Security of public network?

[gaf-mNDKBlG2WHs at public.gmane.org (Jerry Feldman)]

I think more in direct answer to Scott's question, I think that POTS is
probably easier from a technical point of view because it is a simple
copper connection. Anyone with a portable phone can very easily connect
in a lot of different places. Fibre requires more skill and equipment.
That said, I want to comment on Matt's comment about Comcast Voice. I
converted to Comcast Digital Phone a number of years ago. This was a
copper connection to the pole. It was cheaper and more reliable than
Verizon in my neighborhood at the time. When Comcast offered Digital
Voice (VOIP) I switched because it was considerably cheaper. The line
has been clear and reliable. The only problem in the past few years was
that we had a cable outage a few weeks ago, but that was cleared up in
an hour or so. My friend also switched to Comcast, and is very happy
with the service. My mother's condo buiding has FIOS, and other than an
initial problem her Verizon service has been excellent. Note that
Comcast's Digital Voice does not go through the Internet, it is
connected to the phone system at the Comcast office, so there should be
no propagation delay caused by the Internet.

On 06/30/2009 09:06 AM, Matthew Gillen wrote:
> Scott Ehrlich wrote:
>  =20
>> Of the various "Landline" phone methods (though there are likely other=
s):
>>
>> - fiber (i.e. FIOS)
>> - POTS (copper)
>> - VOIP (vonage)
>>
>> Do they have equal weight when it comes to security of residential
>> communication, and the customer can boil it down to price?
>>    =20
>
> Depends what kind of security you're talking about.  None of those use
> encryption, and all can be "wiretapped" by various means.  For someone =
who is
> not law-enforcement, there may be different severity of penalties for s=
omeone
> illegally snooping on the different classes of wire (supposing they get=
 caught).
>
> The CALEA law (and subsequent ammendments to it) updated the Wiretap Ac=
t and
> the ECPA to include the non-POTS options on your list.  So from a
> law-enforcement point of view, they should be equivalent: if the police=
 tap
> any of those without a warrant, any evidence gathered will (should?) be=

> inadmissible in court.  There's an interesting discussion on wikipedia:=

> http://en.wikipedia.org/wiki/Lawful_interception
>
> This also has some good info:
> "http://cyber.law.harvard.edu/privacy/Statutory Summaries for Module IV=
=2Ehtm"
>
> To the last part of your question ("[can] the customer can boil it down=
 to
> price?"), you also have to consider link quality.  I've used all three =
of
> those, plus Comcast's brand of VOIP (note that Verizon patents mean tha=
t
> vonage and comcast avoid the term "VOIP" like the plague).  I've got FI=
OS now,
> and the Verizon box connected to my house telephone wiring doesn't seem=
 to
> have enough juice: I can't have two phones active at once in the house =
(i.e.
> multiple people sharing a line so that both my wife and I can talk to m=
y
> parents at the same time).
>
> Comcast's internet-phone stunk.  It would often cut off the first word =
you
> said, so if you were trying to answer questions with one-word answers, =
it made
> for a difficult conversation.
>
> I used Vonage (over Comcast) for a while, and that wasn't too bad.  A l=
ittle
> bit of the same issue that Comcast's phone service had, but not as bad =
(it
> only happened occasionally, versus every single call for Comcast).  Che=
aper too.
>
> Note that some of the bandwidth shaping technologies that Comcast wants=
 to use
> would really kill vonage, because of the specific way they implement ba=
ndwidth
> limiting (the latency increases would be disasterous for third-party
> phone-over-internet providers like vonage).  I stopped paying attention=
 to
> what Comcast was doing as soon as I could dump them for FIOS, so I'm no=
t sure
> how that situation has panned out (or how it's been evolving).
>  =20

--=20
Jerry Feldman <gaf-mNDKBlG2WHs at public.gmane.org>
Boston Linux and Unix
PGP key id: 537C5846
PGP Key fingerprint: 3D1B 8377 A3C0 A5F2 ECBB  CA3B 4607 4319 537C 5846