Thread Subject: Re: teitac-telecom, PSTN Reference

Note

This archival content is maintained by WebAIM and NCDAE on behalf of TEITAC and the U.S. Access Board . Additional details on the updates to section 508 and section 255 can be found at the Access Board web site.

From: Michaelis, Paul R. (Paul)
Date: Mon, Mar 26 2007 8:30 AM

Tony,

Quoting from the web page you cited:

"Public Switched Telephone Network communications
(PSTN, T.70) The collection of interconnected systems operated by the
various telephone companies and administrations (telcos and PTTs) around
the world. Also known as the Plain Old Telephone System (POTS) in
contrast to xDSL and ISDN (not to mention other forms of PANS)."

Note that this definition seems to say that services such as ISDN are
*not* considered to be PSTN. Presumably, this would also mean that
digital telephones, operating behind a PBX, are not PSTN. These phones
are not IP either. This would mean that the proposal to classify
equipment as either PSTN or IP will leave an entire category of telecom
equipment without an obvious home.

A reason why I am passionate about this issue is that I've wasted a
tremendous amount of time over the past seven years, arguing with people
-- including people in my own company -- about how to interpret the
Section 508 requirements. "This requirement means this!" "No, it means
that!!!" As part of the Section 508 refresh, we need to write
unambiguous requirements that cannot be misinterpreted. Without that,
contract officers will continue to buy whatever the heck they want,
manufacturers will continue to sell inaccessible solutions to the
government, and the likelihood that my company will let me build
anything else will be close to nil.

Getting back to the use of "PSTN" to describe non-IP phones, my fear is
that some people who are looking for an excuse to avoid investing in
accessibility -- including people in my own company -- will claim that
ISDN and digital phones behind a PBX are exempt because they don't fall
into either category. During the next seven years, I want to spend more
time building solutions, and less time arguing about whether they ought
to be built.

-- Paul Michaelis




-----Original Message-----
From: Jasionowski, Tony [mailto: = EMAIL ADDRESS REMOVED = ]
Sent: Monday, March 26, 2007 7:42 AM
To: = EMAIL ADDRESS REMOVED =
Subject: [teitac-telecom] teitac-telecom, PSTN Reference

Paul,
"wired analog and wired non-IP digital" seems odd considering we also
have "cordless analog and digital". I suggest we stick with PSTN for now
unless someone can propose acceptable alternative term. FYI, anyone can
easily find the definition of PSTN at several web sites including:
http://dictionary.reference.com/browse/pstn
Therefore, PSTN may not be so confusing after all.
Tony Jasionowski
Panasonic


Message: 1
Date: Sat, 24 Mar 2007 17:41:44 -0600
From: "Michaelis, Paul R. (Paul)" < = EMAIL ADDRESS REMOVED = >
Subject: Re: [teitac-telecom] Telecommunications Working Group- Draft
1 Template for 508 in WORD
To: "TEITAC Telecommunications Subcommittee"
< = EMAIL ADDRESS REMOVED = >
Cc: = EMAIL ADDRESS REMOVED = , Creagan Tim
< = EMAIL ADDRESS REMOVED = >
Message-ID:

< = EMAIL ADDRESS REMOVED = >

Content-Type: text/plain; charset="us-ascii"

Regarding the reference to "PSTN" versus "IP" in many of the proposed
requirements, I agree that IP telephones should be treated differently
but am concerned that describing the non-IP systems as "PSTN" will
create a lot of confusion.



The "PSTN" is the Public Switched Telephone Network. If asked to
describe a PSTN telephony endpoint, I imagine that most telecom
specialists would say it was an analog device that complies with the FCC
regulations that govern residential telephones. If we use this
definition, I can't remember the last time I've seen a government
procurement of telecom equipment that falls into this category.



There are at least three different categories of non-IP wired telephones
in common use today:



Residential analog. (This is what most folks would classify as a "PSTN
phone.")



Enterprise analog. (The distinction between these phones and
residential analog sets is that the enterprise sets connect to the PSTN
via a PBX or switch that is owned by the enterprise. An important
electrical distinction is that, unlike enterprise analog telephones,
residential telephones are required to do "automatic loop length
compensation" to account for distance-related differences in electrical
resistance and capacitance in the line between the phone and the
associated central office. Another important distinction is that many
enterprise analog solutions are actually hybrid analog/digital sets.
The audio stream is carried as traditional analog, supplemented by
digital signaling that, for example, controls the phone's display.)



Enterprise digital. (This is what is purchased in the vast majority of
non-IP wired telephony procurements by the government. All audio
transmissions entering or leaving the phone are digitally encoded,
typically using the G.711 algorithm that I described in a previous
message. Signaling and call control is also digital, typically using
protocols that are proprietary to the individual manufacturers.)



Wired residential analog systems, wired enterprise analog systems, and
wired enterprise digital systems are able to transmit Baudot TTY signals
with 100% reliability. IP telephony is different.



Depending on network characteristics, the mechanisms that are used in
order to transport voice signals within IP networks are not always
suitable for transporting Baudot TTY tones. Common sources of
impairment include packet loss and the use of voice-optimized audio
compression algorithms, such as G.729. For this reason, a traditional
TTY device used in conjunction with an IP phone (for example, via an
acoustic coupler) is not always going to be reliable.



The good news is that, although not yet in common use, there are
international standards that describe how to transport text and Baudot
tones reliably on IP networks. In general, these techniques do not use
the same audio encoding and transmission mechanisms that are used for
voice. I agree completely that the availability of these techniques,
coupled with the inherent unreliability of Baudot TTY transmissions via
the "voice channels" of IP telephone networks, makes it essential that
we treat IP telephony as a special case in the Section 508 refresh.



Getting back to the original topic of my message... We need a term
other than PSTN to describe non-IP wired telephony. Ideally, it should
be a term or description that would make sense to someone who is not
knowledgeable about telephone engineering. To start the ball rolling, I
suggest "wired analog and wired non-IP digital" as a substitute for
PSTN. (I deliberately use the word "wired" because wireless devices
have special characteristics that may need to be addressed separately.)



-- Paul Michaelis

WebAIM is an initiative of:
Center for Persons with Disabilities (CPD) Utah State University