Silicon Investor (SI) -- The First Internet Community

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Technology Stocks : AUTOHOME, Inc -- Ignore unavailable to you. Want to Upgrade?

To: Frank A. Coluccio who wrote (10869)	6/9/1999 3:10:00 PM
From: ahhaha	Read Replies (5) \| Respond to of 29970

Primer on cable local loop and accommodation of many ISPs Your comments are also addressing what Greg is looking for, a way to accommodate many ISPs on a network originally designed to carry one MSO's offerings. Let's review the basics. The drop which is the connection between the telephone pole supported black inch thick cable tv cable can carry about 100 6 MHz analog tv channels and 500 6 Mhz digital channels. We could say that any of those channels could carry a cable modem broadband signal. Depending on technology maybe we need two of them, one for upstream, one for down stream. Downstream is reception of signal. Upstream is response like clicking a link and getting a response from the remote site computer. The reaction from the remote computer comes back through the downstream channel. Both of these streams can be put on one channel. The home drop is wired in copper coax which is ok for short distances. The home drop is connected to the pole hung cable and that cable runs around the neighborhood covering up to 500 houses in the usual model. We refer to this as the local loop. It has to be wired in HFC which is a optical fiber hybrid. The HFC is needed to carry the many signals added to the cable from each home in the upstream direction. This wasn't a problem when all the signals were downstream only to people's tv sets. It is a problem now with the two way nature of 'Net. The old copper coax in the loop can't cope with this much flow, so it has to be replaced by HFC. Where does the local loop go? It goes to the cable headend. This is the nodal point for distribution of the local loop. It is a building with computers in it. There are wires coming in from the local loops and a big wire going out. The big wire, or cable, or trunk, travels to other buildings which house more computers. This building is a data center and it receives trunks rather than loops. The trunks go in and backbones come out. The data center level is sometimes referred to as metro distribution. The data center 'bones travel to other buildings further away, say hundreds of miles, whose function is to integrate and manage the various bones coming in and going out. The big 'bones are referred to as national backbones and they are monitored and managed at network operation centers. We aren't so interested in the trunk and bone aspect of the network in ATHM, even though this is where the company's technology gets most of its hardware added value. We are interested in the headend and the local loop in discussions about supporting multiple services and possibly multiple ISPs. The MSO owns and controls the local loop and much of the equipment in the headend, so it is TCI or T who is in the hot seat when multiple or open access is the issue. Thus, it is the MSOs who have the responsibility for the "last mile". The question which is under discussion is whether the existing local loop model will support multiple ISPs. Is there enough room in the portion of the spectrum allocated or designated by the FCC to accommodate say 10 major ISPs? The answer is yes. How about 1000? The answer is no, at least not under the current default technology of all MSOs. It is easy to see that if you have hundreds of 6 Mhz channels available in the current spectrum organization that some of them could be reserved for ISP usage. We wouldn't want to squeeze tv too much since it is as important as telegraph was. Are there technologies being developed to handle many ISPs without significant spectrum reorganization? Yes. Some are expensive and some are cheap and some don't work. What has to be done to accommodate 10 ISPs? The headend must be equipped by each ISP and the spectrum must undergo a slight reorganization. There are other ways of doing it too. The MSO could sponsor different ISPs under a provisioning agreement and ISPs would have to make proportionate investments to equip the up scaling of the headend so whereas they wouldn't need differentiated equipment, they would need upgrading of caching server computers and their capacities. Other supporting equipment would also need to be installed. This latter model is the most likely direction things will head. How does this impact ATHM? ATHM has the model of distribution and experience in handling all of this that the copper based ISPs and MSOs don't have. The MSOs know the headend and the local loop. ATHM knows the headend upward throughout the world. Other ISPs will have to link in, or get on ATHM's network. If they try to come in at the trunk level into the headend, it will be expensive. Maybe Sprint can do this, but ELNK can't. Each copper ISP will have to either gain access to the headend through ATHM's existing network or build it. Please make appropriate adjustments to this primer.