrose (talk) 00:42, 22 January 2022 (UTC)

Are you suggesting that there's something significant to this pattern of diagonals? Fabrickator (talk) 07:34, 22 January 2022 (UTC)

Not suggesting anything for use on WP. Clearly the first set of assignments was not by diagonals but essentially the top-left block (rectangle) for the Regional Centers. These HAD to have area codes first before anyone else, because of their essential status in long-haul routing (otherwise the Operator Toll Dialing system could not work anywhere larger than a small region), and indeed they got the first installations of 4XBs. But the 'diagonal structure' does contrast the pattern later when new area codes were scattered somewhat more randomly across the lower triangles in the sequence of assignment, which we can follow since we know the years of assignment for each. This is more or less OR probably, but I can think of one reference that might be helpful with some analysis. It is really out of the scope for the "original" NPA codes, and I wanted to keep the expansion section limited to just the broad strokes of history for context. kbrose (talk) 17:06, 22 January 2022 (UTC)

By the way, the reason I was unclear as to whether "209" or "301 would be considered lower was because I thought that maybe you were considering the "klick count" rather than the area code's actual numeric value. As to the diagonal pattern that you have pointed out, I agree (based on your explanation) that it was essential for the regional centers to be included in the initial implementation of NPA assignments. Fabrickator (talk) 05:58, 23 January 2022 (UTC)

I just can't help but continue to be amused at the explanation as provided in the article. Assuming that kbrose is acting in good faith, he seems stuck on what he presumes to be some kind of inherent preference for a diagonal pattern of code assignment, which accounts later assignments as an extension of the original code assignments. To be clear, nothing should be considered to preclude that there could be multiple considerations in assignment of area codes ... if the intent was to provide low-weight codes to the pre-existing regional centers, and they just happened to have high populations, that doesn't preclude the consideration of population in assignment of codes in the remaining areas.

I remain perplexed about the "obviousness" of diagonal assignment, though of course, I must be missing something, because diagonal assignment corresponds exactly to code weight.

Should the actual metric used to assign low-weight codes be population, the installed based of phones, or the amount of telephone traffic directed towards that area? They are not the same, but without some compelling theory to the contrary, one can posit that the correlation between them would be strong. I believe it was postulated that we should reverse the logic, rationalizing that low-weight codes should have been assigned to the areas with lower populations, because a greater number of calls would be coming from them to the high population areas than vice versa. Perhaps the system engineers had the same blind spot about this logic that I have.

Last but not least, we're supposed to ignore the sources that support the population/weight-based assignment because the sources supporting such claims have all been impeached. And if we accept this premise that such sources have been impeached, then we are merely guessing and/or rationalizing.

If we have no reliable source, theorizing about this would be in violation of WP principles, so we should not be conjecturing about this. But there we are, evidently making unsupported claims about this, regardless of the lack of reliable sources and the asserted dubiousness of the evidence. Fabrickator (talk) 07:19, 22 March 2024 (UTC)

416 & 613

User:Kbrose keeps adding erroneous info to the State, Province, or Region section of area codes 416 & 613. Kbrose writes "northern part resulting from a westward dividing line from Oshawa on Lake Ontario)" and (southern part, including Toronto, below a westward line from Oshawa). To someone with knowledge on the subject, this sounds ridiculous on so many levels.

For one, more than half of southern Ontario was in the 613. Two, "northern part" sounds like a 5th grade remedial book report. They are real provincial divisions. Three, I'm not sure why kbrose decided to arbitrarily choose Oshawa as the division point. It actually was not Oshawa, it was a cluster of different municipalities further east & north. Four, no other suburban town names are used in any other division descriptions in any other area code except for 416/613. Very peculiar and incorrect information.

• 416 should read: Ontario (Part of southern Ontario, including Toronto)
• 613 should read: Ontario (all of Ontario, including Ottawa) or: Ontario (All of Ontario, excluding part of southern Ontario)

Something to this nature. These regions are actually officially named regions with real boundaries, not just arbitrary names that an wikipedia editor with no knowledge of the area (kbrose) should just create out of nowhere. We wouldn't say "the southern part of California", we would say Southern California, because it is an officially defined area.

I request that someone revises these descriptions, as what is shown currently is incorrect. 76.69.7.202 (talk) 17:01, 22 January 2022 (UTC)

"All of Ontario, including Ottawa" is clearly wrong for 613, as that would include NPA 416. NPA 416 was also part of "all of Ontario". Emphasizing "Southern Ontario" here is really no more useful than using "Ontario" as the NPA dividing line was completely independent of other recognized divisions. Oshawa is roughly the easternmost significant community that fell into 416. There are no public documents, AFAIK, that specify the exact line, it was probably just roughly drawn by hand on a map of tributaries, making sure that no toll routes got cut, letting the local administrations work out the details later. I will look at the map of the Trans-Canada line for more clues. Any other major toll lines probably just went south into the US anyways.Writing in a positive sense, "Region south of line XYZ", is far better than "All of A, except part of B) where B is not clearly defined. kbrose (talk) 17:34, 22 January 2022 (UTC)

The way the line appears drawn in the eastern end is a loop around Oshawa, probably including small places with tributaries on the east. Or just west of Oshawa. Newspapers of the early 50s may help. kbrose (talk) 18:58, 22 January 2022 (UTC)

This looks to be quite correct, but it is possible that the most eastern toll switching point in NPA 416 may not have been Oshawa, but Port Hope, about 30 miles east of Oshawa. From the printed maps of the area codes it is pretty much impossible to tell the exact starting point of the line, and Oshawa is a lot bigger place than Port Hope. But I know, that Port Hope had a tributary line from Colborne, another 20 miles east, which was likely the eastern-most end office in 416. But this is based not on data from 1950, but later. In 1950 or earlier Port Hope may have been a tributary to Oshawa, until it got its own switching center. So, Oshawa is a pretty good estimate of the starting point for the 416/613 line. kbrose (talk) 15:29, 23 January 2022 (UTC)

Again, your extreme article ownership continues. "I will look into this" as if you are the judge, jury, and executioner of this article. Oshawa is incorrect. Whether it is more known than Port Hope is irrelevant. Neither should be named as the boundary between the area codes goes well beyond the municipal boundary. 416 included parts of southern Ontario, and 613 included Ontario excluding parts of southern Ontario. Your personal original research on this is again irrelevant. It is abundantly clear that you have zero expertise or knowledge whatsoever in the local geography. Maybe step aside and let editors with some actual knowledge handle this one thanks. 76.69.7.202 (talk) 16:50, 23 January 2022 (UTC)

Well, why don't YOU show some sources for your claims, instead of continuing the constant personal attacks. At least I did look into this in good faith. Which method one uses to locate the boundary can vary, but NPA boundaries are either specified by state borders, or by toll centers. When using toll centers the tributaries are automatically included, without having to go into more detail than necessary. If you know all the details as you claim, then it should have been easy to recognize that your proposal for the text was just wrong, or insufficient. kbrose (talk) 18:32, 23 January 2022 (UTC)

That's not how wikipedia works. You changed the wording not me. You also changed it to something incorrect and unreferenced. It's very clear you don't understand the topic or local geography. I've never encountered someone with such high confidence in a subject they clearly know nothing about. You could be the perfect Dunning–Kruger case study. Now add a source for the rubbish you've added, or restore it to the original text. Otherwise I'd be happy to open another ANI on you if you want to continue your behaviour and article ownership. 76.69.7.202 (talk) 23:08, 3 February 2022 (UTC)

Deprecating Sinks

The references section of Keevers (p.8 [5]) says that William A. Sinks, New Telephone Numbers for Tomorrow's Telephones, Bell Telephone Magazine Vol.40 pp.6–15 (Winter 1959/60) (currently cited in the article) was "intentionally not listed due to errors in content". I've eliminated reliance on Sinks by citing Keevers instead. EEng 06:00, 28 January 2022 (UTC)

proposal to revert

@Piledhigheranddeeper: I propose that the edit of 13:53, 1 September 2022 by User:Piledhigheranddeeper be reverted.

Here is my argument: The phrase "the years immediately following World War II" could be essentially any span of years, presumably without any gaps, whereas "the immediate post-WWII period" implies some fairly short period of time.

The intention here is to leave it somewhat vague, just that this occurred following the war (while avoiding overly casual wording), but it should not be so imprecise as to leave the reader thinking that this might have gone on for what could have been perhaps 7 years, 10 years, or even longer.

I prefer to avoid doing this revert unilaterally, so I'm looking for a "second" to concur on this revert. Fabrickator (talk) 16:41, 1 September 2022 (UTC)

@Fabrickator: I observe that "immediately" is the adverb form of the word "immediate"—essentially the same word in each passage. It's hardly a change at all, except avoiding the string of four straight modifiers, three of them nouns, with an abbreviation. Both of the strings seem to imply the late 40s, and anyone seeking more detail can quickly find it in the body text. I'm willing to discuss more. --Piledhigheranddeeper (talk) 16:58, 1 September 2022 (UTC)

Both the same. EEng 01:12, 2 September 2022 (UTC)