From: heller@deepsoft.com   
      
   At Sun, 06 Jul 2014 11:38:05 -0500 Stephen Sprunk  wrote:   
      
   >   
   > On 06-Jul-14 10:11, Robert Heller wrote:   
   > > Actually it is already possible to push 10s of Gb/s through fiber   
   > > with current technology, using multiple wavelengths.   
   >   
   > Indeed, but how many people need that inside their home or office?   
   >   
   > That is mainly useful for long-haul circuits that already have been   
   > replaced with fiber, but it's not an argument for FTTH over FTTN.   
   >   
   > > Also, there is the matter of attenuation / signal loss over   
   > > distance. I don't think it is possible to run miles and miles and   
   > > miles of 1Gb/s *copper* Ethernet...   
   >   
   > You could, if you were willing to put up with repeaters every few   
   > hundred feet.  Obviously, that's not very practical.   
      
   Right. If you have 50+ miles of road in a town, 1Gb/s *copper* Ethernet is   
   just not going to cut it. Fiber OTOH, has no trouble going miles and miles...   
   The town of Wendell has over 50 miles of road and most of the population gets   
   its phone service out of a concentrator in the center of town. The is   
   *probably* fiber to that concentrator from the CO. But there isn't a DSLAM   
   (and Verizon is NOT going to install one), so ALL the residents can get is   
   bare POTS. And the 50 or so miles of copper is basically crap. Most of it is   
   worn out and overdue for replacement (Verizon is NOT going to replace it, per   
   corp. policy). Even if someone (not Verizon) does install a DSLAM in the   
   center of town, most residents would be lucky to get low-end DSL: like   
   768Kbits down 256Kbits up. Either the *existing* copper is replaced with new   
   copper or with fiber. Given that it *ALL* *has* to be replaced, which makes   
   more sense?  Is it *really* more cost effective to install what is esentually   
   *19th century* technology (simple twisted pair copper, meant for analog audio)   
   vs. 21st centure fiber?  Oh, many of the existing copper drops are probably is   
   just as bad condition as the main cables along the roads.   
      
   Bottom line: while a *copper* infrastructure is probably 'good enough' for   
   current services, replacing old copper with new copper is not going to be   
   cheaper than replacing the old copper with new fiber, given that the old   
   copper is not really usable as is anyway.  I would agree that FTTN is probably   
   more cost effective *when the existing copper* is useable, but when there is   
   either no pre-existing copper or if the pre-existing copper is unusable (eg   
   falling apart), it probably makes better sense to go with FTTH.   
      
   >   
   > S   
   >   
      
   --   
   Robert Heller             -- 978-544-6933   
   Deepwoods Software        -- Custom Software Services   
   http://www.deepsoft.com/  -- Linux Administration Services   
   heller@deepsoft.com       -- Webhosting Services   
      
   --- SoupGate/W32 v1.03   
    * Origin: LiveWire BBS -=*=- UseNet FTN Gateway (1:2320/1)