From: John Higdon <john@zygot.ati.com>
Subject: Re: Unequal Service
Date: 31 Oct 89 18:03:34 GMT
Organization: Green Hills and Cows

In article <telecom-v09i0478m05@vector.dallas.tx.us>, wb8foz@mthvax.cs.miami.
edu (David Lesher) writes:

> About this time I moved elsewhere. But I did learn something.  Always
> write letters to the business office. They HATE that. (I am not sure
> if only the supervisers can read, or they have an elite section....)
> Further, each letter I got back came signed by 	"your service rep."

Letters at least used to be the biggest guns you could fire. Years ago
in Pacific Telephone days, a client chose to venture into the brave
new world of CPE. They had a ComKey 1432 and upgraded to a PBX from an
independent vendor. This was in 1977 or so. Our "service" rep was
supposed to handle the conversion of the lines to ground start trunks,
add additional lines, etc.

 From the beginning, it was a mess. The lines were not converted
properly, or on time; the hunting was messed up; and there were
numerous other problems. Throughout all of this, our rep was
unavailable, or uncooperative. The customer and I were so fed up that
I wrote a letter to the "manager" of the business office. This
triggered a response from someone who was very apologetic and who said
things would be made right.

In the meantime, our "rep" called me and whined about the letter I
wrote. He said that he was now taking considerable heat and tried to
lay a guilt trip on me. I reminded him how we had repeatedly tried to
get him to perform and asked that he look at it from our perspective.
A day or two later, a supervisor called to tell me that our "rep" had
been properly processed and would I be so kind as to tell them if I
had any further trouble. At that point I related the conversation with
the rep in which he whined about my letter. This supervisor was aghast
and got off the phone. Later that day, the sup called back to tell me
that our rep had been let go (!) and that we would have a new one.

I wasn't really too happy about causing someone's termination, but I
was certainly impressed with the action that my humble little letter
seemed to trigger.

        John Higdon         |   P. O. Box 7648   |   +1 408 723 1395
    john@zygot.ati.com      | San Jose, CA 95150 |       M o o !

From: floyd@tanana.polarnet.com (Floyd Davidson)
Newsgroups: comp.dcom.modems
Subject: Re: Rural Internet Access
Date: 17 Jun 1999 17:18:24 GMT

<rassi777@hotmail.com> wrote:
>"kskc" <kskc@kskc.net> wrote:

>> 1.  Did you test with the laptop at the interface box? Using
>> more than one cable? This could prove or disprove the
>> interior wiring as a problem.
>
>Yes, I did.  I was outside the house with the tech. and
>connected to the cable at the interface box.

One consideration to keep firmly in mind, given the above
description, is that v.32bis modem protocols (14.4Kbps) were
specifically designed to function on a *minimally* specified
telephone line.

That means basically that if your line will not support at
least a 14.4Kbps connection using a v.34 or v.32bis modem,
then either the modem is bad or the telephone line does
not meet minimum specifications.  (In fact, a minimal line
today should be able to get 16.8K connections.)

Any connection better than that is purely your good fortune
however!  (Despite the fact that everyone expects to buy a
v.90 modem and get 52K... :-)

Convincing a recalcitrant telephone company, or even one
technician, of may be difficult.  But there are other ways to
cause them to fix your line!  Be a pest!  Report the problem
repeatedly.  If they close their trouble ticket, report it again
two days later.  But have patience and allow them to test and
work at a reasonable pace.  And do not *ever* be abusive or in
any way nasty to them.  In particular do not expect clerks who
answer the telephones to be technical experts.

Keep a log of who you talk to and what they say, and make them
aware that you are doing so.  All telephone companies do that
when they deal with each other, so it will not seem as strange
to them as it might be to you.  Keep track of the time and the
name (initials will do) of everyone you talk to.  List any
specific tests they say they did or will do.  Specifically ask
for an estimated time when any future action is supposed to be
completed.  And then wait until that time has come and gone,
and call back to ask for the current status.

Every time you are told that the line checks OK, repeat the
statement that your modem was specifically designed to function
over a dialup line which meets minimum specifications, and since
you cannot get a connection it is clear that the line is not
meeting those specifications.

Whenever any one individual says they cannot do more, ask to
speak to one higher level of authority.  And then accept the
fact that you will be transfered back down the ladder again.
Even allow yourself to be transfered back to the same person
again more than once, but eventually that is cause to ask to go
another step higher rather than continue a circle jerk.
Hopefully they will quickly tire of that and you will be handed
off to the most adept technician they have.

The ultimate threat, which you do not want to use unless this
drags out for many many weeks, is to write letters to the Public
Utilities Commission or whatever the appropriate regulatory agency
there is called.  Do not threaten to do that until it is apparent
that nothing else will work.  And then only threaten them briefly
before you actually do write such letters.

Generally telephone company people do not like hearing from the
same customer over and over.  You will soon get a nickname,
"the customer from Hell".  And that will cause them to
do *anything* to get rid of you!  So have patience, and continue
to report your line as a trouble.

  Floyd



--
Floyd L. Davidson                                floyd@ptialaska.net
Ukpeagvik (Barrow, Alaska)                       floyd@barrow.com
     North Slope images: <http://www.ptialaska.net/~floyd>

From: floyd@tanana.polarnet.com (Floyd Davidson)
Newsgroups: comp.dcom.modems
Subject: Re: Rural Internet Access
Date: 18 Jun 1999 04:42:42 GMT

Steve Uhrig  <suhrig@bright.net> wrote:
>Floyd Davidson wrote:
>
>SNIP
>
>> One consideration to keep firmly in mind, given the above
>> description, is that v.32bis modem protocols (14.4Kbps) were
>> specifically designed to function on a *minimally* specified
>> telephone line.
>
>	You are making the assumption that the modems are trying a
>V32.bis connection. This may not be the case at all. One of

I wouldn't say that I'm making that assumption so much as I am
instructing him to attempt to make a 14.4Kbps v.32bis
connection.  Clearly if he has not attempted that, then he
should.  Just as clearly the advice on working with the telco is
totally predicated upon demonstating that the line cannot
support a 14.4kbps connection.

>> Convincing a recalcitrant telephone company, or even one
>> technician, of may be difficult.  But there are other ways to
>> cause them to fix your line!  Be a pest!  Report the problem
>> repeatedly.  If they close their trouble ticket, report it again
>> two days later.  But have patience and allow them to test and
>> work at a reasonable pace.  And do not *ever* be abusive or in
>> any way nasty to them.  In particular do not expect clerks who
>> answer the telephones to be technical experts.
>
>	This advice can get him some big charges. Many of the

It won't cost him a dime if the line will not support 14.4Kbps
connections. There was no suggestion that he blindly call in
trouble reports without first verifying his facts.  And no telco
with half an ounce of management (Dilbert would say that is hard
to find though) should allow a customer to be harassed with undue
charges when a defective line is not fixed.

>customers I mentioned above were billed over $400.00 each
>for premise visits. Many phone companies are billing on
>modem complaints when no line problems are found. GTE has

That is entirely reasonable too, _if_ no line problems exist.

But when _known_ line problems do exist and are not found, the
telco is eventually going to face some very serious court cases
if they charge people for their own incompetence.  The fact is
that the common tests done by field personal do not even come
close to providing assurance that a line meets specifications.
(Indeed, a v.32bis/v.34 modem is about the _best_ way to verify
that a line does meet specs, or not.)

And the fact in this case is that he has made test calls with a
laptop connected right at the interface box, which does indicate
there is a line problem.

Incidentally, 15 years ago we were charging leased line
customers $480 to verify C2 conditioning specs if they insisted
and we determined that the line did in meet applicable
specifications.  That policy was designed to discourage certain
large customers (the USAF, for example) who tend to assign the
least experienced personnel they have to do monthly routines on
working circuits, and then log circuits out because C2
conditioning isn't within their ability to do correctly.

We still tend to bill the FAA for trouble calls that turn out to
be COAM equipment.  Since some of my "trouble calls" result in
excess of $2000 spent just for airplane seats, you can imagine
that everyone gets v-e-r-y sensitive anytime I suggest they
check their equipment first.

>instructed their repair clerks to no longer accept reports
>of "I can't connect at XXX speed". If this type of report
>gets by the clerk and to a tech. The tech is to refer the
>report to the supervisor and do nothing else with it. They
>will accept a report of "can't connect at all" but if the
>tech comes out, connects his computer to the line and
>connects at 14.4 You are billed a $43.00 premise visit
>charge. If you report every other day for a month that comes
>out to $645.00 in charges.

Note that, as described, your criteria is exactly what mine is:
the line either will connect at 14.4Kbps or there is a problem.
If the telco, or the customer, can connect at 14.4kbps that is
the end of the story.  There was no suggestion that the telco be
harassed about lines that do not have a problem.  I specifically
pointed out that connections at rates higher than that are
freebies, and not an entitlement even though many seem to expect
it.

It is inappropriate to harass customers with threats of huge
bills when there are problems, too.  If the line won't support
14.4kbps connections then the telco should fix the line, period.

>SNIP
>
>> Generally telephone company people do not like hearing from the
>> same customer over and over.  You will soon get a nickname,
>> "the customer from Hell".  And that will cause them to
>> do *anything* to get rid of you!  So have patience, and continue
>> to report your line as a trouble.
>
>	They will bill you so much that you can't pay the bill and
>they will disconnect your service. Problem solved for them.

If the line will not make 14.4Kbps connections, just imagine
what the PUC (or any good attorney) will do with a telco that
sends you a bill or disconnects your service!  And, imagine the
newspaper articles if there is an elderly person in the house!
I don't think any telco attorney is going to let that kind of
customer treatment stand, and I don't think any telco manager
should even allow it to get as far as the attorney.

  Floyd



--
Floyd L. Davidson                                floyd@ptialaska.net
Ukpeagvik (Barrow, Alaska)                       floyd@barrow.com
     North Slope images: <http://www.ptialaska.net/~floyd>

From: floyd@tanana.polarnet.com (Floyd Davidson)
Newsgroups: comp.dcom.modems
Subject: Re: Rural Internet Access
Date: 18 Jun 1999 22:34:37 GMT

Fred Goodwin, CMA <fgoodwin@eden.com> wrote:
>Floyd Davidson <floyd@ptialaska.net> wrote in
><7kcio2$2nji@enews4.newsguy.com>:
>
>>Steve Uhrig  <suhrig@bright.net> wrote:
>>
>>>     This advice can get him some big charges.
>>
>>It won't cost him a dime if the line will not support 14.4Kbps
>>connections.
>
>If the line supports voice calls and is within noise and loss specs, then
>the telco will bill for the premises visit.
>
>You're making an assumption that a voice line should support 14.4
>connections.  A voice line is guaranteed to support voice, nothing more.

That is not an assumption, it is a fact.

The v.32bis modem protocol was designed specifically to function
at 14.4Kbps over a *minimally* specified line.  You can
emphatically assume that if a 14.4Kbps or better connection
cannot be made that either 1) the modem is defective, or 2) the
line does not meet minimum specifications.

As both Steve and I have mentioned, a 14.4Kbps v.32bis modem is
a great diagnostic tool for telcos.  It may not indicate what
problems exist, but it will definitely provide a fairly positive
yes/no answer to whether a line meets minimum specs or not.

>There is nothing in any tariff I'm aware of that requires POTS lines to
>support 14.4kbps analog modems or any other measure of modem performance.
> If you know otherwise, please share it with us.
>
>If such requirements were in the tariff, we wouldn't even be having this
>discussion; the telcos are clearly obligated to provide the level of
>service stated in the tariff, nothing more.

That begs the point though.  The fact is that if a correctly
functioning v.32bis modem cannot make a 14.4Kbps connection then
_some_ parameter on the line *is* out of specs.  Our problem as
telephone company maintenance people is to correct the problem,
not to force the customer to prove exactly what our problems are
before we move.

That may be less true in your regulated part of the industry
than it is in my unregulated portion, but never-the-less it is
true and will become even more important as the move towards
real competition for local loops progresses.

>>But when _known_ line problems do exist and are not found, the
>>telco is eventually going to face some very serious court cases
>>if they charge people for their own incompetence.
>
>Again, you seem to equate poor 14.4 modem performance with "known" line
>problems.  As I said, if you can call out on the line and its within
>noise & loss specs, then the telco is providing exactly the service
>you're being billed for.

And if that is true then you can be assured that 14.4 modem
performance *will* be able to work.

>>The fact is that the common tests done by field personal do not even
>>come close to providing assurance that a line meets specifications.
>
>The loss & noise tests Icky Pic described are consistent with tariff
>requirements on POTS lines, and there wasn't a single data test among
>them.

So?  What tests are normally performed on a line when it is
reported having trouble?  How many times are full frequency runs
done?  How many times are impulse noise, phase hits,
etc. measured?  And is a balance measurement of some kind done
every time?

Generally, no they are not all done.  And neither are checks made
to assure that digital facilities are not slipping occasionally.

90% of all problems can be solved without going to that extreme,
so it is rare that time is wasted making complete tests.  But when
there _is_ a problem, the test should be done.  Doing half of them
and declaring there can't be a problem is not an appropriate
response.

>>(Indeed, a v.32bis/v.34 modem is about the _best_ way to verify
>>that a line does meet specs, or not.)
>
>Passing 14.4 is a good indicator that your line is within POTS specs, but
>the converse is not true.  Failing to pass 14.4 does not necessarily
>imply the line is bad.

I disagree.

>>And the fact in this case is that he has made test calls with a
>>laptop connected right at the interface box, which does indicate
>>there is a line problem.
>
>If you can take a handset and call out from the demarc (isolating the
>IW), the network line is good.  The original poster never told us whether
>or not he could make voice calls from the demarc (or from within his
>girlfriend's house, for that matter).

He has never suggested that the line will not make voice calls.  He
has attempted data calls from the demarc, and they failed.  He did
not supply enough specifics to be absolutely sure what the significance
of that is however.  As Steve has suggested it is very possible that
his modem was not configured to allow 14.4Kbps connections, but we
don't really know that.

>>It is inappropriate to harass customers with threats of huge
>>bills when there are problems, too.  If the line won't support
>>14.4kbps connections then the telco should fix the line, period.
>
>If the line doesn't meet tariffed specs, the telco should fix the
>problem.  If the line meets those specs but doesn't pass 14.4, that's too
>bad for the customer but the telco met its obligation -- it is *not*
>obligated to pass 14.4 over a POTS line.

But you *can* be assured that if the modem is good and the line meets
specs, the you *will* get at least 14.4Kbps connections.

>If you know of such a tariff requirement, please point me to it.
>
>>If the line will not make 14.4Kbps connections, just imagine
>>what the PUC (or any good attorney) will do with a telco that
>>sends you a bill or disconnects your service!
>
>Again, I'm not aware of any PUC that requires that POTS lines pass 14.4.
>If you know otherwise, please provide a reference.

There is also no tariff for PAR measurement either, but we all know
very well that if it reads 40% there is something wrong.  We don't know
exactly which parameters are out of specs, but we know something is.


Need I point out once again that you sound just like the perfect
telephone company manager Fred!  :-)

  Floyd

--
Floyd L. Davidson                                floyd@ptialaska.net
Ukpeagvik (Barrow, Alaska)                       floyd@barrow.com
     North Slope images: <http://www.ptialaska.net/~floyd>

From: floyd@tanana.polarnet.com (Floyd Davidson)
Newsgroups: comp.dcom.modems
Subject: Re: Rural Internet Access
Date: 19 Jun 1999 00:56:17 GMT

Fred Goodwin, CMA <fgoodwin@eden.com> wrote:
>John Navas <spamfilter@navasgrp.dublin.ca.us> wrote in
><377c955a.44965125@news.swbell.net>:
>>
>>I don't think I've ever seen a case where 9600-14400 couldn't be
>>reliably attained on a clean POTS line.
>
>At the risk of repeating myself, so what?  If your POTS line does not pass
>14.4 but you can call out on it, the telco is providing all the service it
>is obligated to do.

Fred you will get a reputation as the prototype for pointy
haired managers if you keep making statements like that.

If the line will not support 14.4Kbps connections then it
clearly does NOT MEET SPECS.  It is not up to the customer to
prove that, or to even determine which specification is not up
to snuff.

It might happen that due to circumstances there is no way to
provide a given customer with a line which does meet specs!
They have to option of a below standard line or no service at
all.  Certainly in many rural areas that is not as uncommon as
we wish.  But lying about it, badgering the customer to prove
it, or worse yet threatening customers with huge charges is
disgusting.

  Floyd

--
Floyd L. Davidson                                floyd@ptialaska.net
Ukpeagvik (Barrow, Alaska)                       floyd@barrow.com
     North Slope images: <http://www.ptialaska.net/~floyd>

From: floyd@tanana.polarnet.com (Floyd Davidson)
Newsgroups: comp.dcom.modems
Subject: Re: Rural Internet Access
Date: 19 Jun 1999 04:58:08 GMT

Rick Collins <ab309@issc.debbs.ndhq.dnd.ca> wrote:
>floyd@tanana.polarnet.com (Floyd Davidson) wrote:
>
>>That begs the point though.  The fact is that if a correctly
>>functioning v.32bis modem cannot make a 14.4Kbps connection then
>>_some_ parameter on the line *is* out of specs.
>
>Oh well - in for a penny, in for a pound:
>
>Last time I looked, no 14.4 modems were certifed as "test instruments"
>guaranteed to implement V.32bis in a manner necessary to function
>correctly on a "minimal" line.  So - if you can't connect with a modem
>at 14.4 using V.32bis, _just maybe_ the modem (or the remote modem) is
>a poor implementation of the V.32bis spec, and therefore proves
>nothing about the line characteristics.
>
>In other words, your "correctly functioning" are two words which hide
>a whole cornucopia of possible pitfalls.
>
>:-)

You mean to say that I chose the words with care.

If we have an old modem of generic vintage, I wouldn't trust it
for 10.2 milliseconds.

If we have an USR Courier modem...

But the same thing is true for test equipment of any kind.  I've
had people suggest to me that they measured levels using a Radio
Shack VOM.  I figure that is fine, we think that voltage does
exist but have no real idea how much!

If we have an Hewlett-Packard TIMS...

The point stands that if the Courier won't connect at 14.4Kbps
or better, then the TIMS _is_ going to find something wrong with
the line.  No amount of playing "I am the TELEPHONE COMPANY, I
don't have to care" is going to change that.

Our mission is to fix the problem, not to find ways to lord it
over customers.

   Floyd

--
Floyd L. Davidson                                floyd@ptialaska.net
Ukpeagvik (Barrow, Alaska)                       floyd@barrow.com
     North Slope images: <http://www.ptialaska.net/~floyd>

From: floyd@tanana.polarnet.com (Floyd Davidson)
Newsgroups: comp.dcom.modems
Subject: Re: Rural Internet Access
Date: 19 Jun 1999 16:47:47 GMT

In article <3776a26e.113859530@news.swbell.net>,
John Navas <spamfilter@navasgrp.dublin.ca.us> wrote:
>[POSTED TO comp.dcom.modems]
>floyd@tanana.polarnet.com (Floyd Davidson) wrote:
>
>>Our mission is to fix the problem, not to find ways to lord it
>>over customers.
>
>There is obviously no future for you in the telecom business.  <g>

Well, certainly not in the local loop part of the business!  But
since I'm in the LD end of it, I spend a lot of time finding
ways to lord it over Local Exchange Carriers.  ;-)

  Floyd


--
Floyd L. Davidson                                floyd@ptialaska.net
Ukpeagvik (Barrow, Alaska)                       floyd@barrow.com
     North Slope images: <http://www.ptialaska.net/~floyd>

From: floyd@tanana.polarnet.com (Floyd Davidson)
Newsgroups: comp.dcom.modems
Subject: Re: Rural Internet Access
Date: 19 Jun 1999 16:54:16 GMT

Bob Starnes  <rastarnesNOSPAM@hotmail.com> wrote:
>John Navas wrote:
>>
>> [POSTED TO comp.dcom.modems]
>> floyd@tanana.polarnet.com (Floyd Davidson) wrote:
>>
>> >Our mission is to fix the problem, not to find ways to lord it
>> >over customers.
>>
>> There is obviously no future for you in the telecom business.  <g>
>>
>> --
>> Best regards,
>> John  mailto:jnavas@aimnet.com  http://www.aimnet.com/~jnavas/
>>       28800-56K Modem FAQ:  http://www.aimnet.com/~jnavas/modem/faq.html
>
>All this talk makes me wonder how many telco executives live with low
>speed (28800 or less) connections?

Of the 10% that actually use modems, how many of them can figure out
how fast it is connecting?   ;-(

  Floyd



--
Floyd L. Davidson                                floyd@ptialaska.net
Ukpeagvik (Barrow, Alaska)                       floyd@barrow.com
     North Slope images: <http://www.ptialaska.net/~floyd>

From: floyd@tanana.polarnet.com (Floyd Davidson)
Newsgroups: comp.dcom.modems
Subject: Re: Rural Internet Access
Date: 21 Jun 1999 17:40:09 GMT

Fred Goodwin, CMA <fgoodwin@eden.com> wrote:
>Floyd Davidson <floyd@ptialaska.net> wrote in
><7kkg70$fnc@enews5.newsguy.com>:
>
>>We disagree on the significance of that, which is fine.  But if
>>you go back over the past few years of articles on dejanews you
>>are going to find yourself almost totally alone in that stance too.
>>This isn't exactly the first time the subject has been discussed,
>>though you have definitely presented the most vocal argument yet
>>claiming that it is not true.
>
>Steve pointed out several cases where there *is* a regulatory obligation to
>maintain a certain minimum bps performance level on POTS lines.  In those
>cases, I agree the telco must maintain the line to those specs.
>
>Otherwise, it simply has no obligation to do so.


There is no regulatory statement that an HP TIMS is suitable to
determine whether a line meets specs or not either.  Most of us
do use HP test equipment for that purpose though.  We are
satisfied that the specification the TIMS is built to meet means
that it's measurements are valid in the context of regulatory
statements about telco line specifications.

There is a clear parallel there.  Would you deny that the SNR
measurements made with an HP-4935A are valid to determine if a
line meets specifications simply because neither it nor the
exact means it uses to measure SNR are mentioned in any tariff
or regulation?

We compare the specification in the tariff/regulations for test
equipment and compare those to the specifications for the
HP-4935A.  They match.  So does the v.32bis specification.

  floyd

--
Floyd L. Davidson                                floyd@ptialaska.net
Ukpeagvik (Barrow, Alaska)                       floyd@barrow.com
     North Slope images: <http://www.ptialaska.net/~floyd>

From: floyd@tanana.polarnet.com (Floyd Davidson)
Newsgroups: comp.dcom.modems
Subject: Re: Rural Internet Access
Date: 21 Jun 1999 17:57:27 GMT

David Burgess  <burgess@mitre.org> wrote:
>In general, the phone company technicians try to do a good job and actually
>care about the customers.  In general, they are your friend and, sometimes, are
>your only advocate in the phone company.  If they tell you that your phone
>service is working, there is generally no reason to believe it isn't.  That
>isn't to say that the "suits" will treat you the same way, just that the techs
>are just doing their job.  Just like you, they loike to do a good job,
>especially for someone that appreciates their effort.

Techs want to do their job, which is to make trouble tickets
disappear.  They could care less about whether a line meets
every spec in the book or not, and most of them do not
understand even half of what the spec says.

Suits want to do their job, which is to produce a profit by
configuring business processes by which the techs do their job
as efficiently as possible.  Efficient means inexpensive in this
case.

Engineers want to do their job, which includes knowing
everything there is to imagine about a customer's line.  If you
want someone who cares about the customer's line (though perhaps
not a wit about the customer), talk to an engineer.  Engineers
are paid to be concerned about that, and no other telco employee
is.

Now, consider that there are many more technicians than managers
in most telephone companies.  And there are more managers than
engineers.

That gives a bit of perspective on how much a telco is likely
to care about a customer's line meeting all specifications.

(And note too that using US Worst as an example is a fundamental
flaw in itself!)

  Floyd




--
Floyd L. Davidson                                floyd@ptialaska.net
Ukpeagvik (Barrow, Alaska)                       floyd@barrow.com
     North Slope images: <http://www.ptialaska.net/~floyd>

From: floyd@tanana.polarnet.com (Floyd Davidson)
Newsgroups: comp.dcom.modems
Subject: Re: Rural Internet Access
Date: 22 Jun 1999 04:45:01 GMT

Steve Uhrig  <suhrig@bright.net> wrote:
>Floyd Davidson wrote:
>
>> There is no regulatory statement that an HP TIMS is suitable to
>> determine whether a line meets specs or not either.  Most of us
>> do use HP test equipment for that purpose though.  We are
>> satisfied that the specification the TIMS is built to meet means
>> that it's measurements are valid in the context of regulatory
>> statements about telco line specifications.
>
>	I have never seen an HP test set that does all the POTS
>line tests. The tests that it does do are perfectly valid
>for that portion of the line test, but you will need other
>equipment to finish the tests.

So?  There was no claim that it did all tests. I've never seen
any single box which does. The point is that the tariffs and specs
for lines do not mention an HP TIMS for *any* test.  But that is
exactly what they are designed for, and used for.

>> There is a clear parallel there.  Would you deny that the SNR
>> measurements made with an HP-4935A are valid to determine if a
>> line meets specifications simply because neither it nor the
>> exact means it uses to measure SNR are mentioned in any tariff
>> or regulation?
>
>	To take a SNR reading on a pots line with any TIMS I am
>familiar with would require two men to do it. Unless you

I can't see where that is germane, but in fact it can be easily
done by one person (even a woman!) if they have a butt set.

>just calculate it from your tone and c-message readings.

That would work, but it is a lot easier to push the little
button that says SNR.  (Find an HP catalog and look up an
HP-4953A model TIMS.  We have always tended to call it a
"mini-TIMS" because it is small.  It doesn't do envelope
delay, but it does do SNR and PAR tests.)

>This would be a waste of time Floyd, because there is no SNR
>requirement specified for a pots line. You could calculate

In fact there is:  24 dB minimum.

>one, as I suggested, using the standard test results.
>Maximum c-message noise is 20 dbrn or -70dbm and the minimum
>line level at -8.0 would give a SNR of 62 worst case and 70
>best case.

But that isn't the way a line is specified for SNR.

>	While using the customers modem seems to be a good idea, I
>have some reservations. I for one have no idea what extra
>settings to put in my connectoid to get my plain jane modem
>to attempt only a v32.bis connection and I would imagine
>that the typical user of a V.34 or V.90 modem doesn't
>either. If the customer had an old 14.4 modem that they
>could try on the line I would feel much better about the
>results. A proper functioning 14.4 modem in my experience
>could probably connect over a line run over a barbed wire
>fence. I think this is what you are trying to get across in
>your statements. I would have to agree that if a properly
>functioning 14.4 modem can't connect to anything, either the
>modem is bad or the line is bad. The key here is anything.

The point is that when a customer has such a problem and the
telephone company starts babbling about not guaranteeing that
modems work, whimpering that they might charge the customer one
arm and two legs to look at it, and then refuses to accept that
they are indeed selling not just a voice conversation but a
"voice grade circuit" which damned well oughta work with any
functional FAX or dialup data modem (both of which are data and
do the same thing), then the telephone company has their head
where the sun won't shine.

That was acceptable before Judge Green began the process of
putting a stop to it.  Any telco that continues to think in
those terms is about to cease to exist as it is now known.  GTE
for years was one of the worst, and has changed their reputation
significantly in the past decade, as an example.  It appears
they are about to move towards the next step too.

Other companies are not doing that.  I spent 2 and a half
decades working with one that couldn't quite get it right, and
bingo one day the entire top level of management was out looking
for a job.  They renamed it by adding AT&T onto the front!  The
old company ceased to exist because they wouldn't stick their
heads up and look around.  (And that was just about the last
long distance company to be that way.)

The local loop is clearly the next phase.  People can change the
way they think, or they will be run over by a steam roller...

>Some of these POP problems can cause a given modem to not
>connect to the POP in question at all. Dialing another POP
>or a fax machine number will in my experience always result
>in a connection. Probably not as good as the customer would
>like, but a V.34 connection in most if not all cases. I
>would have to say that for the line to prevent a v.32bis
>connection on it's own there will also be a very big problem
>with voice calls as well. Since there would already be a
>voice problem for the phone company to work on, there would
>be no need to do the modem test.

I've been assigned hot jobs which amounted to a line that would
test *perfect* in every way we could imagine testing it, but
would not work at any data rate above 2400 bps.  This happened
to be on a modem bank for access to our dial up service that was
being used by an entire school district, so this was not
something that could be ignored, given the politics.  It turned
out to be an old revision of NEC line cards in a channel bank.
I never did find a single parameter that would not pass muster,
and the only good tool we had was a laptop with a 14.4Kbps modem
in it.

However, the point *still* is not the number of possible false
directions you, I, the customer, or management can be sent it.
The point is that a 14.4Kbps modem is a very good indicator of
whether the line meets specs, and any customer who appears to
have made a reasonably good faith effort at isolating his
problem using a logical approach with a 14.4Kbps modem as a
diagnostic tool should not be ignored, threaten, or browbeat by
the telephone company.

  Floyd

--
Floyd L. Davidson                                floyd@ptialaska.net
Ukpeagvik (Barrow, Alaska)                       floyd@barrow.com
     North Slope images: <http://www.ptialaska.net/~floyd>

From: floyd@tanana.polarnet.com (Floyd Davidson)
Newsgroups: comp.dcom.modems
Subject: Re: Rural Internet Access
Date: 22 Jun 1999 04:54:41 GMT

Fred Goodwin, CMA <fgoodwin@eden.com> wrote:

>Let me ask you: if the telco's brand "X" 14.4 modem connects at 14.4, but
>the customer's brand "Y" modem connects at 9.6, does the line fail?  If so,
>then you are obligating the telco to build and maintain a POTS line that
>will work with every possible combination of user/ISP 14.4 modems, which
>seems just a bit impractical to me.

Fred that is waffling.

I said that if a v.32bis or v.34 modem can get a 14.4Kbps connection then
the line meets specs.  I didn't say it has to connect to your grandmother's
modem, or to my dog's ISP.

A telco that sends tech's out with a laptop and a brand X 14.4 modem is
in a very good position to save a lot of time and trouble for both the
customer and themselves.  When the customer says the line is bad, the
telco tech connects at the demarc and dials up a 14.4Kbps connection
(I rather liked Steve Uhrig's comment about doing it so the could close
the trouble ticket and get a dispatch for the next job).  When the
tech *can* do that, it *is* time to close the ticket.

Until it can be done there is still a problem to fix.

  Floyd


--
Floyd L. Davidson                                floyd@ptialaska.net
Ukpeagvik (Barrow, Alaska)                       floyd@barrow.com
     North Slope images: <http://www.ptialaska.net/~floyd>

From: floyd@tanana.polarnet.com (Floyd Davidson)
Newsgroups: comp.dcom.modems
Subject: Re: Rural Internet Access
Date: 22 Jun 1999 05:12:13 GMT

Rick Collins <ab309@issc.debbs.ndhq.dnd.ca> wrote:
> floyd@tanana.polarnet.com (Floyd Davidson) wrote:
>
>>Fred you can deny that all you like.  The v.32bis spec was *designed*
>>to permit a modem to function on a line that meets *minimum* specs.
>>
>>That was not a spec developed in a vacuum.
>>
>Are you saying that a V.32 modem connected to a  telco line that meets
>(not exceeds) all specifications will _always_, _without exception_
>connect at 14,400?
>
>Or are you saying that was the "intention" of the spec?

Don't be a pedantic asshole Rick.  That is a silly distintion.

Test equipment fails. Modems fail.  So to telephone lines.  But
a v.32bis or v.34 modem that will not make at least a 14.4Kbps
connection over a minimally specified line is indeed a faulty
modem.

Conversely, a line over which a correctly functioning v.32bis
modem cannot make at least a 14.4Kbps connection does not meet
specifications.

Now, it does happen that improvements in the implementation of
the v.34 specification have made it so that almost any recently
manufactured v.34 modem will in fact get *better* than 14.4Kbps
over a minimally specified line.  Hence it becomes even more
emphatic that one can pretty much give blind faith to the idea
that a v.34 modem which cannot connect at 14.4Kbps or higher on
a given line means the line is faulty.

In general, a v.34 modem's connect speed is an excellent
indication of the quality of a telephone line!

>I return to my previous point:  who determines that the _modem_ meets
>the minimum spec for the V.32 "recommended standard"?

The standard is publicly available, just as are the specifications
for telephone lines.  We have no more, or less, reason to trust a
given modem manufacturer than we have reason to trust a given
test equipment manufacturer.  For that reason the only modem that
has been mentioned so far in this thread was a USR Courier, commonly
accepted to be the standard for comparison.  Likewise the only test
equipment specifically mention has been Hewlett-Packard, which is
also recognized as the standard for comparison.

There certainly are equipments with lower prices and less of a
reputation, and there is certain amount of risk that it will
give faulty results more often than higher quality brands.  That
is true either with test equipment or with modems, or with line
equipment too!

  Floyd


--
Floyd L. Davidson                                floyd@ptialaska.net
Ukpeagvik (Barrow, Alaska)                       floyd@barrow.com
     North Slope images: <http://www.ptialaska.net/~floyd>

From: floyd@tanana.polarnet.com (Floyd Davidson)
Newsgroups: comp.dcom.modems
Subject: Re: Rural Internet Access
Date: 22 Jun 1999 20:31:29 GMT

Fred Goodwin, CMA <fgoodwin@eden.com> wrote:
>Floyd Davidson <floyd@ptialaska.net> wrote in
><7kn4ce$2uck@enews4.newsguy.com>:
>
>>So?  There was no claim that it did all tests. I've never seen
>>any single box which does. The point is that the tariffs and specs
>>for lines do not mention an HP TIMS for *any* test.  But that is
>>exactly what they are designed for, and used for.
>
>Agreed - they are designed and used to test for transmission impairments;
>modems are not.

Not quite true.  A test set is designed to measure transmission
parameters.  A modem is designed to function within a range of
impairments.  That means that both of them are useful diagnostic
tools to test for transmission impairments.  The modem can quickly
determine that there are or are not any significant impairments.
A test set can measure parameters to determine what impairment
exists.

If a manager's job description includes designing business
processes that result in efficient use of technician time, then
the use of modems as test equipment should be pretty high up the
list of priorities.

>>The point is that when a customer has such a problem and the
>>telephone company starts babbling about not guaranteeing that
>>modems work, whimpering that they might charge the customer one
>>arm and two legs to look at it, and then refuses to accept that
>>they are indeed selling not just a voice conversation but a
>>"voice grade circuit" which damned well oughta work with any
>>functional FAX or dialup data modem (both of which are data and
>>do the same thing), then the telephone company has their head
>>where the sun won't shine.
>
>"What oughta work" < > what is *required* (recognizing the exceptions that

Recognizing that a FAX modem or a v.32bis modem are designed to work
within the range of *minimum* specifications for a telephone line,
"what oughta work" and what is *required* are identical.

You can refuse to accept that if you like, but smart managers have
learned their lessons about that long ago.

>Steve pointed out); if you ask for more than is required, the telco can
>rightfully bill you extra for it.

>>That was acceptable before Judge Green began the process of
>>putting a stop to it.  Any telco that continues to think in
>>those terms is about to cease to exist as it is now known.
>
>Judge Green had nothing to do with setting POTS line requirements.

Surely you jest.  In the 1960's it was virtually illegal to put
a "foreign" device on a Bell telephone line.  Judge Green changed
that.  Then it was virtually illegal to put one that didn't meet
Bell specifications (which were designed specifically to make it
impossible to do).  Judge Green changed that too.  Which is to
say he had just about everything to do with line specs as they
exist today.

>>However, the point *still* is not the number of possible false
>>directions you, I, the customer, or management can be sent it.
>>The point is that a 14.4Kbps modem is a very good indicator of
>>whether the line meets specs, and any customer who appears to
>>have made a reasonably good faith effort at isolating his
>>problem using a logical approach with a 14.4Kbps modem as a
>>diagnostic tool should not be ignored, threaten, or browbeat by
>>the telephone company.
>
>If a line passes 14.4, then the line is good; it is a logical fallacy to
>say that if a line fails to pass 14.4, then it is bad.  All you can
>conclude from the first, is that if the line is bad, it won't pass 14.4.
>
>I.e., (if a then b) implies (if not b then not a)
>
>(If a then b) does NOT imply (if not a then not b)
>
>Its basic logic.

Your premises are flawed, therefore you conclusion is invalid.

   a + b = c

and

   b + a = c  too.

Where,

  "FAX or v.32bis modem functions correctly"    == a
  "telephone line meets minimum specifications" == b

therefore

 "Modem/FAX can connect at specified rate"      == c

If c is not true, the one or both of a and b must also not be true.
If c is not true, and either a or b is known to be true, then the
other of b or a is proven to be not true.

Hence, if you have a known good v.32bis modem and it cannot make
14.4Kbps connections, the line is bad.

  Floyd

--
Floyd L. Davidson                                floyd@ptialaska.net
Ukpeagvik (Barrow, Alaska)                       floyd@barrow.com
     North Slope images: <http://www.ptialaska.net/~floyd>

From: floyd@tanana.polarnet.com (Floyd Davidson)
Newsgroups: comp.dcom.modems
Subject: Re: Rural Internet Access
Date: 22 Jun 1999 22:31:58 GMT

Fred Goodwin, CMA <fgoodwin@eden.com> wrote:
>John Navas <spamfilter@navasgrp.dublin.ca.us> wrote in
><3788c501.35725346@news.swbell.net>:
>
>>[POSTED TO comp.dcom.modems]
>>fgoodwin@eden.com (Fred Goodwin, CMA) wrote:
>>>
>>>So let me ask you the same thing I asked Floyd: if the telco tester's
>>>14.4 works, but the customer's does not, does the line pass or fail?
>>
>>If the telco tech demonstrates that the line can pass modem data at
>>14.4, then the line presumably passes (the data test).
>
>OK -- but then the telco and the customer are really no better off than
>before, the customer still isn't getting the performance he thinks he
>deserves, right?

At that point they should both be aware that the telephone company is
providing the service they are being paid for.  If the customer's
modem can't achieve the same results, clearly it is faulty.  If it
can while connected to the demarc, but can't when connected via the
normally used house wiring, then clearly the house wiring is suspect.

The point is that they do know a great deal more than before tests
were done, whether they like the results or not is immaterial.

>So why can't the telco give the customer the same answer ("line tests OK" -
>- assuming it actually does so) by using its standard transmission test
>sets (which after all are designed to test for line impairments), rather
>than requiring it to use a modem which isn't designed to detect such
>things?

Simple enough.  If a tech takes about 5 minutes to thoroughly test
a line using a laptop with v.32bis modem attached at the demarc, that
is relatively quick.  If that same test takes 30 minutes that is a
long time.

On the other hand, it takes far more than 30 minutes to even get started
on an exhaustive battery of tests to determine if a line actually does
meet all of the various specifications.  As mentioned before there is
no single test set which will measure every specified parameter.  Plus
tests like impulse noise require at least one 15 minute test period.
Unless one has some very expensive (and difficult to manage) equipment
it is also time consuming to do amplitude attenuation testing.

Perhaps 45 minutes would be an absolute minimum for a very well equipped
tech with a lot of experience.  Given that very few field techs that
work on customer line troubles have that kind of experience, it might
be expected to take twice that long at best, and four times that on
reasonable occasions.

All of that is the reason there are a number of handy dandy "line tester"
sets around which do a quicky set of tests to rapidly give an idea how
the line looks.  They are designed to detect impairments that will
interfere with voice calls though, and do not give an overall picture
of impairments (for example, phase hits can't be heard and are not
measured by such sets).  A v.32bis modem is a very good indicator of
overall circuit quality, but gives absolutely no indication of which
parameters are impaired.

>(Again, I'm referring only to those states that have no requirement that
>POTS lines pass 14.4 -- if a requirement exists, then the telco is clearly
>obligated to meet the requirement.)

That is not true Fred.  In every jurisdiction there are a set of
minimum parameters.  They are virtually the same set of
parameters, with a few minor variations.  That basic set of
parameters can loosely be called a "voice grade circuit".  If that
set of parameters is at least meeting minimum specifications, then
a v.32bis modem will be able to connect at 14.4Kbps over that line.

IF IT CANNOT, THEN THE LINE DOES NOT MEET MINIMUM SPECIFICATIONS.

Clearly the telco is required to meet those specifications under
normal circumstances.  If the line is not meeting specifications
the telco is obligated to correct the problem whether they have
enough sense to use a modem to test with or not.

Again, you can claim otherwise from now until December, but that
isn't going to change the fact that it is so.

  Floyd

--
Floyd L. Davidson                                floyd@ptialaska.net
Ukpeagvik (Barrow, Alaska)                       floyd@barrow.com
     North Slope images: <http://www.ptialaska.net/~floyd>

From: floyd@tanana.polarnet.com (Floyd Davidson)
Newsgroups: comp.dcom.modems
Subject: Re: Rural Internet Access
Date: 23 Jun 1999 00:03:28 GMT

Fred Goodwin, CMA <fgoodwin@eden.com> wrote:
>Floyd Davidson <floyd@ptialaska.net> wrote:
>>Fred Goodwin, CMA <fgoodwin@eden.com> wrote:
>>>Floyd Davidson <floyd@ptialaska.net> wrote:
>>>
>>>>That was acceptable before Judge Green began the process of
>>>>putting a stop to it.
>>>
>>>Judge Green had nothing to do with setting POTS line requirements.
>>
>>Surely you jest.  In the 1960's it was virtually illegal to put
>>a "foreign" device on a Bell telephone line.  Judge Green changed
>>that.
>
>Judge Green did not decide the foreign attachments case; "Hush-a-Phone"
>was decided in '57 and '58.  It was the Carterfone decision in '68 that
>opened up the network to non-Bell CPE, and that was an FCC decision, not
>a judicial one.

I did not say that Judge Green decided any of those cases.  I
also did not say that those cases are what changed the way it
all worked either, and if you are saying that they did then I
disagree.  Clearly Judge Green didn't agree either... because
that is why he took over and _did_ change it all.

>>Then it was virtually illegal to put one that didn't meet Bell
>>specifications (which were designed specifically to make it impossible
>>to do).  Judge Green changed that too.
>
>The FCC instituted its Part 68 registration program in 1975.  See Docket
>No. 19528, First Report and Order, 56 FCC 2d 593 (1975).  That eliminated
>the "connecting arrangement" requirement that AT&T had been using since
>Carterfone.  Again, Judge Green had nothing to do with it.
>
>>Which is to say he had just about everything to do with line specs as
>>they exist today.
>
>POTS requirements existed long before Judge Green took over the
>Divestiture case.  In fact, Judge Green wasn't even the original judge in
>the DOJ's antitrust case that resulted in the '82 Consent Decree and '84
>Divestiture.
>
>So no, I'm not jesting -- but you might want to check your facts again.

Judge Green is the one who came to the ultimate conclusion that none
of the previous attempts at correcting the problem had worked and that
they could not be patched in some way to make them work.  He started
the process that did work... which is what I said to start with.

  Floyd



--
Floyd L. Davidson                                floyd@ptialaska.net
Ukpeagvik (Barrow, Alaska)                       floyd@barrow.com
     North Slope images: <http://www.ptialaska.net/~floyd>

From: floyd@tanana.polarnet.com (Floyd Davidson)
Newsgroups: comp.dcom.modems
Subject: Re: Rural Internet Access
Date: 24 Jun 1999 01:25:14 GMT

Steve Uhrig  <suhrig@bright.net> wrote:
>
>> BTW, how do you loop back a two wire local loop?  That would be
>> a very good trick!
>
>	Very simple. You make the customer buy two lines and then
>tie them together at one end. HA HA  Seriously that was the
>point. The HP is designed to do SNR on a looped circuit.
>Since you can't loop the pots line one PERSON can't do the
>test.

One person who knows how to use the equipment can.

The HP is not limited to doing SNR on a looped circuit.  There
is no requirement that the HP be the source of the tone, and
dialing up a test tone works very well.  That is why I mentioned
using a butt set... to use for dialing.

It only takes one person.

>> You might try refiguring that calculation above for a line
>> that is provisioned on a SLC-96 or similar concentrator.
>> All of a sudden the best case is perhaps 36 dB.
>
>	You are going to have to explain this one in detail. The
>line specs do not change with the type of equipment on the
>line. The noise requirement and level requirement are still
>the same.

You quoted the noise specification for a cable pair, not for a
POTS line.  I'm merely pointing out that if you use any type of
PCM digital transport (for example if the line traverses a
digital switch), then the best SNR that can be obtained is about
36 dB due to quantization noise.  Clearly cable pairs have less
noise than DS0 channels, but DS0's are also used to provision
POTS lines.  Quoting the noise specifications for a cable pair
is not an accurate representation of the parameters that a POTS
customer can expect.

>> >       They key here is functional. Included in functionality is
>> >the absence of inter operability problems with the modem or
>> >machine that they are trying to connect with. Not that I
>> >work on millions of these reports per year, but there has
>> >never been a single case in this area where we could not
>> >connect on the customers line with the modems in our laptop.
>
>> Well certainly not after you fix the problems!  But are you
>> telling me that you could connect on open or shorted pairs?
>> You have never had a pair go open, or become shorted in your
>> area????  I think you just didn't try it until after  you
>> fixed the problem!
>
>	An open or shorted pair is going to knock out voice
>service. Your contention all along has been that the modem
>will find problems that the standard line tests do not find.
>If the customers line is open or shorted they are not going
>to be reporting modem problems. They are going to be
>reporting NO DIAL TONE, CAN'T BE CALLED. You are shooting
>yourself in the foot here Floyd.

I was responding directly to what you said in the paragraph that
I quoted above, not to something stated in a different article
on a different day.  I think both the temporal and spatial
proximity made that clear to everyone who read it, and your
response is just avoiding the obvious by trying to add
confusion.

>> Change your tactics a bit, try the modem first.  Then you
>> will probably find that you never have a problem that is not
>> indicated by the way the modem performs.
>
>	What repairman is going to be stupid enough to try to use
>his modem on a line he already knows is open or shorted? I
>think you are drifting out in left field here trying to find
>some way to prove your theory.

All of your trouble reports are that cut and dried???  Most of
us have problems reported as intermittent, mild, annoying, noisy,
and any number of other adjectives.  Sometimes it takes awhile
to even know there really is a problem!  (And *that* is where
a modem is very handy!)

You are the one who drifted off to left field a long time ago.
The fact is that a lot of technicians have added a modem to
their bag of tools.  Nobody is saying that it is the *only*
tool, or that it is infallible.  None of the other tools fit
that description either.  Your insisting that if you can find a
single instance where it would not prove useful, and therefore
claim that it is _never_ useful, is irrational.

>> I think you are in for a *huge* surprise.  Have you read
>> anything at all about what AT&T's new CEO wants to do?  I'm not
>> so sure that he and AT&T should be the ones they allow to do it,
>> but he has made it clear that his goal is to eat your lunch.
>
>	I was of the opinion that AT&T wants to become a cable
>company because they can't make it in the local telephone
>business. Is this true?

I am not paid to speak on this forum for AT&T.  What they have
published in press releases, and what others have published in
news analysis is that because they have been unsuccessful at
gaining _entry_ to the local telephone business (because LECs
are not willing to give their business away, even if the FCC
and Congress wants them too!), they have decided to attempt
to corner as much of the cable market as possible and use that
as a base to enter the dialtone business.

I wouldn't claim to have the ability to determine if it will be
successful.  But I do think that it is obvious that at some
point in the future, by one avenue or another, there will be a
number of major competitors enter the business using entirely
new paradigms for service delivery.

Additionally it appears that in perhaps as soon as ten years or
perhaps as long as 15 to 20 years the current bread and butter
for IXC's and LECs alike, message switched toll traffic, is
going to disappear almost entirely in terms of revenue
generation.  Some companies clearly believe that cable systems
using IP will be the basis for that change.

>> GTE seems to be positioning itself with exactly that in mind
>> too.  I haven't paid much attention, but for example they just
>> sold off all of their exchanges in Alaska.  These small
>> exchanges are a gold mine for a rural only company that can
>> qualify for NECA tariff and revenue sharing, but for companies
>> whose main base is large or urban exchanges, they are a drain.
>
>Selling off assets in Alaska and other states because they
>no longer provide the income that you require is not  a
>competition problem. Having one of your rural only phone
>companies with all their government handouts coming into the
>same town served by GTE or any other company and trying to
>get customers from the ILEC is competition.

The government has for now exempted those locations from
competition when they are served by companies that meet the
criteria.  GTE doesn't, and therefore those locations are going
to be a severe drain on GTE.  Selling them off is indeed a smart
move.

For other companies, those locations can and will be a virtual
gold mine!

>BTW GTE has
>always been considered a rural phone company. Selling the
>exchanges in your area just means that even GTE doesn't see
>any profit in serving your area. Lets face it they are being
>purchased by Bell Titanic and are getting rid of areas that
>do not meet profit guide lines. Alaska is not the only state
>they are getting rid of. The bottom line is everything,
>NOTHING else matters.

That is true.  But as I have mentioned, for companies in
a different legal position, those same locations will be
very profitable.  GTE certainly didn't have any difficulty
finding a buyer!

  Floyd

--
Floyd L. Davidson                                floyd@ptialaska.net
Ukpeagvik (Barrow, Alaska)                       floyd@barrow.com
     North Slope images: <http://www.ptialaska.net/~floyd>

From: floyd@ptialaska.net (Floyd Davidson)
Newsgroups: comp.dcom.modems
Subject: Re: What to ASK Telco - follow-up
Date: 27 Jul 1999 07:33:09 GMT

Brian <bsilvers@prodigy.net> wrote:

>Not always true.  Depends on the quality of the lines and the switching
>equipment.  The line I'm on right now is a pair gain line.  Connect solid
>at 49,333...  3COM 5687-00 with 5.0.0 flash (11/5/98).  Modem diags
>report only 1 A-D conversion in the loop, high SNR (usually 50-53dB),
>and v.90 peak speeds of either 52000 or 53333...with low BLER counts
>and throughput (depending on ISP network conditions) range from
>4-6K...most hanging out around 5-5.2K...
>
>It is true however that you shouldn't mention "modem" or "internet"
>to any telephone company.  Most all of them will then revert to
>the olden days of when 9600 or 14400 modems were "king" and say
>that they only guarantee up to that level of service....due to fax speeds
>being in that range...

While it is nice to see someone post a verification that "pair
gain" can be implemented correctly for the perspective of an
Internet user, I can't agree with the last paragraph.

Ten years ago it was hard to find anyone at a telco that had
ever heard of the Internet, and very few of them had experience
with dialup modems (as opposed to a great deal of experience
with leased line modems).  But today most telco techs use the
Internet, both at home and at work.  They are not technical
experts on computer networking, but probably have a *far* better
understanding of what is going on than the average Internet
user!  You and others may have more familiarity with all the
terms used here about modems, but they actually understand how
that relates to a telecommunications systems (and the average
computer user, network admin, etc. hasn't got the foggies notion
how it works!).

As to mentioning modems and the Internet to the telecom tech, I
advise that you *do* tell them exactly what is happening.  As to
them making claims that 14.4Kbps is the only level of service
they guarantee... well that is in fact *exactly* the truth.

  Floyd

--
Floyd L. Davidson                          floyd@barrow.com
Ukpeagvik (Barrow, Alaska)