Sist endret 2021.09.10 13:25
Norwegian Unix User Group (NUUG) inviterer til medlemsmøte i NUUG Oslo.
Alle medlemmer og ikke-medlemmer er velkomne. Arrangementet er gratis.
- Tirsdag 13. oktober 2020 kl. 18.30
- Virtuelt møte på internett. Koble til i VLC med adressen https://nuug.no/live.m3u.
Månedens tema er: «0G: Escaping the Surveillance Blackhole with Free Mobile Computing» med Alexandre Oliva. Foredraget vil foregå på engelsk. Kort om presentasjonen:
So-called smartphones have long been recognized as a surveillance problem by the Free Software community, and it feels like, even at the speed of light, no personal data can escape them any more. Still, they offer useful features that many people find valuable enough to make up for their perceived privacy loss, to everyone's detriment. What if we could get the useful features without sacrificing our privacy? We have enough Free Software and Hardware, and we could have surveillance-Free OLPC-like networking with Tor, using onion-like services for incoming calls. Time seems ripe. Who wants one? Who wants to make them? How can we make them viable? We welcome you to learn more about the 0G project.
Alexandre Oliva is a Free Software Evangelist, GNU speaker, Recipient of FSF's 2016 Award for the Advancement of Free Software and FSF and FSF Latin America board member. He is also 0G foreseer of the 0G project, LibrePlanet São Paulo activist, Maintainer of GNU Linux-libre, and co-maintainer of the GNU Compiler Collection, GNU binutils and GNU libc and finally GNU tools engineer at AdaCore.
Spørsmål til foredragsholder kan legges inn på notatsiden mens presentasjonen pågår.
Medlemsmøtet vil også bli direktesendt over internett. For mer informasjon se https://wiki.nuug.no/grupper/video/streaming om direktesending av video.
Minner forøvrig om NUUGs IRC-kanal #nuug på irc.freenode.net, mailinglisten firstname.lastname@example.org for diskusjon om aktiviteten i NUUG og at NUUGs kalender er tilgjengelig på iCalender-format via <https://www.nuug.no/adict/ical.php?organizer=NUUG>.
The talk has been transcribed thanks to Quiliro Ordóñez. Here it is:
Ladies and gentlemen, dear passengers:
This is your captain speaking.
Thanks for flying with us.
We've just had a successful launch.
And we're now approaching cruise speed
away from the surveillance blackhole.
On the viewer screens
in front of your seats,
you can see images
taken by our flight camera.
The surveillance blackhole is in it.
But you can't see it.
You see, most blackholes
have halos around them.
That's light that nearly falls on it,
but still escapes.
But this is the surveillance blackhole.
No personal data escapes from it.
Not even at the speed of light.
So you see nothing.
That's why we're flying away from it.
So please keep your phones in airplane mode
and disable any data-gathering apps,
or you might be pulled back.
As you do that,
you might experience a weird
and very pleasant feeling of weightlessnes.
We call that 0G.
To avoid floating about in the canvan,
please remain seated for the remaining
of the operation of the flight,
with your seatbelts fastened.
We're still quite a while away
from our destination.
But I am sure you're going to love it,
... if we make it.
Please enjoy the flight.
And thanks for flighing with us.
Thanks for joining our dream.
Have a good flight.
All right, so lets get started.
This is now Alex Oliva speaking.
Thanks for inviting me to be here.
I'm very excited to speak here
and about this project.
I'm going to talk about surveillance devices
like the iSauron that has cameras (eyes)
and microphones and GPS.
And it collects data about you
and sends it out to ...
well you can probably recognize the logo.
Now, people carry devices like this around.
But I'm not only talking about the iThings.
Not because of the surveillance;
but because they actually offer
very desireable features.
Now the thing is:
couldn't we get just the features we want,
without the surveillance?
I'm going to talk some more about that.
But I believe
that we can make a viable alternative ...
much better actually,
than the personal communicators
from Star Trek.
That will enable us
to escape the surveillance blackhole.
Time is ripe for us.
We have the hardware.
We have the software.
We have the network.
And it's not that they're ready.
It'll take some work.
But I believe it's worth the effort.
Then we'll get the desireable features
and not the surveillance.
So, lets do that.
We live in 1984.
Not GNU's 1984; the George Orwell's 1984.
Our telescreens now are portable.
And we carry them voluntarily.
But just like in 1984,
they're always watching you.
They have cameras.
They have microphones.
And, because you can carry them around,
they also have biometrics
and location services.
And they are always on;
you cannot turn them off
... like in 1984.
And they are always listening.
Now, it's not everything like 1984.
They only had one Big Brother.
We have many.
We install different trApps
in our mobile phones.
And each one of them
serves a different master.
There's no way our personal data
can escape this way.
We have know for quite a long time
about this problem.
But we haven't managed
to fix it somehow.
But you've got to remember:
SMART...from smart phone.
Surveilling Most Actions 'Round Tentant.
And this is important: Tenant
because most people don't realize
that they don't own the devices they carry around
because they're actually controlled by third parties.
And, in the very 1984 style:
smart is idiotic.
And idIoTic is spelled with an IoT in it.
So, we've known about the problem
for quite a while.
And yet we can't resist.
Why is that?
Bradley Kuhn and Karen Sandler
talked about that at FOSDEM
a few years ago.
And they actually inspired me
to start 0G because of that talk.
So, I thank them for that.
The thing is,
these devices offer very useful features.
Why should we give them up?
I mean the features; not the devices.
Why should we give up features such as:
sending and receiving messages,
text, audio, video,
calls, social media, web,
looking up information on the web.
Why should we give that up
or even consider giving that up?
Maps, directions to get to places.
Very useful, why not?
You can actually
get this kind of information
without being surveilled,
without anyone tracking you
and running only free software.
We have free software for that today.
What we're missing
are the devices and the network.
Well, we've had some smart phones
designed to run free software
and others that free software
was ported to work on.
But most of them still have
some piece, in general of firmware,
that talks to the phone network
running nonfree software.
The one piece that communicates
on your behalf,
the one that the phone company
uses to track you
That is not a coincidence.
Now, we have to do better in this regard.
But if you think about it,
you might realize that,
in order for the phone company
to deliver messages to you
or get calls to your attention,
they have to know where you are, right?
They have to track you, right?
So, no problem there?
Well, guess what? They don't.
If you think you need
the phone company to track you,
to get calls to you,
you've got to rethink the network.
Tor, the Tor Project
actually provided me with the answer.
Most people who use Tor,
the onion router,
use it for anonymization,
to contact other sides
or peers anonymously.
It will route your connection
through other nodes
until none of the nodes knows
where the connection is from
or where it is to,
except for the very ends.
They know who's starting
and who's ending the signals.
You know, onion services
are the opposite of that.
Instead of you reaching out to a peer,
you get someone out there
to listen for communications for you.
So they contact another node
out there on your behalf.
And they will send back,
in a way, the request
all the way to your server.
The Pirate Bay is a famous web site
that uses this feature
and has remained untraceable
for several years.
I use that too to connect
to my home server over ssh,
regardless of whatever
what my ISP might think
of my having an ssh,
or willing to get ssh incomming connections.
I'm behind a dynamic IP address
that changes every so often
and it's actually, for some times,
an unroutable address.
And yet, Tor doesn't care!
It just figures out and it just works.
So you can reach your home server.
It can be a FreedomBox.
It can be just a web server ...
like our web server, the 0G web server
is running on the web server back home,
behind the same IP address
and it's reachable thanks to Tor.
"Hey, why not use the same feature
on ultraportable communicators?"
You announce the address
where you want to be called.
Not as a phone number,
but as a Tor onion address.
And then people call you there.
The thing is Tor
only supports only TCP, not UDP.
So some services
would have to be revisited.
But there are other ways to build 0G,
part of this interactive call.
Oh, by the way, you can see
some primitive logos for 0G on this slide.
I know they are not the finished.
You can help us with that.
That will be appreciated.
As for calls, I know that
GNUnet has peer to peer infrastructure.
And one of the applications
built on top of that
can actually make calls.
GNUnet can route calls
through pretty much anything.
So that is another way to go
about the call functionality.
GNU Jami can do that too.
So that is another way to go.
GNU Jami is entirely peer to peer.
So there are several way to go about
having your friends find you
without the phone company track you.
Now, if you're communicating
through the phone company,
(like using their data link)
they can still track you
through your SIM card.
So, better not use that.
What can we use
for non-tracked communication, then?
WiFi. Community WiFi.
Several cities are setting up
community WiFi services
that we might use, rely on.
It would be really cool
to have something like OLPC meshes.
One Laptop per Child project
had this notion of
schools having open WiFi
or open mesh networking signal
for students who lived nearby to pick up
And they would relay those signals
to students who would lived farther away.
And so the network
would end up covering the entire city.
Very cool stuff!
But that implementation
depended on nonfree firmware.
It'd be cool to replicate that
in freedom respecting hardware and software.
Something very cool, very awesome
that Daniel Peterson,
one of our earliest collaborators,
has been working on
(besides this logo that he proposed for 0G)
is longer range radio.
He is working on a modem
that you could have on your phone,
possibly as a USB external device
(if not built into the phone)
that could get you
hundreds of kilometers of reach
in low density populated areas.
Very little bandwidth
for this low frequency
But you could,
in this design we're talking about,
use that low frequency
to ask for help from the road,
if your car breaks,
like sending a text message.
Or to negotiate connections to other frequencies
that offer a lot more bandwidth.
So your could actually get internet service
or calls or whatever, if there is coverage,
if there are others willing
to route your communications.
I think that's pretty cool.
Now let me talk about hardware.
In my mind, 0G is a concept.
I'm not thinking of a product.
I'm thinking of a specification.
Software vendors and hardware vendors
and networking vendors, can adhere too.
Time is ripe for this.
We have a lot of technology available.
But I'm not looking to specify,
like one type of CPU.
Though, I want something free
as RISC-V, as possibly free.
I'm not specifying GPUs.
But I look forward to
Libre-SOC efforts for CPU and GPU.
WiFi: Well we don't have
a lot of free WiFi available out there.
Atheros 5k and 9k are probably the best choices,
especially for operation without infrastructure,
as in peer to peer mode.
Although it seems desireable
to avoid GSM or whatever,
it might make sense to,
at least start with devices
that also support that kind of network.
So, I wouldn't rule them out.
But it's probably useful for development.
As in, it's the sort of device
that you're going to find out there,
even for freedom-concerned suppliers.
But I'd like to see something
which could, at least,
disable the modem by hardware.
Now, what in my mind would be really cool,
is something around the lines of the EOMA68 project.
Same people who are now working on Libre-SOC.
What they figured out is:
"Hey, we could have a very small computer like a card;
the sort we used to expand laptops."
But in this size case,
it is an entire computer.
I'm just talking about the form factor.
And you can put that
inside a desktop case or a laptop case.
That was what they were planing to do
when they successfully crowdfunded ...
and, I am not sure, they already delivered that.
But I thought:
"Hey, why not put the same card inside a phone?
Why not have the phone have the connectors
to become a desktop or to become a laptop,
so the laptop would be just the case
which connects to the phone
and has a keyboard, a mouse or a touchpad,
a screen, aditional storing,
networking (wired even)?
And while it offers power to recharge the phone.
And then it's a single computer.
You can travel for a conference
and connect to the projector
directly from your phone for your presentation.
I'd really like to see that.
Ha! Who wouldn't? "
I started, when I first presented this, as:
"Who would want one of these?"
And now the question is more like:
"How many of these would you like?"
Now, even if a lot of people in our community
are interested in these devices,
someone who would like to make them
would probably need a broader bussines plan
to make them viable.
So I have some advice for them.
But, as I said, I'm not in the business
of making portable
or ultraportable communicators.
But I have some thoughts to share,
if anyone is interested.
My references mainly
are the One Laptop per Child project.
They designed devices to offer them to school kids,
to sell them to governments in large numbers
because you needed to make them in large numbers
to make them viable, back then.
Now, we don't need such large numbers, like millions,
to make hardware devices viable nowadays.
But it still makes sense
to sell hundreds of thousands of units
to governments all over the World.
You've just got to make them very sturdy
for childen...and cute...and very usable
These are very desireable features for a device.
Another piece of advice
that I pass on
from the One Laptop per Child project
is to focus, not so much on performance,
but on cost.
If you are going to sell
a million units to a government,
don't make them more costly
than 100 dollars a unit.
I think that's a reasonable number to aim for.
And then use public education
to grow the network that everyone else
is going to rely on for communication,
just like OLPC did.
And, of course,
why would governments buy this device,
rather than commercial,
proprietary, surveillance stuff
that others are offering?
Well, who would want to push
surveillance telescreens onto school kids?
Only very evil governments would do that, right?
So there's our chance.
We can sell the oposite of ad-ucation.
(That's my daughter's term.)
But we can even use even strong copyleft
in the portable communicator operating system
to try to avoid it being contaminated
by nonfree software.
Not sure if we can even do that.
But it's something to keep in mind.
And we've got to think in privacy-preserving apps,
ideally peer to peer
to avoid centralization
and Services as Software Substitutes.
For that we need a pervasive network.
So that's what what I think
a bussiness-oriented 0G hardware manufacturer
should keep in mind.
There's some more...
Something that bit the OLPC project really hard
was some partners
that had various conflicts of interest.
There were resentatives from Intel and Microsoft
in the OLPC board of directors.
And when the OLPC project
was negotiating huge sales
to governments all over the World,
somehow Intel and Microsoft learned
about those potential deals.
And lets say, made a different offer to them.
We should expect very fierce oposition
for this kind of revolutionary project.
So, I expect phone service providers
would dislike the revamping these services
this would force.
And some hardware manufacturers
might be unhappy.
So whoever works on this
should be aware of these risks.
I should mention that phone service providers
shouldn't be entirely unhappy
because probably it will take quite a while
for lots of people to jump into this permanently.
And in the meantime, a lot of us
will probably hire forwarding services
such as JMP.chat that delivers SMS over XMPP
and calls over SIP
from and to the regular phone system.
So, that's how you can get a regular phone number
on your 0G device, when there is one.
But the main point I want to make is
that OLPC was revolutionary.
But it didn't survive.
However netbooks were a trend that it sat.
And that it has remained with us
for quite a while.
So this revolutionary plan that I have
might not mean
that any specific device will remain.
But I hope the idea will.
And I am willing to build and to be the change.
Even if eventually it seems like we fail,
I think: if we make a dent,
that's already a great thing
because surveillance is something we cannot tolerate.
And if we succeed in decentralizing the net,
a lot of innovation
that will tend to keep it decentralized,
will evolve and hopefully
resist against to monopolize it again.
So, let's conquer mobile freedom! Shall we?
Well, thanks for listening.
Keep in mind 0G is sort of concept
encompassing Tor network-overlays-like
and draws a lot of ideas
from the One Laptop per Child project
and from the Open Moko phones
It's kind of a cross of all these
and a number of other ideas.
But it's mainly a concept.
And we need the people
to realize that concept
from the varios fronts.
If you would like to collaborate,
we are in the 0G channel on Freenode.
There is this web site
where you can learn more information from the project.
And if you don't want to collaborate
and do your own thing in 0G ideas,
that's fine too.
It's a little weird, though.
Well, I'm now ready to take questions,
if you can reach me, if I am available.
I look forward to them.
I look forward to seeing you in the channel.
And be free.
That's what this is about: Freedom!
And human rights are all about human freedoms and rights.
And surveillance is the oposite of that.
So let's escape the surveillance blackhole,
building this. Right?
Thank you so much! :-)
Original talk: Copyright 2019-2020 CC By-SA international:
Transcription: Copyright 2020 CC By-SA international: Quiliro Ordóñez