In traditional cable systems, any power carried on the distribution plant was terminated at the tap, to lessen the risk of electricity-related accidents for technicians and homeowners. The emergence of lifeline cable phone service (circuit-switched) brought with it a device known as the “power passing tap,” which, as it name implies, passed power to the side of the house. This was to provide plant power to the unit, should the power grid go down and the phone need to stay live.
TCP/IP is a package of different rules that define how data move from one network layer to another. There are seven different layers (see “OSI stack”) and they all count toward the end game, which is moving data around so that they’re all in one piece at their target destination.
Just as protocol in the etiquette sense provides guidelines on how to behave in certain situations — to say “thank you,” to cover your mouth when you cough — IP is a set of rules that tells data hardware how to behave. It’s part procedure, part practice and part policy.
Mostly, IP puts addresses on packets of data. In a high-speed-data network, for example, cable modems and users are identified, in part, by their IP addresses. IP doesn’t baby-sit the data to make assure their safe passage to another destination. That’s on the to-do list for the TCP part of TCP/IP.
Circuit-switched telephone service, provided by MSOs including AT&T Broadband (now Comcast), Cox and Cablevision, works by mounting network interface units on the side of subscribing homes, to provide as many as four phone lines of service. The network interface units (NIUs) connect over the upstream cable path to a headend “host digital terminal,” or HDT, which in turn connects to a traditional telephone switch to connect to the public switched telephone network (PSTN).
Voice-over-IP leverages the two-way Internet Protocol (IP) path put in place with cable modem services. It runs “on top” of cable modems, to move phone calls to, ultimately, a managed private backbone.
The term “thick client” almost always travels with its antonym, “thin client.” Usually, “thick client” is the good guy, and “thin client” is the bad guy (unless it’s accountants talking.) “Thin clients” are usually described as such because they’re too scrawny for whichever new application or service.
Thick clients are the opposite — sufficiently populated with memory and processing power, able to accommodate new services, and usually several hundred dollars more expensive. That makes the whole “thick v. thin” discussion more of a supplier-inspired way of distinguishing “now versus next.” Thin is now; thick is next. What’s thick today is thin tomorrow, and what’s thin today was thick yesterday.
What’s the baseline resource footprint, to decide whether a box is thick or thin? None exists, officially, but the conversation seems to flow around OCAP (OpenCable Applications Platform) techniques. OCAP (version 1.) requires an 8 x 16 Megabyte memory footprint (where the “8” is flash memory, and the “16” is dynamic random access memory, or DRAM).
More is better. In processing oomph, OCAP needs at least a 130 MHz processor — but faster is better.
2) In cable, “thin client” currently means the lowest common denominator in digital set-top feature sets. It’s the opposite of the “thick client,” meaning a digital box loaded with processing power and memory, and capable of doing multiple, complex applications.
Thick v. thin has less to do with physical size than it does with capabilities: How much processing power, graphics capability, and memory resources are available.
Metaphorically, thick versus thin is like Americans. Generation by generation, we get taller and heavier. Photographs of ancestors show smaller, shorter people; your grandmother’s china hardly seems big enough to hold today’s ample servings. You order a cup of coffee in another country, and embarrass yourself with your surprise at its thimble-like capacity (at which point you know for sure that you are an American).
Ditto for set-tops. Motorola’s DCT-1000 was gigantic compared to the various analog units that constituted the set-top landscape, in 1995. Those very boxes, great-grandfathers now, seem tiny by functional comparison to today’s digital boxes.
The next chapter of the thin client will likely coincide with the digital transition, and the development of low-cost, digital-to-analog converters, perhaps the size of a deck of cards, that seamlessly convert the huge installed base of analog TVs and VCRs into units that can display a digital picture.
In the 2005 timeframe, those “DTA” (Digital Terminal Adaptor) gizmos were envisioned as having enough resources to offer a guide, digital music, all digital video channels, and (theoretically) OCAP-based access to on-demand titles. As time progressed, the feature set slimmed D:A conversion only, to keep costs down.
In totality, a contemporary cable system, in 2005, contains equivalent capacity for about 5 Gbps of throughput.
In cable’s advanced services mix, digital video signals get compressed, then sent. They’re compressed with MPEG-2, and sent using MPEG-2 transport protocol. In high-speed Internet and voice-over-IP applications, there’s the payload itself, and the transport protocol, which is TCP (transmission control protocol; see TCP/IP.)
Tree-and-branch networks were the norm for cable television systems until the introduction of fiber optic technologies (the “F” in “HFC”). Even so, today’s hybrid-fiber coax systems still plausibly use tree and branch, from the optical node outwards, toward subscribing homes.
The fiber optic lines replace traditional coaxial trunks, and extend to transceiver nodes serving an average of 500 passings (homes). From the node to the homes, coaxial cable extends and branches off to the homes. In older tree-and-branch cable networks, as many as 50 amplifiers, in cascade, worked to get signals to homes. The trouble with this approach is that noise is also boosted along the way, which degrades signal quality.
There are several ways to envision a “trusted domain.” In a sense, the movement toward multi-room DVR — where you start a show in one room, decide you want to watch from another room, pause it, go to the other room, and resume — is a “trusted domain.”
Another example is the digital box with a “spigot” (a digital connector with copy protection), used to pour stored shows onto a portable player (picture a sort of “video iPod.”) Or, consider a PC, connected to broadband Internet, which serves as an intermediary between a piece of downloaded content, and a portable media player.
The “trusted domain” term originated within the NGNA (Next Generation Network Architecture) effort of Comcast, Cox, and Time Warner Cable, in 2003.
T-commerce can happen in at least 2 different ways. First, merchandise can be offered as part of a suite of “first screen” or “walled garden” menu items — picture a shopping tab nestled somewhere between “local weather” and “TV listings.” Clicking on the shopping tab elicits a visual list of things to by, either by category or by merchant. (This is bigger in the U.K. than the U.S., so far.)
Second, merchandise offers can be embedded directly into a television program. This is sometimes called “program synchronous” interactivity, or a “bound app.” A pointer, often called a “trigger,” is embedded into the vertical blanking interval of an analog TV signal, or in a data sidecar of a digital TV signal. As the program with the commerce offer airs, software in the set-top box recognizes any incoming triggers and converts them into a clickable icon.
In both cases, TV-commerce requires a fairly sophisticated set-top box, with at the very least a two-way connection between the headend and the home. In many cases, the two-way path needs to be an Internet Protocol (IP) path, such as a combination set-top/cable modem or DSG (DOCSIS set-top gateway) device.
So far, the biggest success story in TV commerce has little to do with merchandise, or Americans. Its happening “over the pond,” where it’s legal: Gambling. The European “Sky” service, delivered via direct broadcast satellite, said in 2003 that 50% of its interactive revenues came from betting and gambling. So far, that’s still taboo in most of the U.S.
From a telco central office out to neighborhoods, many twisted pairs are bundled together in a sheath. This, coupled with the comparatively light shielding on the actual wires, could present a problem for telco xDSL services, as penetration levels rise.
Unshielded twisted pair wiring (UTP) is also the physical medium for Ethernet networks, and especially 10/100baseT networks. These wires can carry traffic roughly 100 meters; shielded twisted pair has extra metal shielding to protect it from interference.
© 2000-2016 translation-please.com. All Rights Reserved.