Before the Federal Communications Commission Washington, D.C. 20554
In the Matter of Implementation of Sections 255 and 251(a)(2)
of the Communications Act of 1934, as Enacted by the Telecommunications
Act of 1996
Docket No. 96-198
The Trace Center of the University of Wisconsin-Madison and Gallaudet University's Technology Assessment Program have extensive experience in engineering, research, consultation and guideline development regarding accessibility of technology to people with disabilities. The two groups collaborate on a Rehabilitation Engineering Research Center on Telecommunications Access, funded by the National Institute on Disability and Rehabilitation Research (NIDRR).
The opinions expressed here represent those of the authors and not those of the host institutions or funding agencies.
The Trace Center and Gallaudet's Technology Assessment Program thank the FCC for the opportunity to comment on emerging technologies that affect the accessibility of telecommunications for people with disabilities. By anticipating the effects of new technologies and shining a light on emerging telecommunications access issues, the FCC can encourage industry to build in accessibility during the development of standards and products.
One of the fundamental issues in this inquiry is whether the FCC has jurisdiction with regard to telecommunications accessibility when IP is used in telephony. We assert that this is unequivocally the case. When IP technology is used to orginate, route, and terminate telecommunications as defined in the Telecom Act, it is subject to the requirements for accessbility and compatibility where readily achievable.
In the following section we outline the basis for the FCC's jurisdiction in this area. Following this section we address the NOI's specific questions on access issues, FCC response, industry response, other new technologies, and other issues.
The definition of telecommunications in the Telecom Act is:
The term 'telecommunications' means the transmission, between or among points specified by the user, of information of the user's choosing, without change in the form or content of the information as sent and received.
This definition is a functional definition and not technology-specific.
The definition of information services in the Act is:
The term 'information service' means the offering of a capability for generating, acquiring, storing, transforming, processing, retrieving, utilizing, or making available information via telecommunications, and includes electronic publishing, but does not include any use of any such capability for the management, control, or operation of a telecommunications system or the management of a telecommunications service.
This definition excludes conversations such as those typically conducted during telecommunications. It also is not technology-specific.
Section 255 covers telecommunications. IP in and of itself is not an information service. The ability to place phone calls is involved. Conversation is the primary function. The underlying technology and the mode of communication (voice, text, etc.) are not specified as limiting factors in the definition of telecommunication. Transmitting phone calls via IP does not change them from telecommunications to information services. IP is a transport protocol and is not confined to Internet technologies. It can be (and is being) used for telecom transport as well.
The definition of telecommunication does not limit itself to conversations using voice. In fact, conversations using text (such as using TTY's) were not only envisioned but were commonplace and protected elsewhere in Federal legislation. Thus, in considering what constitutes telecommunication it is urged that telecommunication include conversations that are carried out using speech, sounds, text and any other modalities that are used for carrying out a conversation.
There are already services available where an individual can talk into a standard POTS telephone on one end of a conversation and the person at the other end (in a noisy environment or in a meeting) can view the conversation in text. The second person can then respond (silently) in text, which is converted to speech for the person using the phone. Because half of the communication is occurring in text and half is occurring in speech, does that mean that only one person is telecommunicating?
In other cases, one person may be talking and the other person listening (speech in one direction), but the individual types the responses back, which are viewed by the first person (text in the other direction). Is this telecommunication in one direction but not telecommunication in the other?
The clearest action, and the one that will be most robust in light of the continuing technical advances would be to recognize all forms of conversation over technology as being telecommunication. We would propose that if a device or service permits phone calls (to a phone number or its future equivalent or alternative) that device or service is undeniably within the realm of telecommunications and, as such, would be covered by Section 255.
When Congress passed the Telecommunication Act, it was their intent that Section 255 apply to those things which people used to call up and converse with each other whether they were shaped like footballs or speakerphones or antique phones or objects of art. It is clear that they tried to separate information services from telecommunication services. However, they did not provide any indication that a telephone call that went over a piece of fiber optic network should be treated differently than a phone call that went over copper wire or one that was transmitted using a microwave link. The distinction was in whether it was a telecommunication versus information service, not whether it was carried over one particular type of transmission technology or another.
In the not-too-distant future it is possible that all telecommunications will be handled by adapting IP to voice traffic. If the term Internet is used to refer to the information backbone in our country in the future, then there will likely not be any phone calls or telecommunication of any type that will not occur over these information networks (collectively known as the Internet). It should be noted that this 'Internet' will be all digital - but not necessarily all TCP/IP. (Just as all legs of the current 'Internet' are not TCP/IP. Other technologies are used to transport over wireless, satellite and even some long line trunk portions of the net.)
Even today, some POTS to POTS phone calls are carried at least part of the journey (if it is a long one) in a packet based system. Telecommunications equipment comprises a complex myriad of different technologies that are used to enable communication by users between two points of their choosing.. The fact that IP is one of these technologies is usually completely invisible to the users.
In some cases, the same IP connection is used to provide both Internet access and what looks like regular telephone service from a location such as a business. That is, telephone users at a location will pick up a phone and dial a phone number believing they are using a "standard phone" to call another "standard phone." In fact, the phone they are using may communicate with the telephone company using a TCP/IP line. The phone at the other end of their call may either be a standard POTS phone or another similar phone that looks and behaves like a POTS phone. Should employees or customers of one company (or hotel) be covered but not those of another?
There seems to be a misimpression that transmitting phone calls via IP changes them from phone calls into information services. If digitizing a call or transporting it over a digitized network changes telecommunication into an information service then a large percentage of today's phone call are already not telecommunication.
IP-based networks will (surprisingly quickly) evolve into the information AND communication system for the entire country, which will permeate our businesses, schools, homes and lives in the same way that electricity does today. This system will carry all of our information services AND our telecommunication services. It will extend to wired and wireless systems and will even operate over the standard power wiring in our homes and businesses.
It is already possible to connect products, including telephones, throughout your house by simply plugging them into the electrical outlet and using the electrical wiring from the house for the network wiring. This will include the ability to connect both traditional PSTN (true today) phones or IP telephones (in the near future) by simply plugging them into a standard wall three-prong power socket.
In the NOI it was stated that there were two basic ways of providing IP telephony (paragraph 177). One was phone-to-phone and one was computer-to-computer. In fact, IP will be used throughout telephony for reasons of cost savings and other advantages of mixing data and voice communications with one technology. We can see IP telephony occurring in all of the following cases:
...and any combination of the above.
As pointed out in the FCC Office of Plans and Policy working paper #31, "legacy regulations should not automatically be imposed on new technologies." In many cases they may no longer apply or be necessary. Some have read this to say that they should never be applied. We suggest that the use of IP as a transport protocol for telecommunications does not in any way alter the fundamental need for telecommunications to be accessible. Nor does the incorporation of IP as the transport protocol ensure accessibility of the instruments at either end or the transport "wire" itself. Thus, moving from old to new technologies for transport in this case (disability access) is not a case where carrying regulations or requirements forward is undesirable or unneeded.
Working paper #31 also states: "When Internet based services replace traditional legacy service, begin to deregulate the old instead of regulating the new." Again, there are issues (such as competition) where this may be true. But it should not be taken as a maxim. There are issues where it is also clearly desirable to carry regulations forward, updating them as necessary. These would include areas that deal with health, safety and human rights. For example, in a slightly different arena, if email replaces mail it would not make sense to drop regulations around mail fraud. Rather, fraud regulations should be carried forward and updated as necessary to cover fraud perpetrated using email. If a new type of transmitter is designed, RF interference with existing technologies or emergency services should not necessarily be dropped. Similarly, it is doubtful that Congress intended for section 255 to be dropped if the transport mechanism for carrying the phone calls across the country were changed or if combination phone-and-something-else appliances became popular in homes, hotels or businesses.
As a result of the above blurring of lines between the various technologies (as seen by the user), it is important that basic product regulation not be tied to the technologies but rather to the function or service that is provided. Otherwise, it is likely that a consumer could switch from one telephone service to another apparently identical service (which allowed them to use the same phones in their home as they did before) and then later find that protections that were available to them last month were no longer available because the "new phone service" used a technology exempt from the regulations. In the case of some consumers with disabilities, this may mean that the phone service or product they just signed a contract for is not usable for no apparent reason.
Although in the past, it was possible to find telephones which did not include computer technology, there are essentially no telephones today that are produced which do not have computer technology in them. The cost to produce small central processing units has dropped to the point where it is even possible to find small computers in electric shavers where they do nothing but help control speed and battery charging. Similarly, even the least expensive phones have small processors in them that control all of the functioning of the phones.
We are also likely to see functions, currently carried out by personal computers, that will increasingly be carried out by information appliances. We are already seeing devices on the market which send and receive e-mail, browse web pages, and make phone calls, but which have none of the typical characteristics associated with personal computers (floppy disk drives, etc.). In fact, some of the devices have handsets and look more like telephones than computers. Increasingly, we are going to see a continuum of products that no one would suspect had computer technology in them (light switches, coffee makers, doorbells) all the way up to devices that clearly look like our current envisionment of a computer. There will be no clear breaks in this continuum and it will be difficult or impossible to decide where a computer begins and where it ends.
In addition, we are also likely to see homes where all of the electrical devices in the home are all interconnected together in a network which is likely to include, at least some place in the network, one or more things which might be thought of as computers. Further, the function of the various items in the environment will be affected by the other items and the computers.
In other homes, there may be nothing that resembles a computer as we think of it today. Instead, it would just be a number of different information and communication appliances scattered throughout the house. Individuals will be able to make and receive phone calls from a variety of different types of devices, including those that are both audio only and audio/video.
As a result, it is not useful to try and separate telephones or communication into computer based and non-computer based. This will only invite confusion as the lines between computers and information in communication appliances continues to blur.
Instead, it is recommended that telecommunication be treated as a function. Any products that are designed and marketed to be used to carry out telecommunication be recognized as being covered under the Act (at least those portions of them which are used for telecommunication).
The specific answers to this question are best answered by industry.
Press reports (see Pappalardo, D., and Green, T., Jan. 11, 2000) indicate that major carriers are readying for significant deployment of IP for telecommunications carriage in the next 18 months. CPE using IP for telephony is not currently widely used outside organizations, according to the article: "A recent study by Faulkner Information Services indicates that only 7% of enterprise customers have deployed any voice over IP, and those who have installed it in only small pilot projects in least-critical environments."
It appears that, although the pace of change is rapid, and there are plans for widespread application, the deployment has not yet begun in earnest. There is therefore an opportunity for industry to solve any problems that exist and to avoid new barriers before they become troublesome.
We do not have statistics on usage patterns by people with disabilities. If the service is inherently accessible to an individual with a disability (for example, someone who uses a wheelchair and has no limitations of use of the arms), then we have no reason to believe that usage differs from that of the general population. Where a product does present a barrier, it is unlikely that you will find persons with that disability availing themselves of the service.
Two of the main problems with tracking usage patterns and barriers with IP telephony are that (1) the user is not necessarily the customer (may encounter the technology at work or while traveling), and (2) the source of the problem is nearly impossible for the user to trace.
A better line of questioning might be "What is the usage pattern in the general population (i.e., as in Question 1 above)? What is or will be the impact on society at large or on anyone who could not use the technology? How widespread would use be in general? Will the technologies be or become pervasive or standard? Will use (or inability to use) the technologies affect employment? Will they affect our culture of communications? If a telecommunications technology is widespread, unavoidable, able to affect employment, and part of our culture, then it should be a high priority with regard to accessibility. Unfortunately, if we wait until a technology is already widely deployed, we will be in a retrofit situation.
For example, with voice mail, IVR, and wireless technologies, the technologies have become widespread before access issues were addressed or solved. In the case of IP telephony, there is an opportunity for industry to solve them in advance of their widespread deployment.
It is clear that telecommunication functionality is moving from its current locations (in CPE, in PBXs, Central offices etc.) and being distributed across the entire continuum of CPE to (multiple interacting) network services. Further, in most cases, the individuals who are using these services will be completely unaware as to whether or not they are located in a central location or at the customer's location. Finally, the impact to the consumer, their inability to access these services and functionalities, will be the same whether they are central or at the customers location.
It is also clear from the wording of the Telecommunication Act that the drafters intended the telecommunication system to be accessible from end to end, because b customer premises equipment, network equipment, and services were covered in the scope of 255.
This question also asks a specific question about functionality which resides in a computer connected to a PBX rather than in the PBX itself. As discussed above, referring to something as being in a computer versus in a PBX is a rather constructed way of looking at things. First of all, the PBX itself contains a computer--in fact, a fairly large number of computers (microprocessors) working together to originate, route, and terminate telecommunications.. Adding another processor to this cluster and delegating some of the tasks off to this other processor is really no different than inserting another board into the PBX case. It should be considered the same as any other hardware or software upgrade. The future of organization based telephony is likely to be a TCP/IP version of the PBX. The open architecture, ease of changing components, and ability to integrate voice telecommunications with data are attractive features. (Riezenman 2000)
Unless these future versions of the devices which are used to carry out the telecommunications functions of today's business phones are required to follow 255 regulations (where readily achievable) then employees with disabilities working at these locations and customers who must use these facilities will lose basic phone accessibility as companies shift to these new technologies..
This would be unfortunate since the way these new telecommunication products are being designed (and the technologies being used to design them) make it easy to design them to be cross-disability accessible without impacting price or functionality for other users. In fact, most all of the accommodations would also be a benefit to large numbers of other users (as curbcuts and closed captions on television are today).
We've just carried out a rather extensive study of different industries looking at when and why companies build accessibility into their standard products (practice universal design or not) and what factors seem to account for it. Although there are many different factors that seem to account for exactly when and where accessibility is considered in standard product design. However the only two factors that appear to have reliably affected this practice (building access into standard products):
It is not that companies don't have a sense of duty or feel that building access in would be important. It is just that there are so many things that need to be done and it is so competitive that:
It is therefore critical that the FCC play a role in this process if anything is to actually happen to ensure that future technologies are accessible.
Specifically, the FCC must:
Phone to phone IP telephony is in its infancy. An assessment of impact would require cooperative efforts of the user community and industry. One area that has been documented is the problem of garbling of TTY conversations. Other issues are more hypothetical.
Some voice-over-IP technology garbles TTY. Currently the problem is very limited because the technology is not widely used in the telephone network. However, when carriers increasingly employ IP telephony in their networks, the extent of the problems may increase. CPE is of concern because it may affect consumers' employment (if seeking employment in an organization with an inaccessible telecommunications system or carrier) and safety (e.g., when traveling and staying in a hotel with an inaccessible phone system).
Access problems must be identified by people who use the technology. But there is no way for either party on a garbled TTY call to know the source of the problem. If a TTY user places a call to another TTY and garbling occurs, most users are likely to think the TTY is broken. Some may know enough about text telephones to investigate whether background noise is causing the problem (particularly if the device is coupled acoustically) but since most text telephone users cannot hear well, there is typically no way to find this out with any certainty. Few would suspect that network technologies were the source of the problem, because this has rarely been the case in the past. To complicate things further, wireline network technologies will not necessarily be to blame. The garbling could be the result of a LAN/PBX of one party's employer or a hotel in which one of the parties is staying.
Transmission problems on a voice call can often be described in terms of the sound - static on the line, distorted voice, noisy background. With TTY garbling, there are no such noticeable qualitative differences in the garbling of characters. This makes diagnosing a problem even more difficult.
It is important to note that TTY as a service is not limited to two-way live chat by text. Another important function in this service is generally called voice carry-over (VCO) and hearing carry-over (HCO), in which one party uses voice and the other, text. Alternating voice and text (or in some forms, having simultaneous voice and data available during the entire call) is an important component of TTY over voice networks. Currently, VCO and HCO would appear to be affected only to the extent that TTY is garbled. However, in future implementations of voice and text, the situation is uncertain. The protocol would need to support both functions on a call; ideally simultaneous use of both channels. VCO and HCO are required by the FCC under the TRS rules. Both must continue to be supported. In fact, it would appear that digital technologies would enhance the mixing of the two, but there is no guarantee that the industry would support it.
We raise the question of intelligibility of voice over IP to people who are hard of hearing. Compression of speech for efficiency/cost savings is a strong trend in telecommunications. Listeners' judgment of speech quality over packet-switched technologies is measured through a standardized metric called the Mean Opinion Score (MOS). The listeners in evaluations of codecs are hearing people. People who are hard of hearing, especially those with severe hearing loss, may find it more difficult to understand speech that has been compressed and that is judged of lower quality than "toll" quality speech. We do not know whether this will be a problem; in fact it is possible that less-noisy connections could improve accessibility. We invite industry to work with us in assessing the effects on people who are hard of hearing.
If speech quality is lower with some VoIP applications, there could also be an adverse effect on people whose speech is impaired. They could find themselves less able to be understood by others. This would include hearing people with speech disabilities, as well as people who are deaf or hard of hearing and who use VCO. Again, this type of problem would be difficult if not impossible for consumers to trace.
The definition of telecommunications implies that conversation is covered as a major function of telecommunications. We propose that mode of conversation not be limited to voice, but include text and video telephony. This concept of "Total Conversation" has been devised by Gunnar Hellstrom as a way to ensure accessibility of multimedia telecommunications in the future, across different types of technology. (See Appendix **). The service definition for Total Conversation is now being crafted in ITU-T Study Group 16, and exists more informally in other ITU documents.
In addition, we urge the FCC to establish text telephony, with VCO/HCO as a component, as a class of service. This is proposed to help ensure that deaf people have a stable means of telecommunication just as voice users do, as technologies change. This step would also assist in compatibility efforts by establishing standards. Finally this step would assist in the migration of text telephony from analog to digital technology.
The legislative history on text telephony is not limited to Section 255. Text telephony is also the subject of the Telecommunications Enhancement Accessibility Act of 1988, ensuring access for TTY users to Federal Government, and the Americans with Disabilities Act of 1990, which required text telephone accessibility to 9-1-1 service; established a telephone relay service to open up telecommunications network to text telephone users; and in guidelines establishes telecommunications access requirements in public accommodations. Although the FCC has limited jurisdiction under these laws, the FCC would be contributing to problems of implementation of these laws if it did not shepherd text telephone functionality (via compatibility or direct accessibility) through the changing technologies of telecommunications.
We find that, with every industry segment addressing compatibility for the first time, the same information is needed: key functionality of TTY, signal characteristics, services that must be supported. We have taken steps, working with the TTY industry, to formalize in an industry standard the characteristics of Baudot TTY signals in products in the field. A draft standard has recently been favorably balloted and should become a TIA standard shortly. But the TTY market has changed in the past few years with the diffusion of proprietary protocols, which perform better than traditional Baudot but pose new compatibility issues and issues of intellectual property. The service definition for TTY would need to include important functions such as VCO/HCO, visual means of monitoring call progress including whether answered by voice, ring signaling, and a means of letting a hearing person (e.g., 9-1-1 agent) know that a TTY is on the line, ability to successfully transfer the calls, and other functions.
Video telecommunications should also be defined as a service, as it meets the definition of telecommunications and for some people it is the only means of giving a counterpart to voice service (through sign communication by deaf people).
In addition to phone-to-phone IP telephony, addressed in 179-A above, other implementations of IP telephony (and other computer-telephony integration technologies) will present some new barriers as voice communications begins to incorporate more visual elements.
In the area of visual impairment/blindness there is a very severe threat of loss of access. Newer phones are increasingly using techniques such as "soft buttons" and touchscreens. Soft buttons are buttons which appear along the edge of a display and which do not have any set function. The function of the buttons changes from moment to moment with the current function being displayed on the screen. For example, there may be two buttons immediately below the LCD screen which are labeled (on the LCD screen) as "directory" "and redial." As soon as you begin to dial numbers, the functions of these two buttons immediately change to "backspace" and "okay." Once the call is in progress, the functions of the two buttons change to "transfer" and "hold." As a result, an individual who is blind has not idea what the function of these buttons is from one minute to the next unless they can see the display (or some other provision for access is provided). Touchscreens are also increasingly appearing on products. The problem here is twofold: (1) the individual cannot feel where the buttons are and (2) they cannot identify what they do since they change from one minute to the next as the screens change.
There are very easy mechanisms for making both types of phones accessible to people with low vision and blindness. However, unless telephones that operate over packet networks are covered (as are all other phones) then, as all phones evolve into packet based technologies, they would evolve out from under the Section 255 regulations. One might assume that the people with disabilities could simply purchase the accessible phones (if there are any on the market). Unfortunately, a large number of the phones that an individual must use are those they find in where they work, in the hotel rooms, in public places, and even in homes they are visiting. In all these cases they do not have the ability to choose but must be able to operate the phones they encounter.
We defer to industry on this question, noting only that there are some standardisation activities to make the gateways between the circuit-switched technologies and packet networks be "text telephone aware." (See our discussion of text telephony standardization and migration, section 8E, below). But the main effort in that standardisation is currently financed by the Rehabilitation Engineering Research Center on Telecommunications Access through Gallaudet University.
Standards are important, but they must be implemented before being of any importance to the user.
Using the definition of phone to phone IP telephony as described in the NOI, there are no known advantages that would boost accessibility. The type of phone to phone IP telephony described in the NOI involved the use of a standard wire line phone which was connected to an IP trunk at the phone company then later converted back to an analog phone and delivered via another standard POTS phone. Since this, essentially, involves the chaining of the two technologies, the concept of "weakest link" applies. Those things that would be possible over a purely analog network would not apply since it is a packet network in the middle. Those advantages which might exist over a purely packet network would not apply since both ends are analog. Thus, there are no known advantages over pure analog system (except possibly the greater quality of signal over long calls due to the digital transmission in the middle. However, this quality would be no greater over IP than it would over be over ATM or other digital transmission technologies.)
In contrast, pure phone to phone IP telephony would provide many potential advantages. If IP communication was used from end to end, then there are a number of potential advantages over an analog end to end phone. the ability to select one or more communication modes offers the potential for access by users who can perform some functions but not others. Because the products are heavily based in software and because they are still being developed, building in access could still be done without retrofitting.
Some examples below highlight the functions of communications and conversation.
Yes, most definitely. In fact, if it is clear that IP telephony is just a subset of telephony and, therefore, covered under the Section 255, most of the accessibility provisions can be incorporated into the standards and into standard practice from the beginning so that accessible phones are the normal practice.
Yes, but only for pure IP telephony (that is, telephony that is IP from end to end). Situations where conventional telephony is chained serially with IP telephony will continue to have the limitations of conventional telephony.
The advantage of this technology is that it is open and software based. If developed early enough, many of these features will be readily achievable. Will market forces alone make it happen? No one knows the answer to that question, but if history is a guide we can project that some new functions will provide enhanced accessibility, while others will erode it. We look at the example of digital wireless telecommunications, whose access problems are being resolved for TTY users through FCC orders, but whose access problems for hearing aid wearers are not yet solved. On the other hand, many people with other types of disabilities, such as those that affect mobility, find low-cost wireless telecommunications a tremendous aid to accessibility and independence: Accessibility is enhanced for them.
As assistive technology evolves to include IP connectivity, there will be great benefits. Individuals who are not physically able to operate IP telephones due to extreme physical disabilities will be able to use special assistive technologies to control the phones and/or to connect directly to the IP telephony system with their aids.
There are already standards such as the Alternative Interface Interaction Protocol (AIIP) being developed by the National Center for Information Technology Standards. However, these standards need regulations to see that they are implemented and built into products. The V.18 standard for telephony modems is a good example of a standard which was developed, but without regulation, has not been implemented
The question as posed seems to assume that telecommunications consumers will be able to choose whether or not to use IP telecommunications. However, as the use of IP spreads throughout telecommunications, and so incompatibility of assistive technology could cut off the ability to communicate using the telecom products and services they encounter on a daily basis. This would be a serious access problem. We hope that industry-wide, the problem can be solved before it begins to affect consumers.
Because the possibility of garbling TTY is known, compatibility with TTY is a key issue in this proceeding. It is our view that text TTY telephones should be accommodated by technologies just as voice telephones are. The new voice technologies being developed are backward compatible with analog wireline phones. If this is the case, they should be compatible with analog TTYs as well.
At the same time, we recognize the obvious need to migrate away from analog devices to digital devices in as effective a manner as possible, without losing functionality in telecommunications during the transition. This is a goal which many share. Text telecommunications is inherently digital and should merge readily with other digital communications.
Some of the commenters in this proceeding have been working in the background with some engineers in industry to try to move forward with compatibility between TTYs and other text communication protocols leading to the eventual migration from TTY. The first effort in this regard was an effort by Gallaudet to meet with TTY manufacturers in the early ‘90s to discuss better, standardized implementations of ASCII in TTYs. Recognizing that the issue is an international one, Richard P. Brandt took the issue of compatibility between modems and TTYs to ITU-T Study Group 14, of which he was at the time Vice Chairman. The result of his initiative was Recommendation V.18, which addresses not only compatibility between domestic analog TTYs and data communications equipment, but also with international text telephone protocols.
This work within ITU was continued and expanded into the all-digital environment by Gunnar Hellstrom of Sweden. His work and the work of industry in crafting digital text channels and protocols are exemplary. See "Total Conversation and Text Telephony in the IP Revolution", appended.) The development of standards for digital text chat to be used in a variety of communications media, whether IP or not, is a major step in the right direction. It looks ahead to the future and lays a path for migration and integration of text chat into all forms of multimedia. (We note as an aside that his early work on accessibility within the ITU was supported by the Swedish counterpart to the FCC.)
As Hellstrom notes, "It is a good habit of the industry to unite in creating standards for communication methods and compete in creating the smartest implementations of the standards" (Hellstrom, 1999). The contributions of engineers in the standards groups to the development of T.140, V.18, and emerging standards have been admirable. It is in the latter area - competition during implementation -- that we have lingering concerns. To date, the implementation of V.18 has not occurred in "mainstream" products, or for that matter in any domestic products, although it was initially approved in ITU-T in 1994. British Telecom funded development and testing of V.18 and even developed a testing protocol - but still the technology has not been picked up. T.140 has not been implemented either, to our knowledge. There is a growing number of technical standards for accessibility, but only those required by government are being implemented.
While commenting on text conversation standards, we note that mobile/wireless telecommunications is yet another area in which migration to digital text telephony is advisable and should be considered now, as the new generations are being planned and developed. We should aim for a single system of text telecommunications worldwide once end-to-end digital telephony is in place if we want to see an eventual migration from the universal TTY solution to text communication using more standard technologies.
Almost all very long calls are digital in the middle. All of the big carriers probably use high enough quality in their digital legs that transmission quality should pose no problem, even for TTY signals and for those with poor speech or hearing. However, smaller carriers may, in an attempt to gain a cost advantage, provide lower quality digitalization that could cause problems for individuals with speech or hearing disabilities as well as those who use TTY's.
When trying to project the use of various kinds of IP telephony by persons with disabilities, it is probably best to look at the projected usage by everyone. In general, if any significant portion of the standard population is using IP telephony, then this will be mirrored in the disability population. Also, if it is being used in public places, then those will be the systems available to people with disabilities as well. If they are not accessible, this will be a real problem. If they are not required to be accessible, then accessibility is likely to be irregular at best.
With regard to the extent to which IP telephony will take over for conventional switch circuit telephony. It is clear that in the long run the trend is toward some type of packet based transmission of information throughout the networks. There is a tremendous amount of installed based, however, which may provide some inertia. However, in the light of competition, installed basis can quickly present an operational or competitive disadvantage. If this happens, they can be quickly swept away.
As noted in the preface above, there are not really two separate kinds of IP telephony. Instead, there is a continuum that goes seamlessly and continuously from what might be thought of as phone based IP telephony at one extreme and what might be thought of as a personal computer based IP telephony at the other. Since there is this continuum, it will not be possible to clearly divide the range of devices into one category or another. In addition, any attempt to do so will result in a great deal of confusion by FCC staff, industry and consumers given the fact that even the phone based IP telephony would be carried out with phones which include the same computer technologies as found on PC's. There would seem to be no reason to try to differentiate between the two.
One might argue that computer based systems are programmable and, therefore, more flexible than phone based IP telephony. In fact, only an extremely small number of consumers will be capable of reprogramming their computers to behave differently from the way they are programmed in the standard software they purchase. Both computer program based and phone based telecommunication products can be programmed from the beginning to provide additional modes or interface variations which would accommodate people with a wide range of disabilities.
There are a wide range of new technologies that are emerging which would impact the availability of accessible services and equipment. Three key ones are:
These three technologies provide individuals with disabilities with the ability to move about and yet stay connected with telecommunication products and telecommunication networks without having to be able to find and manipulate connectors. This will allow individuals with severe or extreme disabilities to be able to use special assistive technologies with the new telecommunication systems that are emerging.
For individuals with mild to moderate disabilities, the fact that essentially all new telecommunication products have microcomputers built into them, which control their behavior, means that it is fairly straightforward for next and next, next generation of telecommunication products to behave flexibly to meet the needs of a wide range of users.
If the equipment originates, routes or terminates telecommunications, it is in some way connected to the rest of the network - unless it is a closed telecommunications system used only within an organization (and there are few such systems).
The statute covers telecommunications. Because of its function, this type of equipment meets the definition of telecommunications equipment and CPE. The ability to place phone calls is involved. Artificially separating telecommunication products and services into covered and not-covered categories based on the underlying technology alone would not only cause serious accessibility gaps but also lead to much confusion and a very uneven playing field for different companies. It will also lead to a situation where virtually identical products could be covered or not covered simply based upon the technology used to implement them or the networks used to interconnect them. It could also result in phone conversations where the products at one end of a call were covered but not the phones at the other end of the call; even where the phones in one room or area of a hotel would be accessible but not the phones in another room or area of the same hotel. People with disabilities would no longer be able to pick up a phone and call with any degree of confidence. People on some parts of the network would not be able to reach the relay service or 9-1-1, if calling from a locale with a system that garbles TTY either locally within the CPE or within the network.
Federal Communications Commission. (1999). The FCC and the Unregulation of the Internet. (Office of Plans and Policy Working Paper No. 31). Washington, DC: Jason Oxman.
Hellstrom, G. (1999) Total Conversation and Text Telephony in the IP Revolution. Presentation to the VON coalition meeting with Accessibility Actors and the FCC. Monday, December 13, 1999, Washington DC
Pappalardo, D. and Green, T. (January 11, 2000) Voice over IP gathers steam. Available: http://www.cnn.com/2000/TECH/computing/01/11/voice.ip.idg/index.html
Riezman, M. J. (2000). Communications IEEE Spectrum, 37(1), pp33-38.
Respectfully submitted,
Gallaudet Technology Assessment Program
Gregg C Vanderheiden Ph.D.
Appended Document:
Total Conversation and Text Telephony in the IP revolution
by Gunnar Hellström, Omnitor.
Accessible Information Technology consultant
Rapporteur in ITU-T Q9/16 Accessibility to Multimedia Systems and
Services
E-mail:
gunnar.hellstrom@omnitor.se
www.omnitor.se