The World of Computer Software/The World of Computer Software.iso/tc13-021.zip/TC13-021.TXT

Text File | 1993-01-12 | 20KB | 514 lines

TELECOM Digest     Tue, 12 Jan 93 02:32:20 CST    Volume 13 : Issue 21

Index To This Issue:                     Moderator: Patrick A. Townson

    Re: Format of ZIP Code Bars on Envelopes (Sean Case)
    Re: Format of ZIP Code Bars on Envelopes (Steve Hutzley)
    Re: Format of ZIP Code Bars on Envelopes (Sean Malloy)
    Re: Format of ZIP Code Bars on Envelopes (Zealand R. Hatch)
    Re: Format of ZIP Code Bars on Envelopes (Michael Rosen)
    Database of Area Codes, NPA-s and Zipcodes Wanted (John Villalovos)
    Re: Hunt Groups (John Rice)
    Re: Hunt Groups (Bill Huttig)
    Re: Hunt Groups (Barton F. Bruce)
    Re: Hunt Groups (Steve Forrette)
    Re: Hunt Groups (Stephen Diercouff)
    Re: Hunt Groups (Dave Levenson)
    Re: Hunt Groups (Jim Knight)
----------------------------------------------------------------------

From: [e-mail] (Sean Case)
Subject: Re: Format of ZIP Code Bars on Envelopes
Date: 12 Jan 93 00:08:56 GMT
Organization: Australian National University


[e-mail] writes:

> Curious enough to hunt for another factoid ... does anyone know what
> fraction of the entire snail mail in the US is generated by business
> and how much by Aunt Agatha ?

> Should give some insight into how much mail the automatic sorters
> actually can ever benefit from. And if the dreams of USPS to read
> human handwriting are so important?

Here in Australia, most envelopes are sold with preprinted boxes for
the (four-digit) postcode. Australia Post claim that their equipment
can read hand-scrawled postcodes so long as they're in the boxes.
Typescript addresses don't have to use the boxes, but pushing the
postcode out to the right of the address is supposed to help.  They
don't say what sort of error rate they have, though.

I heard a story some time back that the good ol' USSR had envelopes
with a mark sense area for the postcode (possibly using a
seven-segment layout). Mail with unreadable postcodes was discarded.
That's certainly one approach ... is this just obsolete propaganda?


Sean Case   [e-mail]

------------------------------

From: [e-mail] (Steve Hutzley)
Subject: Re: Format of ZIP Code Bars on Envelopes
Organization: Digital Equipment Corporation
Date: Mon, 11 Jan 1993 23:22:33 GMT


In article <[e-mail]>, [e-mail].
com (say,halim s) writes ...

> I would like to find out the encoding for the ZIP code bars on USPS
> mail envelopes.

> I could figure out this much:

> Ones are long bars, zeros are short bars.  Start and end with a one.
> Each digit has five characters.  I could figure out what zero through
> nine correspond to.

> The only remaining part is the check digit at the end.  Only one check
> digit is used both for five digit regular ZIP code and nine digit
> extended code.

> Now my question is: What is the encoding/decoding rule for this single
> check digit?

> Does anyone know or could you tell me a reference for this "standard"?
> I am sure some address generator software would have this algorithm.

See {PC Magazine} a few months ago (November/December), they had the
whole scheme in there. Its kind of an inverted BCD where the LONG ONES
are 0 and the SHORT ONES are 1 (I think), and there were CLOCK pulses
in there to compensate for the speed variances of the reading machines.


Steve

------------------------------

From: [e-mail] (Sean Malloy)
Subject: Re: Format of ZIP Code Bars on Envelopes
Date: 12 Jan 1993 04:30:42 -0600
Organization: Texas A&M University, College Station, TX


> Also, Is there a summary of several bar codes ftp-able somewhere ?

As an aside: On wuarchive.wustl.edu in /mirrors/win3/fonts/truetype
are two fonts that might relate to this topic:

usps_bar.zip and barcod39.zip

I make no claims about their accuracy, but if you're running MS
Windows 3.1 then they might help.


Sean C. Malloy  -  Texas A&M University  -  [e-mail]

------------------------------

From: [e-mail] (Zealand R. Hatch)
Subject: Re: Format of ZIP Code Bars on Envelopes
Date: 11 Jan 93 19:28:57 GMT
Organization: Amdahl Corporation, Sunnyvale CA


I'm not sure what this has to do with telecom but having recently
investigated this I'll share what I know.

There is a group of publications available from the Post Office.

Publication 25: Designing Business Letter Mail (Aug 1992)

  This publication replaced the previous Pub. 25 "A Guide to Business Mail
  Preparation".  The major sections from the index are.

  1.0  Introduction
  2.0  Basics of OCR and Barcode reading
  3.0  Physical Characteristics of Automation Compatible Letter Mail
       ("Machinability")
  4.0  Addressing for Automation
  5.0  Postnet Barcode
  6.0  Facing Identification Mark (FIM) Patterns
  7.0  Preparing Reply Mail for Postal Automation

  It's well writen and has all of the information you need to "produce" a
  barcode and put it where it needs to go on the envelope.

Publication 28:  Postal Addressing Standards

  The major sections from the index are.

  I    Introduction
    A  Background
    B  Overview
  II   Postal Addressing Standards
    A  General
    B  Last Line of the Address
    C  Delivery Address Line
    D  Rural Route Addresses
    E  Highway Contract Route Addresses
    F  General Delivery
    G  Postmaster Addresses
    H  Post Office Box Addresses
    I  Puerto Rico Addresses

  This publication is "a comprehensive guide to all styles of addressing".
  Not required reading if what you're interested in is barcodes.

Publication 67:  Automation Plan For Business Mailers

  This publication outlines the Post Office's plans for business mail.
  Interesting as a general overview.


Your local Postmaster should be able to obtain these publications for
you without cost.

As to the question of how does this help with the hand written
address, the SORTING of letters is being converted to barcodes and a
letter is sorted a bunch of times before it is put in your mail box.
So the plan is when a letter initially comes into the post office it's
first checked for a barcode. If it is there the letter goes off to the
sorters. If not, the address is checked to see if the City, State and
ZIP Code are machine readable. If they are the barcode is applied to
the envelope and its off to the sorters. If not the letter is queued
for manual reading. Again the barcode would be applied and then its
time for the sorter.  Not all post offices have all of the equipment
needed for all of these steps yet.  And it will be 1995 before all
letter mail will be automated.


I hope this helps; Happy Mailing,

Zealand Hatch    - 408 746-8720 -   {where_ever}!amdahl!zrh
Amdahl Corporation  -or-  [e-mail]

------------------------------


From: [e-mail] (Michael Rosen)
Subject: Re: Format of ZIP Code Bars on Envelopes
Organization: University of Denver, Dept. of Math & Comp. Sci.
Date: Tue, 12 Jan 93 04:59:38 GMT


[e-mail] writes:

> In TELECOM Digest 12-928, [e-mail] (say,halim s)
> asks:

>> I would like to find out the encoding for the ZIP code bars on USPS
>> mail envelopes.

> Ask the post office for the circular on bar code.  It's a small
> pamphlet of about 5 pages, and it explains how all of the codes work,
> as there are two codes; one for ZIP code and one for ZIP+4 code.
> There is no charge for the pamphlet.  If your local Post Office
> doesn't have it, try a main office in a large city.

I have that booklet.  The title is "A Guide to Business Mail
Preparation," Publication 25.  Here is the address for comments or
suggestions and I believe this is where I ordered it from:

Marketing Dept
Regular Mail Services Div
U S Postal Service HQ
475 L Enfant Plz SW Rm 5541
Washington DC 20260-6336

They describe the specs for the Postnet bar code and the FIM patterns.
As for the algorithm, I'll have to find my {2600} issue with the article
(where I found the address for the publication originally).  There's
52 bars in a ZIP + 4 Postnet bar code, with two long bars on either
end.  I can't remember the exact methodology for the bars in the
middle at the moment, I'll repost when I find the {2600} issue with the
article.

Also, I have a program I ftp'd at one time called envlj.  Scan for it
via archie, etc.  It will print both FIM patterns and Postnet bar
codes when you supply a complete ZIP + 4 and only a Postnet bar code
when using a five-digit ZIP.


Michael Rosen    Tau Epsilon Phi - George Washington University
[e-mail]    [e-mail] or @lambada.oit.unc.edu

------------------------------

From: [e-mail] (John Villalovos)
Subject: Database of Area Codes, NPA-s and Zipcodes Wanted
Date: 12 Jan 93 07:13:58 GMT
Organization: Oregon State University


I was wondering if anyone could point me in a current or fairly
current list of Area Codes, NPA's and Zipcodes.  I wanted to be able
to set up an application where people could tell me their Zipcode and
I could figure out their Area code and possibly some of the NPA's for
the Zipcode.  I once had a program like this but I don't anymore.  It
was put out by AT&T and the data was somewhat out of date.


Thanks,

John Villalovos   Certified Netware Engineer
[e-mail]   [phone]


[Moderator's Note: I'll tell you who keeps a great data base on this:
our reader Carl Moore has lots of files on this. I'm sure when he sees
this he will contact you directly.   PAT]

------------------------------

From: [e-mail]
Subject: Re: Hunt Groups
Organization: Teradyne Inc., Telecommunications Division
Date: Mon, 11 Jan 93 23:14:45 GMT


In article <[e-mail]>, [e-mail] (Rob
Boudrie) writes:

> Is it possible for two numbers serviced by the same physical CO to be
> placed on a hunt group even though they has different exchange
> prefixes?

> [Moderator's Note: I've never heard of a hunt group including lines on
> different prefixes, even if they were in the same CO.  Anyone?   PAT]

Not a hunt group, but 'forward on busy' should work if the lines are
in the same switch. It accomplishes the same end.


John Rice        K9IJ    [e-mail]


[Moderator's Note: But John, the big difference between the two at
least where Illinois Bell is concerned is that 'forward on busy' is
handled like any call-forwarding, with a unit (or more) charged on
each call. On the other hand, IBT gives 'hunt' for free. Same
identical feature, but the one is charged for call after call after
call, making it quite expensive if you get several thousand calls each
day with 99 percent or more of them going somewhere other than the
lead number. IBT tried to get *me* to take that feature on a voicemail
system I look after instead of traditional hunting. I said nuts to
that and insisted on regular hunting (and checked the bill the next
month to be sure they obeyed!)   PAT]

------------------------------

Date: Mon, 11 Jan 93 21:57:38 -0500
From: [e-mail] ( Bill Huttig)
Subject: Re: Hunt Groups


In SouthernBell land there are rotaries on two or more exchange ... I
would recomend (if you can get it from your carrier) that you get call
forward on busy between the lines ... line 1 on busy forwards to line
2 line 2 on busy forwards to line 1 ... I have my lines set up that
way and I only have to pay $1 for each forward on busy ... a rotary
would have made be pay about $16 more per line ... and call forward
busy does the same thing.


[Moderator's Note: Well this seems to be an applications thing based
on the volume of calls received. IBT does *not* charge any special
fees for hunting. It is free as lomg as all the lines are on the same
exchange. You can pay $1 per call forwarded on busy if you wish, but I
thought it was too expensive when IBT said each forwarded call would
cost one 'unit' -- about 3.5 cents here. Even at three cents per call
forwarded, I'd wind up paying several hundred dollars per month for
this service, which is useless considering the same thing is free from
IBT if you ask for 'hunting' instead.  PAT]

------------------------------

Date: Tue, 12 Jan 1993 01:19:12 -0500 (EST)
From: Barton F. Bruce <[e-mail]>
Subject: Re: Hunt Groups
Organization: Cambridge Computer Associates, Inc.


[e-mail] (Rob Boudrie) wrote:

> Is it possible for two numbers serviced by the same physical CO to be
> placed on a hunt group even though they has different exchange
> prefixes?

Sure is! but depends on local CO equipment. Or there are tricks you
can play.

Most CO switches cover several exchange codes. Telco may bitch and
scream, but often will hunt between exchanges on their own just to
make their life easier or to get out of some messup they caused. You
may also get number in a hunt group that are 'theoreticals' that have
NO exchange mere customers can dial (can YOU dial 017-9325? I thought
not!).

If you can't get them to do it or it is IMPOSSIBLE (may be on
different CO switches, not just exchanges), then, if you have NO
message unit implications and NO problems paying for a phone line that
only is a number, get call forwarding on the isolated single one and
set it to forward and GROUND both Tip and Ring to PERMANENTLY busy it.
If they bitch, tell repair you are having some equipment problems and
you DID busy it, but if they could DO IT FOR YOU while you get your
equipment straightened out, and that you will CALL THEM when you need
it restored, they probably will leave it that way for ten years.

All calls to that number will then forward to the other numbers were
you can have several lines.

If you can get few line centrex service, you may be able to get line
in use forwarding - more a secretarial feature than hunting.

> [Moderator's Note: I've never heard of a hunt group including lines on
> different prefixes, even if they were in the same CO.  Anyone?   PAT]

Many times. No problem.

------------------------------

From: [e-mail] (Steve Forrette)
Subject: Re: Hunt Groups
Date: 12 Jan 1993 02:15:55 GMT
Organization: Walker Richer & Quinn, Inc., Seattle, WA


In article <[e-mail]> [e-mail] (Terry

Kennedy) writes:

> However, the various call forward functions will work across
> prefixes, or across switches if desired. So the subscriber could order
> call forward on busy (or on busy/no answer) and then set whatever form
> of hunt group up that he wanted. This could be done either with
> customer-selected for- warding (the kind usually explained in the
> phone book, where the customer sets the forwarding, or with fixed
> forwarding, where it is set up in the switch and the customer can't
> change it).

One problem with using call-forward-on-busy as a creative way to
create hunt groups of more than two lines is that many telcos restrict
it to handle no more than one forwarded call at a time.  This
restriction may also be placed on Remote Call Forwarding numbers and
regular (immediate) call forwarding.

With US West in Washington State, RCF costs around $18 per month plus
message units, and can handle only one call at a time.  You can
purchase additional "circuits" for $18 per month each.  But, if you
can find somewhere to terminate a loop, you can get an unmeasured
business line for around $32 per month, add call forwarding for around
$4.  Then, for $36, you get up to 20 concurrent calls forwarded with
no message units.  The 20 call maximum was what call forwarding was
configured for on my switch - the person at the business office had to
check this out when I asked, and I got the impression that different
switches had it set to different values.  Pacific Bell on the other
hand, sets all of their switches to the same value: they only allow
one concurrent call for regular call forwarding.

Also, if you want to have a busy or no-answer forwarding go to a
number that's on a different switch, it is technically a different
feature: "busy call forwarding extended."  It was not widely available
until a couple of years ago.  Since this feature is a necessary part
of the telco's own voicemail offering, it is now magically available
most everywhere now.  But some switches may not yet support it, so
busy-forwarding may be restricted to the local switch.


Steve Forrette, [e-mail]

------------------------------

From: [e-mail] (Stephen Diercouff)
Subject: Re: Hunt Groups
Organization: tfm Associates, Ltd.
Date: Tue, 12 Jan 1993 03:32:39 GMT


[e-mail] (Steve Forrette) writes:

> In article <[e-mail]> [e-mail] (Rob
> Boudrie) writes:

>> Is it possible for two numbers serviced by the same physical CO to be
>> placed on a hunt group even though they has different exchange
>> prefixes?

> I had a similar setup recently, but strictly speaking, it was
> busy-transfer rather than hunting.  It was on a 5ESS switch.  The fact
> that I wanted the busy-transfer was known at the time I placed the
> order for the two lines, and the computer happened to select numbers
> on two different prefixes.  I would imagine that on a modern SPC
> switch, the prefix difference would not matter.

I have three numbers, with three different prefixes, in a rotary hunt
group. The latter two were assigned at the time I ordered the hunt
group. If it makes any difference, the switch is a 5ESS, and the
territory is USWest.


Stephen Diercouff, tfm Associates, Ltd., Bellingham WA  voice: [phone]
Internet:  [e-mail]                                    fax: [phone]
UUCP    :  uunet!nwnexus!tfm!sgd
Snail   :  P.O. Box 5084/Bellingham WA 98227-5084

------------------------------

From: [e-mail] (Dave Levenson)
Subject: Re: Hunt Groups
Organization: Westmark, Inc.
Date: Mon, 11 Jan 1993 18:52:42 GMT


In article <[e-mail]> by [e-mail], Pat
adds:

> [Moderator's Note: I've never heard of a hunt group including lines on
> different prefixes, even if they were in the same CO.  Anyone?   PAT]

In Brooklyn, NY, I have a customer with a 12-line hunt-group (which is
used by callers to access a voice-response banking system).  Most of
the lines in this hunt group are 718-388-xxxx numbers, but three of
them are in another prefix.  When New York Telephone was installing
them, I asked the installer if there was a mistake in the line
numbering, and if he would verify that hunting worked correctly
throughout the 12-line group.  He verified the directory number of
each of the lines, and he and I worked together to verify that hunting
worked as ordered.

The serving central office is a Northern Telecom DMS-100 which was cut
into service less than three years ago.  The two prefixes are served
by the same switch.


Dave Levenson        Internet: [e-mail]
Westmark, Inc.        UUCP: {uunet | rutgers | att}!westmark!dave
Warren, NJ, USA        Voice: 908 647 0900  Fax: [phone]
[The Man in the Mooney]

------------------------------

From: [e-mail] (Jim Knight)
Subject: Re: Hunt Groups
Organization: UMCC
Date: Tue, 12 Jan 1993 04:06:15 GMT


M-Net A Public Access Unix System here in Ann Arbor has two different
prefixes in it's trunk hunt.  We had no trouble getting the phone
company to add two numbers from another site into our trunk hunt.  We
are in the 313 area code, and have both 994 and 996 prefixes in our
trunk hunt.


Jim

------------------------------

End of TELECOM Digest V13 #21
*****************************