OT: internet phones
Brian K. White
brian at aljex.com
Wed Apr 19 20:15:59 PDT 2006
----- Original Message -----
From: "Bill Campbell" <bill at celestial.com>
To: "filePro" <filepro-list at lists.celestial.com>
Sent: Wednesday, April 19, 2006 10:00 PM
Subject: Re: OT: internet phones
> On Wed, Apr 19, 2006, Fairlight wrote:
>>On Wed, Apr 19, 2006 at 05:00:57PM -0700, Enrique Arredondo may or may not
>>have
>>proven themselves an utter git by pronouncing:
>>>
>>> VERIZON RIP, VONAGE Rocks
>>
>>Let me know how that works out for you when a core router goes up and down
>>for the better part of 3 days, and you can't figure out why your phone
>>service doesn't work because, "[My] network connection is up..."....
> ...
>>I'd take a pass on that risk, personally. I can only assume that those
>>taking that kind of leap have never actually dealt with 1st and 2nd peer
>>carriers in any amount of detail regarding outages. Run an ISP for a year
>>and a half--it's an extreme eye opener. Given the way the carriers don't
>>cooperate and communicate on some levels, it's a miracle anything gets
>>passed anywhere.
>
> I'll second Mark on this. I've seen many instances where network outages
> or bottlenecks have made it impossible to get to some sites, or they were
> very slow.
>
> I've seen cases where I couldn't get to a customer sites in Tennessee and
> North Carolina directly, but could if I first ssh'ed into another
> customer's site in Indiana. The first two are on Sprint backbones, and
> the
> links between our provider and Sprint weren't talking at the time.
>
> In another situation, I was installing an ISP system in San Leandro,
> California, and the connection back to our office was approximating a
> 2,400
> baud dialup. I could see the bottleneck using traceroute, but getting
> their backbone provider to do something about was hopeless (a major ILEC
> in
> that part of the world).
>
> There's still far too much of Lily Tomlin's ``I'm the phone company, and I
> don't have to care'' around. It probably won't get any better as the
> number of major backbone providers decreases, and competition diminishes.
As true as all that is, what about simply "Do you really want to rely on ANY
network connection the day you have to call 911?"
My cable goes out all the time for several hours at night because they just
assume fewer people are using the net then and they can do maintenance and
upgrades.
It's been out for days at least 2 or 3 times in the past 6 months.
Maintenance and upgrades are nice and all, and necessary, but Verizon
manages to do it without knocking out my T1's somehow, so presumably it can
be done for cable & dsl too. And presumably it's simply harder & more
expensive.
That said, I haven't had a land line at home in 7 years. Just a cell phone.
For the last 4 or so I've also had a voip station of one type or another
that connects to the phone system at the office and ultimately uses the
office T1 or pots lines.
And the voip station has always been about like the weather. Usually it's
fine, sometimes it's crappy, sometimes it's completely dead.
911 will probably be my cell phone, which does rely on having a good
battery.
But the good battery is under my control, and since it's a Treo650 which I
use for several things both recreational and professional besides a phone, I
don't have to worry about forgetting to keep it charged any more than
forgetting to eat.
Here's a question, I wonder how fast you get connected to 911 via pots vs t1
vs voip vs cell ?
I know it can take several seconds for a cell call to get going sometimes,
and sometimes it fails even when reception is good and I have to retry and
the retry will work.
And once in a while my cell just locks up or reboots adding several more
seconds.
Consumer routers, and voip boxes (or combined units) all lock up/crash once
in a while also.
The day I have to call 911 I do not want to discover I must first get down
to the cellar and power-cycle the network.
There's another point, a plain phone requires only the phone system to be
up. A voip phone requires both the power and the net. More requirelments =
more probability to fail.
Basically it's fine as long as you don't care if the thing works only most
of the time.
Brian K. White -- brian at aljex.com -- http://www.aljex.com/bkw/
+++++[>+++[>+++++>+++++++<<-]<-]>>+.>.+++++.+++++++.-.[>+<---]>++.
filePro BBx Linux SCO FreeBSD #callahans Satriani Filk!
More information about the Filepro-list
mailing list