Re: 10 years from now... (was: internet futures)

[Stephen Frost]

Greetings,

* Javier J (jav...@advancedmachines.us) wrote:
> Since FiOS still doesn't do ipv6 (I don't bother checking anymore) I've
> used tunnelbroker since I was stuck on Comcast.

Called last week and, no, FiOS *still* doesn't do ipv6.  Seriously
ridiculous.

> I'm not running BGP since that's overkill for my home lab needs. just a
> tunnel with the /64 they give you and an addition /48.

Yeah, same, and fully get that renumbering will be annoying if FiOS ever
actually does ipv6 but I suspect it'd be worth it.  The HE.Net tunnel is
pretty good but there's still sometimes it goes out even though my FiOS
v4 is working fine. :/

> If I have to renumber, there are maybe just 4-5 places where an ipv6 is
> manually set. I'll just setup a new tunnel and change the router
> advertisement settings.

I'd probably just overlay the new addresses and then update DNS and wait
a couple of weeks, wouldn't really be that much of an issue.  If HE
actually takes away the tunnel, that'd be unfortunate but I'd just
remove the v6 addresses from DNS, most stuff should fall back to v4 just
fine.

Thanks,

Stephen


signature.asc
Description: PGP signature

Re: 10 years from now... (was: internet futures)

[Javier J]

Since FiOS still doesn't do ipv6 (I don't bother checking anymore) I've
used tunnelbroker since I was stuck on Comcast.
I'm not running BGP since that's overkill for my home lab needs. just a
tunnel with the /64 they give you and an addition /48.

If I have to renumber, there are maybe just 4-5 places where an ipv6 is
manually set. I'll just setup a new tunnel and change the router
advertisement settings.

- J

On Tue, Mar 30, 2021 at 4:53 AM  wrote:

> So, I assume you have PI IPv6 space and doing BGP with HE?
> In other case, if anything will happen to HE (they close they
> tunnelbroker service) you will have to renumber.
>
>
> -- Original message --
>
> From: Javier J 
> To: b...@uu3.net
> Cc: nanog 
> Subject: Re: 10 years from now... (was: internet futures)
> Date: Mon, 29 Mar 2021 13:57:20 -0400
>
> I've had an IPV6 tunnel from Hurricane Electric for 10+ years I think.
> IPv4 will probably live as it does now in my network, mostly for management
> / interserver coms for legacy hardware/software that doesn't support ipv6.
>
>
> On Fri, Mar 26, 2021 at 5:31 PM  wrote:
>
> > Oh, sorry to disappoint you, but they are not missing anything..
> > Internet become a consumer product where data is provided by
> > large corporations similary to TV now. Your avarage Joe consumer
> > does NOT care about NAT and that he cant run services or he does NOT
> > have full e2e communication.
> >
> > Yes, you are right, NAT was a second class internet for a while but
> > now it seems that we cannot live without it anymore :)
> > I dont really see other way how I can connect LAN to internet now.
> > Using public IPs? Thats so terrible idea. How can I be el-cheappo
> > dual-homed then?
> >
> >
> > -- Original message ------
> >
> > From: Mark Andrews 
> > To: Andy Ringsmuth 
> > Cc: Grant Taylor via NANOG 
> > Subject: Re: 10 years from now... (was: internet futures)
> > Date: Sat, 27 Mar 2021 08:00:38 +1100
> >
> > There are more smart phones in use in the world today the world than can
> > be
> > addressed by IPv4. Complaining about lack of IPv6 deployment has been
> > legitimate for a long time. Telcos shouldn˙˙t have to deploy NATs. Homes
> > shouldn˙˙t have to deploy NATs. Businesses shouldn˙˙t have to deploy
> NATs.
> >
> > NATs produce a second class Internet.  We have had to lived with a second
> > class Internet for so long that most don˙˙t know what they are missing.
> --
> > Mark Andrews
> >
>

Re: 10 years from now... (was: internet futures)

[borg]

So, I assume you have PI IPv6 space and doing BGP with HE?
In other case, if anything will happen to HE (they close they
tunnelbroker service) you will have to renumber.


-- Original message --

From: Javier J 
To: b...@uu3.net
Cc: nanog 
Subject: Re: 10 years from now... (was: internet futures)
Date: Mon, 29 Mar 2021 13:57:20 -0400

I've had an IPV6 tunnel from Hurricane Electric for 10+ years I think.
IPv4 will probably live as it does now in my network, mostly for management
/ interserver coms for legacy hardware/software that doesn't support ipv6.


On Fri, Mar 26, 2021 at 5:31 PM  wrote:

> Oh, sorry to disappoint you, but they are not missing anything..
> Internet become a consumer product where data is provided by
> large corporations similary to TV now. Your avarage Joe consumer
> does NOT care about NAT and that he cant run services or he does NOT
> have full e2e communication.
>
> Yes, you are right, NAT was a second class internet for a while but
> now it seems that we cannot live without it anymore :)
> I dont really see other way how I can connect LAN to internet now.
> Using public IPs? Thats so terrible idea. How can I be el-cheappo
> dual-homed then?
>
>
> -- Original message --
>
> From: Mark Andrews 
> To: Andy Ringsmuth 
> Cc: Grant Taylor via NANOG 
> Subject: Re: 10 years from now... (was: internet futures)
> Date: Sat, 27 Mar 2021 08:00:38 +1100
>
> There are more smart phones in use in the world today the world than can
> be
> addressed by IPv4. Complaining about lack of IPv6 deployment has been
> legitimate for a long time. Telcos shouldn˙˙t have to deploy NATs. Homes
> shouldn˙˙t have to deploy NATs. Businesses shouldn˙˙t have to deploy NATs.
>
> NATs produce a second class Internet.  We have had to lived with a second
> class Internet for so long that most don˙˙t know what they are missing. --
> Mark Andrews
>

Re: 10 years from now... (was: internet futures)

[Eric Kuhnke]

I am doing this right now. A starlink CPE is a fairly ordinary DIA link
that exists in cgnat space from the perspective of whatever router you plug
into it. The starlink indoor 'router' is optional.

Whatever you plug into the high power PoE injector will be given a DHCP
lease and a default route out to the world. People who don't want to DIY a
dual-WAN solution with their own Linux box or pfsense or similar can use
things like peplink routers.

In difficult to reach places in the world I can see starlink with a
low-bandwidth traditional VSAT link as backup/failover as a fairly common
configuration. Or starlink as primary and local HSPA+/HSDPA/LTE (whatever
3GPP related) as a failover.



On Mon, Mar 29, 2021 at 12:37 PM Matt Erculiani 
wrote:

> I wouldn't be the least bit surprised if anyone out there was trying to
> mix their StarLink kit and existing broadband service to optimize
> performance and/or add redundancy though.
>
> The underlying technologies will change, but what people try to do with
> them will remain relatively unchanged.
>
> Back 20 years ago people were talking about their Frame Relay P2P
> services, now they talk about their Ethernet P2P services.
>
> -Matt
>
> On Mon, Mar 29, 2021 at 1:10 PM Aaron C. de Bruyn 
> wrote:
>
>> On Mon, Mar 29, 2021 at 11:39 AM Matt Erculiani 
>> wrote:
>>
>>> I think the best way to think about what 10 years from now will look
>>> like is to compare 10 years ago to the present:
>>> https://mailman.nanog.org/pipermail/nanog/2011-April/thread.html
>>>
>>
>> Multi-homing your DSL connection?
>> I can't wait to multi-home my 10x10 array of StarLink satellites in a few
>> years...
>>
>> -A
>>
>
>
> --
> Matt Erculiani
> ERCUL-ARIN
>

Re: 10 years from now... (was: internet futures)

[Crist Clark]

It’s not like Starlink is anything brand new. Iridium and Globalstar both
do Internet from LEO. It wasn’t their primary service, voice was/is, but
they could do it in a half-a** manner.

Starlink isn’t going to become big in China without bowing to the GFW
‘cause how do you bill for it if you don’t meet the local authorities’
rules?

On Mon, Mar 29, 2021 at 1:44 PM Nathan Stratton  wrote:

> I mix Starlink and Comcast over two openvpn tunnels to my datacenter in
> Ashburn.
>
> ><>
>
> nathan stratton
>
>
> On Mon, Mar 29, 2021 at 3:38 PM Matt Erculiani 
> wrote:
>
>> I wouldn't be the least bit surprised if anyone out there was trying to
>> mix their StarLink kit and existing broadband service to optimize
>> performance and/or add redundancy though.
>>
>> The underlying technologies will change, but what people try to do with
>> them will remain relatively unchanged.
>>
>> Back 20 years ago people were talking about their Frame Relay P2P
>> services, now they talk about their Ethernet P2P services.
>>
>> -Matt
>>
>> On Mon, Mar 29, 2021 at 1:10 PM Aaron C. de Bruyn 
>> wrote:
>>
>>> On Mon, Mar 29, 2021 at 11:39 AM Matt Erculiani 
>>> wrote:
>>>
 I think the best way to think about what 10 years from now will look
 like is to compare 10 years ago to the present:
 https://mailman.nanog.org/pipermail/nanog/2011-April/thread.html

>>>
>>> Multi-homing your DSL connection?
>>> I can't wait to multi-home my 10x10 array of StarLink satellites in a
>>> few years...
>>>
>>> -A
>>>
>>
>>
>> --
>> Matt Erculiani
>> ERCUL-ARIN
>>
>

Re: 10 years from now... (was: internet futures)

[Nathan Stratton]

I mix Starlink and Comcast over two openvpn tunnels to my datacenter in
Ashburn.

><>
nathan stratton


On Mon, Mar 29, 2021 at 3:38 PM Matt Erculiani  wrote:

> I wouldn't be the least bit surprised if anyone out there was trying to
> mix their StarLink kit and existing broadband service to optimize
> performance and/or add redundancy though.
>
> The underlying technologies will change, but what people try to do with
> them will remain relatively unchanged.
>
> Back 20 years ago people were talking about their Frame Relay P2P
> services, now they talk about their Ethernet P2P services.
>
> -Matt
>
> On Mon, Mar 29, 2021 at 1:10 PM Aaron C. de Bruyn 
> wrote:
>
>> On Mon, Mar 29, 2021 at 11:39 AM Matt Erculiani 
>> wrote:
>>
>>> I think the best way to think about what 10 years from now will look
>>> like is to compare 10 years ago to the present:
>>> https://mailman.nanog.org/pipermail/nanog/2011-April/thread.html
>>>
>>
>> Multi-homing your DSL connection?
>> I can't wait to multi-home my 10x10 array of StarLink satellites in a few
>> years...
>>
>> -A
>>
>
>
> --
> Matt Erculiani
> ERCUL-ARIN
>

Re: 10 years from now... (was: internet futures)

[Matt Erculiani]

I wouldn't be the least bit surprised if anyone out there was trying to mix
their StarLink kit and existing broadband service to optimize performance
and/or add redundancy though.

The underlying technologies will change, but what people try to do with
them will remain relatively unchanged.

Back 20 years ago people were talking about their Frame Relay P2P services,
now they talk about their Ethernet P2P services.

-Matt

On Mon, Mar 29, 2021 at 1:10 PM Aaron C. de Bruyn 
wrote:

> On Mon, Mar 29, 2021 at 11:39 AM Matt Erculiani 
> wrote:
>
>> I think the best way to think about what 10 years from now will look like
>> is to compare 10 years ago to the present:
>> https://mailman.nanog.org/pipermail/nanog/2011-April/thread.html
>>
>
> Multi-homing your DSL connection?
> I can't wait to multi-home my 10x10 array of StarLink satellites in a few
> years...
>
> -A
>


-- 
Matt Erculiani
ERCUL-ARIN

Re: 10 years from now... (was: internet futures)

[Aaron C. de Bruyn via NANOG]

On Mon, Mar 29, 2021 at 11:39 AM Matt Erculiani 
wrote:

> I think the best way to think about what 10 years from now will look like
> is to compare 10 years ago to the present:
> https://mailman.nanog.org/pipermail/nanog/2011-April/thread.html
>

Multi-homing your DSL connection?
I can't wait to multi-home my 10x10 array of StarLink satellites in a few
years...

-A

Re: 10 years from now... (was: internet futures)

[Michael Thomas]

On 3/29/21 11:36 AM, Matt Erculiani wrote:



We might be talking a lot more about PRKI as it becomes compulsory, 
maybe 400G transit links will start being standard across the 
industry. If we're lucky (or unlucky, depending on how you look at it) 
maybe a whole new routing protocol will be introduced and rapidly gain 
popularity.


One interesting observation is that QUIC has the potential to open the 
floodgates for new purpose built transport protocols for things other 
than http that have their own requirements. It also shows that it can 
navigate the problem of pleading kernel code and firewalls that block 
unknown (it it) IP protocol numbers. It's my guess that those were what 
really sunk SCTP. Another thing that is coming up is that with 
increasingly high bandwidth, the TCP checksum is showing its age and 
we'd probably like to leverage crypto-grade hashes instead of being at 
the mercy of a 40 year old algorithm.


Mike

Re: 10 years from now... (was: internet futures)

[Matt Erculiani]

I think the best way to think about what 10 years from now will look like
is to compare 10 years ago to the present:
https://mailman.nanog.org/pipermail/nanog/2011-April/thread.html

- BGP issues/hijacks: for sure, despite everyone's best efforts
- Jokes and fun stories: I sure hope so!
- Current events (and their impact on NetOps): Yep...
- Request for support form ISPs with a complete lack of quality frontline
support: 100% guaranteed
- Questions about routing standards and associated practices (RRI, RPKI,
etc.): Also for sure
- DNS problems: It's always DNS, so yes
- Vendor recommendation for ABC: Yes, new technologies means new players,
old players either keep up or get replaced.
- Outages of network providers and popular online services: There will
ALWAYS be outages, and we'll always break them down piece by piece for the
operators to confirm, deny, or never comment.
- Popular vendor doing something bad and the community reacts: Also 100%
guaranteed
- Big acquisitions: As long as money exits, things will be bought and sold,
including companies, big and small.
- Staffing questions: Might decrease as automation goes up, but these still
pop up
- How do I get IPv4: Beaten to death already; yes, will likely still be a
thing (and associated discussions about IPv6)
- What device should I use for XYZ: Until moore's law hits a plateau, we'll
always need better gear

You know, posts from 10 years ago, (and 20 years ago for that matter
) don't
seem a whole lot different than they are today with the obvious exception
of the underlying technologies.

Tech that only exists in university labs today will start to see enterprise
applications, smaller outfits will start to standardize technology that
used to only be available to the big fish, and there will be all sorts of
new hotness we're excited about on the horizon, just like 10 years ago...

We might be talking a lot more about PRKI as it becomes compulsory, maybe
400G transit links will start being standard across the industry. If we're
lucky (or unlucky, depending on how you look at it) maybe a whole new
routing protocol will be introduced and rapidly gain popularity.

I've joined the group within this past 10 years, and I sure am looking
forward to 10 more years of learning, constructive discussion, and
entertainment, hopefully in that order of occurrence.

Here's to 10 more, ya bunch of nerds,
-Matt


On Fri, Mar 26, 2021 at 1:42 PM Michael Thomas  wrote:

>
> On 3/26/21 12:26 PM, Mark Tinka wrote:
>
> If the last decade is anything to go by, I'm keen to see what the next one
> brings.
>
> Mark.
>
>
> So the obvious question is what will happen to the internet 10 years from
> now. The last 10 years were all about phones and apps, but that's pretty
> well played out by now. Gratuitously networked devices like my dishwasher
> will probably be common, but that's hardly exciting. LEO internet providers
> will be coming online which might make a difference in the corners of the
> world where it's hard to get access, but will it allow internet access to
> parachute in behind the Great Firewall?
>
> One thing that we are seeing a revolution in is with working from home.
> That has some implications for networking since symmetric bandwidth, or at
> least quite a bit more upstream would be helpful as many people found out.
> Is latency going to drive networking, given gaming? Gamers are not just
> zitty 15 year olds, they are middle aged or older nowadays.
>
> Mike
>


-- 
Matt Erculiani
ERCUL-ARIN

Re: 10 years from now... (was: internet futures)

[Javier J]

I've had an IPV6 tunnel from Hurricane Electric for 10+ years I think.
IPv4 will probably live as it does now in my network, mostly for management
/ interserver coms for legacy hardware/software that doesn't support ipv6.


On Fri, Mar 26, 2021 at 5:31 PM  wrote:

> Oh, sorry to disappoint you, but they are not missing anything..
> Internet become a consumer product where data is provided by
> large corporations similary to TV now. Your avarage Joe consumer
> does NOT care about NAT and that he cant run services or he does NOT
> have full e2e communication.
>
> Yes, you are right, NAT was a second class internet for a while but
> now it seems that we cannot live without it anymore :)
> I dont really see other way how I can connect LAN to internet now.
> Using public IPs? Thats so terrible idea. How can I be el-cheappo
> dual-homed then?
>
>
> -- Original message --
>
> From: Mark Andrews 
> To: Andy Ringsmuth 
> Cc: Grant Taylor via NANOG 
> Subject: Re: 10 years from now... (was: internet futures)
> Date: Sat, 27 Mar 2021 08:00:38 +1100
>
> There are more smart phones in use in the world today the world than can
> be
> addressed by IPv4. Complaining about lack of IPv6 deployment has been
> legitimate for a long time. Telcos shouldn˙˙t have to deploy NATs. Homes
> shouldn˙˙t have to deploy NATs. Businesses shouldn˙˙t have to deploy NATs.
>
> NATs produce a second class Internet.  We have had to lived with a second
> class Internet for so long that most don˙˙t know what they are missing. --
> Mark Andrews
>

Re: 10 years from now... (was: internet futures)

[Mark Tinka]

On 3/29/21 07:21, Eric Kuhnke wrote:

The US State Department is already a large customer for dedicated 
transponder capacity, in C-band hemispheric and Ku beams in some weird 
places in the world.


As a randomly chosen example if you take a look at the roof of the UK 
embassy in Kabul, there's a nice 4 meter size Andrew/Commscope compact 
cassegrain dish up there. Pretty typical thing already for embassies, 
the big difference would be that that they'll have more market options 
for high-throughput service.


Exactly... those would be the kinds of places grumpy dudes with guns 
can't just show up. However, that does not prevent the local gubbermint 
from jamming signals to the extent they can without infringing on the 
sovereign rights of the consulate.


Mark.

Re: 10 years from now... (was: internet futures)

[Eric Kuhnke]

The US State Department is already a large customer for dedicated
transponder capacity, in C-band hemispheric and Ku beams in some weird
places in the world.

As a randomly chosen example if you take a look at the roof of the UK
embassy in Kabul, there's a nice 4 meter size Andrew/Commscope compact
cassegrain dish up there. Pretty typical thing already for embassies, the
big difference would be that that they'll have more market options for
high-throughput service.


On Sun, Mar 28, 2021 at 10:18 PM Mark Tinka  wrote:

>
>
> On 3/29/21 02:23, Eric Kuhnke wrote:
>
> >
> > I am not saying it is an impossible problem to solve, but any system
> > intended for that sort of purpose would have to be designed for
> > circumvention, and not a consumer/COTS adaptation of an off the shelf
> > starlink terminal.
>
> Behind the walls of an embassy, perhaps :-).
>
> Mark.
>

Re: 10 years from now... (was: internet futures)

[Eric Kuhnke]

By definition and orbital mechanics, low earth orbit things don't "park"
anywhere. There's an equal number of starlink satellites over Mongolia
right now as there are over the same latitude locations in the US and
Canada.

https://satellitemap.space/

This also becomes intuitive once one plays Kerbal Space Program for a few
hours...




On Sun, Mar 28, 2021 at 9:16 PM Keith Medcalf  wrote:

>
> Net to mention, of course, that the Low Orbit constellation would need to
> be "parked" over China (or where-ever you want to access it).  I am quite
> sure that "shooting down" such low orbit stationary vehicles would not be
> too difficult.  And if they are owned by an adversary who has no permission
> to fly those objects in your airspace, I doubt that anything could be done
> about it.
>
> If I owned a bunch of low orbit satellites costing millions of dollars
> each, I would not want to "park" them in low orbit over a hostile territory.
>
> Then you also have the requirement to maintain positive control over the
> satellites which, unlike those in geostationary orbits, need to be under
> continual thrust and control in order to stay "parked".  I doubt that any
> "private" (ie, non-Government organization) could afford to do so without
> the cooperation of the state over which they are parking.
>
> --
> Be decisive.  Make a decision, right or wrong.  The road of life is paved
> with flat squirrels who could not make a decision.
>
> >-Original Message-
> >From: NANOG  On Behalf Of
> >Eric Kuhnke
> >Sent: Sunday, 28 March, 2021 18:24
> >To: na...@jima.us
> >Cc: nanog@nanog.org
> >Subject: Re: 10 years from now... (was: internet futures)
> >
> >I would also concur that the likelihood of Starlink (or a Oneweb, or
> >Kuiper) terminal being used successfully to bypass the GFW or similar
> >serious Internet censorship, in an authoritarian environment, is probably
> >low. This is because:
> >
> >a) It has to transmit in known bands.
> >
> >
> >b) It has to be located in a location with a very good, clear view of the
> >sky in all directions (even a single tree obstruction in one section of
> >the sky, relative to where the antenna is mounted will cause packet
> >loss/periodic issues on a starlink beta terminal right now). Visually
> >identifying the terminal would not be hard.
> >
> >
> >c) Portable spectrum analyzers capable of up to 30 GHz are not nearly as
> >expensive as they used to be. They also have much better GUIs and
> >visualization tools than what was available 6-10 years ago.
> >
> >
> >d) You could successfully train local law enforcement to use these sort
> >of portable spectrum analyzers in a one-day, 8-hour training course.
> >
> >
> >e) The equipment would have to be smuggled into the country
> >
> >f) Many people such as in a location like Iran may lack access to a
> >standard payment system for the services (the percentage of Iranians with
> >access to buy things online with visa/mastercard/american express or
> >similar is quite low).
> >
> >
> >
> >There are already plenty of places in the world where if you set up a
> >1.2, 1.8 or 2.4 meter C, Ku or Ka band VSAT terminal using some sort of
> >geostationary based services, without appropriate government "licenses",
> >men with guns will come to dismantle it and arrest you.
> >
> >I am not saying it is an impossible problem to solve, but any system
> >intended for that sort of purpose would have to be designed for
> >circumvention, and not a consumer/COTS adaptation of an off the shelf
> >starlink terminal.
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >On Sat, Mar 27, 2021 at 8:31 PM na...@jima.us <mailto:na...@jima.us>
> >mailto:na...@jima.us> > wrote:
> >
> >
> >   Please don't forget that RF sources can be tracked down by even
> >minimally-well-equipped adversaries.
> >
> >   - Jima
> >
> >   -Original Message-
> >   From: NANOG  ><mailto:jima...@nanog.org> > On Behalf Of scott
> >   Sent: Saturday, March 27, 2021 19:36
> >   To: nanog@nanog.org <mailto:nanog@nanog.org>
> >   Subject: Re: 10 years from now... (was: internet futures)
> >
> >
> >   On 3/26/2021 9:42 AM, Michael Thomas wrote:
> >   > LEO internet providers will be coming online which might make a
> >   > difference in the corners of the world where it's hard to get
> >access,
> >   > but will it allow internet access to parachute in behind the
> Great
> >   > Firewall?
> >   
> >   > How do the Chinas of the world intend to deal with the Great
> >Firewall
> >   > implications?
> >
> >
> >   This is what I hope will change in the next 10 years.  "Turning off
> >the
> >   internet" will be harder and harder for folks suppressing others,
> >many
> >   times violently, and hiding it from everyone else.  A small-ish
> >antenna
> >   easily hidden would be necessary.
> >
> >   scott
> >
> >
> >
> >
>
>
>
>
>

Re: 10 years from now... (was: internet futures)

[Mark Tinka]

On 3/29/21 02:23, Eric Kuhnke wrote:



I am not saying it is an impossible problem to solve, but any system 
intended for that sort of purpose would have to be designed for 
circumvention, and not a consumer/COTS adaptation of an off the shelf 
starlink terminal.


Behind the walls of an embassy, perhaps :-).

Mark.

RE: 10 years from now... (was: internet futures)

[Keith Medcalf]

Net to mention, of course, that the Low Orbit constellation would need to be 
"parked" over China (or where-ever you want to access it).  I am quite sure 
that "shooting down" such low orbit stationary vehicles would not be too 
difficult.  And if they are owned by an adversary who has no permission to fly 
those objects in your airspace, I doubt that anything could be done about it.

If I owned a bunch of low orbit satellites costing millions of dollars each, I 
would not want to "park" them in low orbit over a hostile territory.

Then you also have the requirement to maintain positive control over the 
satellites which, unlike those in geostationary orbits, need to be under 
continual thrust and control in order to stay "parked".  I doubt that any 
"private" (ie, non-Government organization) could afford to do so without the 
cooperation of the state over which they are parking.

--
Be decisive.  Make a decision, right or wrong.  The road of life is paved with 
flat squirrels who could not make a decision.

>-Original Message-
>From: NANOG  On Behalf Of
>Eric Kuhnke
>Sent: Sunday, 28 March, 2021 18:24
>To: na...@jima.us
>Cc: nanog@nanog.org
>Subject: Re: 10 years from now... (was: internet futures)
>
>I would also concur that the likelihood of Starlink (or a Oneweb, or
>Kuiper) terminal being used successfully to bypass the GFW or similar
>serious Internet censorship, in an authoritarian environment, is probably
>low. This is because:
>
>a) It has to transmit in known bands.
>
>
>b) It has to be located in a location with a very good, clear view of the
>sky in all directions (even a single tree obstruction in one section of
>the sky, relative to where the antenna is mounted will cause packet
>loss/periodic issues on a starlink beta terminal right now). Visually
>identifying the terminal would not be hard.
>
>
>c) Portable spectrum analyzers capable of up to 30 GHz are not nearly as
>expensive as they used to be. They also have much better GUIs and
>visualization tools than what was available 6-10 years ago.
>
>
>d) You could successfully train local law enforcement to use these sort
>of portable spectrum analyzers in a one-day, 8-hour training course.
>
>
>e) The equipment would have to be smuggled into the country
>
>f) Many people such as in a location like Iran may lack access to a
>standard payment system for the services (the percentage of Iranians with
>access to buy things online with visa/mastercard/american express or
>similar is quite low).
>
>
>
>There are already plenty of places in the world where if you set up a
>1.2, 1.8 or 2.4 meter C, Ku or Ka band VSAT terminal using some sort of
>geostationary based services, without appropriate government "licenses",
>men with guns will come to dismantle it and arrest you.
>
>I am not saying it is an impossible problem to solve, but any system
>intended for that sort of purpose would have to be designed for
>circumvention, and not a consumer/COTS adaptation of an off the shelf
>starlink terminal.
>
>
>
>
>
>
>
>
>
>On Sat, Mar 27, 2021 at 8:31 PM na...@jima.us <mailto:na...@jima.us>
>mailto:na...@jima.us> > wrote:
>
>
>   Please don't forget that RF sources can be tracked down by even
>minimally-well-equipped adversaries.
>
>   - Jima
>
>   -Original Message-
>   From: NANOG <mailto:jima...@nanog.org> > On Behalf Of scott
>   Sent: Saturday, March 27, 2021 19:36
>   To: nanog@nanog.org <mailto:nanog@nanog.org>
>   Subject: Re: 10 years from now... (was: internet futures)
>
>
>   On 3/26/2021 9:42 AM, Michael Thomas wrote:
>   > LEO internet providers will be coming online which might make a
>   > difference in the corners of the world where it's hard to get
>access,
>   > but will it allow internet access to parachute in behind the Great
>   > Firewall?
>   
>   > How do the Chinas of the world intend to deal with the Great
>Firewall
>   > implications?
>
>
>   This is what I hope will change in the next 10 years.  "Turning off
>the
>   internet" will be harder and harder for folks suppressing others,
>many
>   times violently, and hiding it from everyone else.  A small-ish
>antenna
>   easily hidden would be necessary.
>
>   scott
>
>
>
>

Re: 10 years from now... (was: internet futures)

[Eric Kuhnke]

I would also concur that the likelihood of Starlink (or a Oneweb, or
Kuiper) terminal being used successfully to bypass the GFW or similar
serious Internet censorship, in an authoritarian environment, is probably
low. This is because:

a) It has to transmit in known bands.

b) It has to be located in a location with a very good, clear view of the
sky in all directions (even a single tree obstruction in one section of the
sky, relative to where the antenna is mounted will cause packet
loss/periodic issues on a starlink beta terminal right now). Visually
identifying the terminal would not be hard.

c) Portable spectrum analyzers capable of up to 30 GHz are not nearly as
expensive as they used to be. They also have much better GUIs and
visualization tools than what was available 6-10 years ago.

d) You could successfully train local law enforcement to use these sort of
portable spectrum analyzers in a one-day, 8-hour training course.

e) The equipment would have to be smuggled into the country

f) Many people such as in a location like Iran may lack access to a
standard payment system for the services (the percentage of Iranians with
access to buy things online with visa/mastercard/american express or
similar is quite low).


There are already plenty of places in the world where if you set up a 1.2,
1.8 or 2.4 meter C, Ku or Ka band VSAT terminal using some sort of
geostationary based services, without appropriate government "licenses",
men with guns will come to dismantle it and arrest you.

I am not saying it is an impossible problem to solve, but any system
intended for that sort of purpose would have to be designed for
circumvention, and not a consumer/COTS adaptation of an off the shelf
starlink terminal.









On Sat, Mar 27, 2021 at 8:31 PM na...@jima.us  wrote:

> Please don't forget that RF sources can be tracked down by even
> minimally-well-equipped adversaries.
>
> - Jima
>
> -Original Message-
> From: NANOG  On Behalf Of scott
> Sent: Saturday, March 27, 2021 19:36
> To: nanog@nanog.org
> Subject: Re: 10 years from now... (was: internet futures)
>
>
> On 3/26/2021 9:42 AM, Michael Thomas wrote:
> > LEO internet providers will be coming online which might make a
> > difference in the corners of the world where it's hard to get access,
> > but will it allow internet access to parachute in behind the Great
> > Firewall?
> 
> > How do the Chinas of the world intend to deal with the Great Firewall
> > implications?
>
>
> This is what I hope will change in the next 10 years.  "Turning off the
> internet" will be harder and harder for folks suppressing others, many
> times violently, and hiding it from everyone else.  A small-ish antenna
> easily hidden would be necessary.
>
> scott
>
>
>
>

Re: 10 years from now... (was: internet futures)

[scott]

On 3/26/2021 9:42 AM, Michael Thomas wrote:

LEO internet providers will be coming online which might make a
difference in the corners of the world where it's hard to get access,
but will it allow internet access to parachute in behind the Great
Firewall?

How do the Chinas of the world intend to deal with the Great Firewall
implications?

This is what I hope will change in the next 10 years.  "Turning off the
internet" will be harder and harder for folks suppressing others, many
times violently, and hiding it from everyone else.  A small-ish antenna
easily hidden would be necessary.


On 3/27/2021 5:30 PM, na...@jima.us wrote:

Please don't forget that RF sources can be tracked down by even 
minimally-well-equipped adversaries.


Spread spectrum?  ;)

https://en.wikipedia.org/wiki/Spread_spectrum

scott
 

RE: 10 years from now... (was: internet futures)

[na...@jima.us]

Please don't forget that RF sources can be tracked down by even 
minimally-well-equipped adversaries.

- Jima

-Original Message-
From: NANOG  On Behalf Of scott
Sent: Saturday, March 27, 2021 19:36
To: nanog@nanog.org
Subject: Re: 10 years from now... (was: internet futures)


On 3/26/2021 9:42 AM, Michael Thomas wrote:
> LEO internet providers will be coming online which might make a 
> difference in the corners of the world where it's hard to get access, 
> but will it allow internet access to parachute in behind the Great 
> Firewall?

> How do the Chinas of the world intend to deal with the Great Firewall 
> implications?


This is what I hope will change in the next 10 years.  "Turning off the 
internet" will be harder and harder for folks suppressing others, many 
times violently, and hiding it from everyone else.  A small-ish antenna 
easily hidden would be necessary.

scott

Re: 10 years from now... (was: internet futures)

[scott]

On 3/26/2021 9:42 AM, Michael Thomas wrote:
LEO internet providers will be coming online which might make a 
difference in the corners of the world where it's hard to get access, 
but will it allow internet access to parachute in behind the Great 
Firewall?


How do the Chinas of the world intend to deal with the Great Firewall 
implications? 



This is what I hope will change in the next 10 years.  "Turning off the 
internet" will be harder and harder for folks suppressing others, many 
times violently, and hiding it from everyone else.  A small-ish antenna 
easily hidden would be necessary.


scott

Re: 10 years from now... (was: internet futures)

[Michael Thomas]

On 3/27/21 2:50 AM, Valdis Klētnieks wrote:

On Fri, 26 Mar 2021 12:42:20 -0700, Michael Thomas said:


dishwasher will probably be common, but that's hardly exciting. LEO
internet providers will be coming online which might make a difference
in the corners of the world where it's hard to get access, but will it
allow internet access to parachute in behind the Great Firewall?

At which point, we get to see two very different types of LEO engage
in mortal combat



PREEEYYY FIREWORKS!!!

I'm sure somebody's thought about this, but are these LEO networks 
intended to have the downlink at home?  How do the Chinas of the world 
intend to deal with the Great Firewall implications?


Mike

Re: 10 years from now... (was: internet futures)

[Valdis Klētnieks]

On Fri, 26 Mar 2021 12:42:20 -0700, Michael Thomas said:

> dishwasher will probably be common, but that's hardly exciting. LEO
> internet providers will be coming online which might make a difference
> in the corners of the world where it's hard to get access, but will it
> allow internet access to parachute in behind the Great Firewall?

At which point, we get to see two very different types of LEO engage
in mortal combat


pgpa0yUCiLvX5.pgp
Description: PGP signature

Re: 10 years from now... (was: internet futures)

[Michael Thomas]

On 3/26/21 3:31 PM, Mark Tinka wrote:



On 3/26/21 23:30, b...@uu3.net wrote:


Oh, sorry to disappoint you, but they are not missing anything..
Internet become a consumer product where data is provided by
large corporations similary to TV now. Your avarage Joe consumer
does NOT care about NAT and that he cant run services or he does NOT
have full e2e communication.


Yep - infrastructure is now implied, to the extent that customers even 
forget who are they paying for connectivity.




Yes, you are right, NAT was a second class internet for a while but
now it seems that we cannot live without it anymore :)
I dont really see other way how I can connect LAN to internet now.
Using public IPs? Thats so terrible idea. How can I be el-cheappo
dual-homed then?


As long as infrastructure continues to dilly-dally, software will fill 
in the gaps, even if it may cause more breakage in the eyes of the 
networking purists.



I think the question these days is NAT or not. It's double NAT or not.

Mike

Re: 10 years from now... (was: internet futures)

[Michael Thomas]

On 3/26/21 2:00 PM, Mark Andrews wrote:

There are more smart phones in use in the world today the world than can be 
addressed by IPv4. Complaining about lack of IPv6 deployment has been 
legitimate for a long time. Telcos shouldn’t have to deploy NATs. Homes 
shouldn’t have to deploy NATs. Businesses shouldn’t have to deploy NATs.

NATs produce a second class Internet.  We have had to lived with a second class 
Internet for so long that most don’t know what they are missing.


I thought a fair chunk of mobile phones were using ipv6?

Mike

Re: 10 years from now... (was: internet futures)

[Mark Tinka]

On 3/26/21 23:30, b...@uu3.net wrote:


Oh, sorry to disappoint you, but they are not missing anything..
Internet become a consumer product where data is provided by
large corporations similary to TV now. Your avarage Joe consumer
does NOT care about NAT and that he cant run services or he does NOT
have full e2e communication.


Yep - infrastructure is now implied, to the extent that customers even 
forget who are they paying for connectivity.




Yes, you are right, NAT was a second class internet for a while but
now it seems that we cannot live without it anymore :)
I dont really see other way how I can connect LAN to internet now.
Using public IPs? Thats so terrible idea. How can I be el-cheappo
dual-homed then?


As long as infrastructure continues to dilly-dally, software will fill 
in the gaps, even if it may cause more breakage in the eyes of the 
networking purists.


Mark.

Re: 10 years from now... (was: internet futures)

[Mark Tinka]

On 3/26/21 23:00, Mark Andrews wrote:


There are more smart phones in use in the world today the world than can be 
addressed by IPv4. Complaining about lack of IPv6 deployment has been 
legitimate for a long time. Telcos shouldn’t have to deploy NATs. Homes 
shouldn’t have to deploy NATs. Businesses shouldn’t have to deploy NATs.

NATs produce a second class Internet.  We have had to lived with a second class 
Internet for so long that most don’t know what they are missing.


Well, it's simple - apps will continue to route around the inadequacies 
of infrastructure; even more so, over the next 10 years.


Mark.

Re: 10 years from now... (was: internet futures)

[Mark Tinka]

On 3/26/21 22:12, Andy Ringsmuth wrote:


Ten years from now? Easy. We’ll still be talking about the continued shortage 
of IPv4 address space and (legitimately) complaining about why IPv6 still isn’t 
the default addressing/routing methodology for the Internet worldwide.


Thankfully, the users who benefit from the connectivity, somehow, won't 
care about any of those problems.


Mark.

Re: 10 years from now... (was: internet futures)

[Mark Tinka]

On 3/26/21 21:42, Michael Thomas wrote:

So the obvious question is what will happen to the internet 10 years 
from now. The last 10 years were all about phones and apps, but that's 
pretty well played out by now. Gratuitously networked devices like my 
dishwasher will probably be common, but that's hardly exciting. LEO 
internet providers will be coming online which might make a difference 
in the corners of the world where it's hard to get access, but will it 
allow internet access to parachute in behind the Great Firewall?


One thing that we are seeing a revolution in is with working from 
home. That has some implications for networking since symmetric 
bandwidth, or at least quite a bit more upstream would be helpful as 
many people found out. Is latency going to drive networking, given 
gaming? Gamers are not just zitty 15 year olds, they are middle aged 
or older nowadays.




I'm expecting some kind of gubbermint drive in many parts of the world 
(especially the developing world) to get as much free Internet in the 
hands of citizens as they possibly can, largely driven by the effects 
the Coronavirus had on economic productivity last year. This won't go 
down easy, though.


Mark.

Re: 10 years from now... (was: internet futures)

[Robert L Mathews]

On 3/26/21 2:00 PM, Mark Andrews wrote:

> Telcos shouldn’t have to deploy NATs. Homes shouldn’t have to
>deploy NATs. Businesses shouldn’t have to deploy NATs.

But NATs are good: https://youtu.be/v26BAlfWBm8

(Since we're speaking of things from ~10 years ago...!)

-- 
Robert L Mathews, Tiger Technologies

Re: 10 years from now... (was: internet futures)

[borg]

Oh, sorry to disappoint you, but they are not missing anything..
Internet become a consumer product where data is provided by
large corporations similary to TV now. Your avarage Joe consumer
does NOT care about NAT and that he cant run services or he does NOT
have full e2e communication.

Yes, you are right, NAT was a second class internet for a while but
now it seems that we cannot live without it anymore :)
I dont really see other way how I can connect LAN to internet now.
Using public IPs? Thats so terrible idea. How can I be el-cheappo
dual-homed then?


-- Original message --

From: Mark Andrews 
To: Andy Ringsmuth 
Cc: Grant Taylor via NANOG 
Subject: Re: 10 years from now... (was: internet futures)
Date: Sat, 27 Mar 2021 08:00:38 +1100

There are more smart phones in use in the world today the world than can be 
addressed by IPv4. Complaining about lack of IPv6 deployment has been 
legitimate for a long time. Telcos shouldn˙˙t have to deploy NATs. Homes 
shouldn˙˙t have to deploy NATs. Businesses shouldn˙˙t have to deploy NATs.

NATs produce a second class Internet.  We have had to lived with a second 
class Internet for so long that most don˙˙t know what they are missing. -- 
Mark Andrews

Re: 10 years from now... (was: internet futures)

[Mark Andrews]

There are more smart phones in use in the world today the world than can be 
addressed by IPv4. Complaining about lack of IPv6 deployment has been 
legitimate for a long time. Telcos shouldn’t have to deploy NATs. Homes 
shouldn’t have to deploy NATs. Businesses shouldn’t have to deploy NATs. 

NATs produce a second class Internet.  We have had to lived with a second class 
Internet for so long that most don’t know what they are missing. 
-- 
Mark Andrews

> On 27 Mar 2021, at 07:14, Andy Ringsmuth  wrote:
> 
> 
>> 
>>> On 3/26/21 12:26 PM, Mark Tinka wrote:
>>> If the last decade is anything to go by, I'm keen to see what the next one 
>>> brings.
>>> Mark.
>>> 
>> 
>> 
>> So the obvious question is what will happen to the internet 10 years from 
>> now. The last 10 years were all about phones and apps, but that's pretty 
>> well played out by now. Gratuitously networked devices like my dishwasher 
>> will probably be common, but that's hardly exciting. LEO internet providers 
>> will be coming online which might make a difference in the corners of the 
>> world where it's hard to get access, but will it allow internet access to 
>> parachute in behind the Great Firewall?
>> 
>> One thing that we are seeing a revolution in is with working from home. That 
>> has some implications for networking since symmetric bandwidth, or at least 
>> quite a bit more upstream would be helpful as many people found out. Is 
>> latency going to drive networking, given gaming? Gamers are not just zitty 
>> 15 year olds, they are middle aged or older nowadays.
>> 
>> Mike
> 
> Ten years from now? Easy. We’ll still be talking about the continued shortage 
> of IPv4 address space and (legitimately) complaining about why IPv6 still 
> isn’t the default addressing/routing methodology for the Internet worldwide.
> 
> 
> -Andy

Re: 10 years from now... (was: internet futures)

[Andy Ringsmuth]

> On 3/26/21 12:26 PM, Mark Tinka wrote:
>> If the last decade is anything to go by, I'm keen to see what the next one 
>> brings.
>> Mark.
>> 
> 
> 
> So the obvious question is what will happen to the internet 10 years from 
> now. The last 10 years were all about phones and apps, but that's pretty well 
> played out by now. Gratuitously networked devices like my dishwasher will 
> probably be common, but that's hardly exciting. LEO internet providers will 
> be coming online which might make a difference in the corners of the world 
> where it's hard to get access, but will it allow internet access to parachute 
> in behind the Great Firewall?
> 
> One thing that we are seeing a revolution in is with working from home. That 
> has some implications for networking since symmetric bandwidth, or at least 
> quite a bit more upstream would be helpful as many people found out. Is 
> latency going to drive networking, given gaming? Gamers are not just zitty 15 
> year olds, they are middle aged or older nowadays.
> 
> Mike

Ten years from now? Easy. We’ll still be talking about the continued shortage 
of IPv4 address space and (legitimately) complaining about why IPv6 still isn’t 
the default addressing/routing methodology for the Internet worldwide.


-Andy