Sunday, August 25, 2013

Things I don't understand

Thanks for bookmarking my new blog.  So far it sucks but I'm gonna keep after it. :)

I'm going to change a few things right off the bat, using my reserved right there in the fine print.  I do have more long-form things to post -- the General Systems Theory piece will be at minimum fun, and a bit mind-bending, if you're into that kind of thing.

But I'm not sticking to just long-form stuff, because I realized the advantage of a blog is its clear ownership. On public media, you risk looking like a dork if you stand on a soapbox.  But a personal blog is a place you can try that stuff, especially if you invite discussion, as I always try to do.  So that's the criterion: stuff I wouldn't post on Facebook between the food porn and puppy pix.

To sum up: I'll keep bloviating here about things I believe I understand, and be open to reasons to rethink these beliefs.

But there's another category of things, ones that I admit I simply don't understand. I invest time in trying, because logic and evidence seem to dictate that they should be different from the way they actually are.   My brain glitches when I see these incomprehensible things, and it turns out there are a lot of them.  So maybe "Things I don't understand" can be a regular blog topic if people find it interesting.

First off, I accept that there are things I'm just not wired to understand.  For example, over-the-air television feels exactly like magic to me.  But there are people who do understand it, so I don't actually need to.

No, I'm talking about things I actually have tried hard to understand, but simply cannot reconcile to logic.  Some of them are kind of strange, ok.  Many are in the areas of economics or politics, which makes sense; there are also a lot of them in the areas of roads and transportation.

For example, this one has bugged me for decades.  You see a lot of dented guard rails, smudged jersey barriers, and scarred trees, but you don't see that many accidents or dented vehicles, do you?  Even considering many vehicles over time, that doesn't quite reconcile to me. I know the math must work, based on the evidence.  But it just feels that those numbers don't match, you know?

Ok, that's a silly example.  Here's another -- one that I actually know something about and have thought through, yet still can't reconcile.

In 1988, New England Telephone was running out of numbers in the 617 area code, which covered central and eastern Massachusetts.  They were faced with two choices:
  1. They could institute 10-digit dialing, and the new 508 area code would overlay 617
  2. They could split 617 apart, and assign everyone outside of the Boston metro area to 508
If you think about that for even a second, you realize that overlaying the numbers is better because:
 a) Everything about 617 continues to be true
 b) 508 just extends 617 capacity
 
So with overlay, anyone with an existing 617 number (which was everyone) would see no change.  New numbers would be assigned to 508, but be in the same well-defined geographic area as 617.  Essentially, 508 and 617 become the same thing as 617 always was by itself, so your explanation is very simple.

One more critical piece of data: Overlay had the disadvantage of forcing everyone from 7-digit dialing to 10-digit dialing, but it was mathematically certain that 10-digit dialing was coming anyway.  The explosion in first, fax numbers, and then mobile numbers, was well understood by everyone in the industry.  10-digit dialing was inevitable, and it was going to be soon.

You know what happened.  They chose to split 508 out from 617.  This forced millions of people -- from Provincetown at the tip of Cape Cod, to the Quabbin area where I live (one town from area code 413), to Salisbury on the New Hampshire border -- to change the way EVERYBODY contacted them.  They needed to print up new forms, cards, stationery, etc., and change the way they were listed in myriad places (remember, no internet listings in those days).

I remember the arguments at the time.  There were sweet old ladies with rotary phones, who would be horribly inconvenienced by dialing three more numbers every time they wanted to call their families.  I'm not kidding, that was the argument against all technical and logical counterarguments.

Folks, all those old ladies are DEAD now.  And we are stuck with a completely unintelligible area code system in Massachusetts.

To make it worse, they kept doing it.  In 1997 (less than a decade later), 617 was overloaded and they split out 781 (suburban Boston ring), and 508 was overloaded and they split out 978 (northern towns).  So again, they made everyone change everything.  Those in 978 got to change for a second time.

But the final irony is that, by 2001, they apparently had somebody in charge who could grok this concept.   Overlays were added to 617 (857), 781 (339), 508 (774), and 978 (351).  Presumably it will operate this way going forward.  But who knows?  Maybe phone numbers go away altogether, subsumed by the much more flexible and comprehensive identity models of today's systems.

Anyway, those are some examples of things I don't understand.  I have some more I'll add in future posts (political primaries WTF?).  I welcome your comments or corrections.

4 comments:

  1. Dave Ross solved the Area Code mystery for me over on Facebook. He suggested that the phone company chose the bad option because it's cheaper for them, under the assumption that for-profit organizations are always driven by cost.

    One problem there. In addition to being wrong as a policy/economic decision, the code-splitting option cost *way* more for New England Telephone. Reprogramming all the switches, communicating to and supporting the public, that's a big deal at that scale and it cost a fortune. That expense pales in comparison to the costs incurred by customers, but it's a critical point.

    I completely forgot that New England Telephone was, and is, a regulated monopoly. Its profit is determined by a formula based on its (justifiable) costs. So they just needed to prove that its choice was justifiable, and I'm sure they had plenty of experts say that it was. I worked for the phone company's parent from 1989-1993 (NYNEX, later Bell Atlantic, now Verizon), and saw all sorts of gold-plating in the telco division that you avoided in the unregulated parts. (I sold phone systems, but we worked with the telco's Centrex/T1 sharks.)

    So it never was about the sweet little old ladies, it was about generating nonrecurring profit over a period of what, six or eight quarters. New England Telephone acted completely rationally, given the incentives offered. It's like the chapter in "Freakonomics" that shows real estate agents sell harder, and charge more, for their own properties than for their customers'. NET's transient profit was much smaller than the overall costs to the public, but it was more than enough to drive their choice.

    THAT explains it perfectly. See, these are answers that exist, folks, it just takes collaboration to bring them out. :) Thanks Dave!

    ReplyDelete
  2. finally able to post via google account.

    ReplyDelete
  3. with a CAPCHA each time. Arghhh!

    ReplyDelete

Play nice please. Mean stuff will be deleted.