[93842] in North American Network Operators' Group
Re: Undersea fiber cut after Taiwan earthquake - PCCW / Singtel /
daemon@ATHENA.MIT.EDU (Frank Coluccio)
Thu Dec 28 19:09:52 2006
From: Frank Coluccio <frank@dticonsulting.com>
To: frank@dticonsulting.com, Joel Jaeggli <joelja@uoregon.edu>
Reply-To: frank@dticonsulting.com
Date: Thu, 28 Dec 2006 18:08:59 -0600
Cc: nanog-post@rsuc.gweep.net, nanog@merit.edu
Errors-To: owner-nanog@merit.edu
Joel Jaeggli wrote:
>The relevant charts and or current navigation software have the cables
>well marked because mariners have an obligation under several
>international treaties (going back to 1884) not to hit them...
All very good points. Thanks for pointing this out. Just so we're clear, wh=
ile I
was passing along information provided by a third party I wasn't suggesting
whether obfuscating, or attempting to hide, or make public, location inform=
ation
posed a problem or had any merit one way or the other. You've presented som=
ewhat
of a Gordian knot here, it would appear.=20
Consider, while it's extremely difficult for a customer to obtain informati=
on
concerning his or her own overland fiber routes without the signing away of=
their
first born, here we have, ironically, a situation where underwater cables o=
f far
greater import are clearly marked on charts and made freely available throu=
gh
software under the international treaty terms. Very interesting, but as you
suggest, a condition that's existed for a very long time.
Where continental shelves come into play, this problem has been partially
mitigated since 1970, when TAT-5 between R.I. and Spain was implemented usi=
ng the
first "sea plow" to bury the cable several feet down off Rhode Island. That
having been said, not all shorelines sit on top of continental shelves, and=
even
where trenching is used on a shelf, it, too, poses its own perils during re=
pairs,
when dangling cable ends are lost in a mire of mud clouds that take hours to
settle after divers use water pressure to find them while kicking up sedime=
nt on
the seabed.=20
In other situations some operators resort to using radio communications and
sometimes small craft to ward off ships from cabling lanes. Years ago piper=
cubs
were used to buzz encroaching ships and drop tons of leaflets on them warni=
ng
them to stay away. Today they'd probably get shot down.
Right about here I'd expect a visit from Sean Gorman, who has had his own s=
hare
of grief to deal with in this respect while plotting the nation's overland =
fiber
routes. In fact, the title of his book is tres apropos to the problem you
highlighted:
Networks, Security And Complexity: The Role of Public Policy in Critical
Infrastructure Protection by Sean P. Gorman (Hardcover - Sep 5, 2005)
http://tinyurl.com/y8za2t
Frank=20
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
On Thu Dec 28 18:18 , Joel Jaeggli sent:
>Frank Coluccio wrote:
>>=20
>> Kidding aside, these "errors" are actually intentional, and the publishe=
r makes
>> no bones about it at the bottom of the page. See disclaimer under the So=
uth
>> Atlantic Ocean:
>>=20
>> "Cable Routes do not represent all subsea cable networks and do not refl=
ect
>> actual location of cables"
>
>The relevant charts and or current navigation software have the cables
>well marked because mariners have an obligation under several
>international treaties (going back to 1884) not to hit them... If you
>have the tools to go on a "fishing trip" you have the tools to find the
>cable. If you obfuscate the location of cables I can plead ignorance
>when I drag it up with my achor.
>
>http://mapserver.maptech.com/mapserver/nautical_symbols/L4.html
>
>Like with back-hoeing through fiber, if you think hitting a submarine
>cable is bad there's plenty other stuff out there that has potentially
>disastrous consequences, gas lines, oil lines, well heads, high voltage
>power lines, and of course lots of other things that fall into the
>category of navigational hazards.
>
>joelja
>--=20
>------------------------------------------------------------------------
>Joel Jaeggli Unix Consulting joelja@uoregon.edu
>GPG Key Fingerprint: 5C6E 0104 BAF0 40B0 5BD3 C38B F000 35AB B67F 56B2
