Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 327
327 manuscript a list of all objects discussed in the manu- script (i.e., a key object list in addition to a keyword list). As mentioned previously, the committee described in Section V could play a useful role in coordinating these data-base activities. VI I I . TELECOMMUNICATIONS. The Panel investigated the increased access to computa- tional facilities and data bases that might result from enhanced telecommunications. Telecommunications services may be divided into three regimes depending on the bandwidth of the service. At the low end are communications over ordinary voice-grade telephone lines at rates up to 9600 baud. In the mid- range are services that use a single digital voice chan- nel at 56 kilobits per second (kbps), and on the high end are services at 1.544 megabits per second (Mbps) (24 digital voice channels plus control bits) and up. The cost of the voice-grade line is just the cost of a long- distance telephone call between the same two endpoints. At the high end, only dedicated channels are available at a typical cost of $50 per mile per month. In the mid- range, packet-switching networks are available. These networks can route packets (a packet can contain up to 128 characters or 1024 bits) to one or more nodes on the network, perform code translations, hold packets until the destination can receive them, and perform other simi lar services. Costs are based on a fixed connect charge (typically $2100 per month) and a packet charge (typically O.OOOS¢ per packet). It is not expected that the costs of these services will drop dramatically over the next decade but, rather only at a rate on the order of 10 percent per year in real dollars. The Panel identified several potential uses for telecommunications: 1. Remote access to the data archives described earlier, 2. Remote access to the data bases described earlier 3. Remote access to centralized computational facilities, and 4. Remote operation of telescopes. The typical application of remote access to data ar- chives would be to transfer data from the centralized ar- -

OCR for page 327
328 chives to the users' local facility for local analysis. This application does not require high-bandwidth telecom- munications. The 56-kbps bandwidth is adequate to allow "overnight delivery" of several large images (202 Mbytes in 8 h). At 5 X 10-4¢ per packet it is already cheaper to send an image in this manner than to mail a tape. The main drawback is the $2100 per month connect charge. This is far too large to be borne by most astronomy depart- ments. If it is possible to spread this charge over sev- eral university departments, then access by remote link may become a cost-effective method of retrieving data from archives. Voice-grade lines are adequate for most accesses of the data bases described earlier. Terminals and modems (which provide the interface between terminals or com- puters and the telephone lines) are proliferating rapidly among astronomers, so this method of access to data bases will become a reality when the data bases are upgraded to support remote access as recommended the previous section. Remote access to centralized computational facilities is, of course, already a reality. Access over voice- grade lines is adequate for analyses of one-dimensional data or for those theoretical computations that do not generate a large quantity of output requiring interactive examination. However, voice grade or even 56-kbps lines are inadequate for interactive image analysis or for those theoretical computations that generate large quantities of output requiring interactive evaluation. mese latter cases require high-bandwidth communications, and the cost of such services effectively precludes this mode of opera- tion. For the foreseeable future, interactive image analysis and large-volume interactive theoretical compu- tations require that either the user must travel to the centralized facility or the computational facilities must be distributed among the users, as already recommended here. Remote operation of instruments is, of course, neces- sary for all unmanned spacecraft. Remote operation of ground-based telescopes can be conceived of in two modes. Telescopes that are at relatively inhospitable or inacces- sible sites may be operated remotely from a control center within the observatory headquarters. Also, remote opera- tions may enable observers to participate in an observing run (i.e., to examine quick-look data and make decisions concerning the short-term course of the observing program) from a terminal at or near their home institutions without traveling to the observatory site. A 56-kbps line, if