It's hard enough figuring out who we are now.
In the future, we'll also have to contend with who else is you too.
So today we will be looking at the concept of transferring or uploading your mind to
a computer, and we'll be digging into basic science of how we might do that along with
the many philosophical and ethical concerns that often come up on the topic.
We will also be looking at some that sadly tend to get skipped a lot and raise even more
existential concerns.
Now this is hardly a new topic for us.
After the Fermi Paradox and how to get into space and all the cool and huge things you
can build up there, cybernetics, AI, transhumanism and so on is probably the topic we look at
the most.
However we typically treat mind uploading as a quick sub-topic of a larger topic and
skim over it, and I felt it deserved a primary focus for a change.
It's a popular topic in fiction too, and if you throw in brain transplants into robots
or other bodies, it goes back to the dawn of science fiction.
Typically it gets a pretty pessimistic portrayal, and I'm not going to be portraying it in
the rosiest of lights today either, but in most fiction simply by doing it one inevitably
goes down a path toward madness or evil.
In more recent years it's gotten a better portrayal, often as one with no downsides
or problems, which I think is not a fair view either.
New technology generally solves more problems than it creates, but it still creates new
problems.
That's a major theme of our look at post-scarcity civilizations this year, that even if you
avoid utter catastrophes with new technology, even apparent utopias still have their problems.
For my part I appreciate fiction that tries to take an even-handed approach to new tech
and how it might alter our lives, rather than sugar-coat or wave away problems or just assume
a given technology is unusable and inherently evil.
One of the best new authors for that is Dennis E. Taylor, who is a welcome addition to the
hard sci-fi genre.
I was introduced to his work by members of our audience here and have recommended his
first trilogy, the Bobiverse, in prior episodes.
I've gotten to know the author since and was lucky enough to get an advanced copy of
his newest novel, the Singularity Trap, and as soon as I finished listening to it I messaged
him how great it was and realized afterwards it was 6 in the morning.
It's a great book, one that covers a lot of the themes we discuss on the channel, and
while he treats the topics seriously, there's also always a light-hearted and humorous tone
to his work I appreciate.
The Singularity Trap just came out, and is exclusively available on Audible for now,
but you can pick up a free copy of the Singularity Trap today, and also get a 30-day trial of
Audible, just use my link, Audible.com/Isaac or text isaac to 500-500.
In order to discuss mind uploading we first must ask what a mind is, where it is, and
what you'd be moving it to.
A key point to hit from the outset is that the mind and the brain are not synonymous,
the one is not necessarily the other.
This is good place to introduce a term that gets used a lot in these discussions, and
that's substrate.
Substrate is a term with a lot meanings, depending on your field, but typically means the thing
lying underneath something else that it's built on, or in printing, the medium something
is printed on, like paper or parchment.
In the case of our minds, the only substrate we currently have is our brains, that squishy
grey lump in your head.
In the future, that could change, though.
Now, it's not unusual to say that the mind is essentially software and the brain the
hardware it runs on, but that's oversimplified and why we prefer to use the term substrate.
I can print words on to paper but I can also print them on a T-shirt or the side of a mug,
in this digital age we often tend to forget how much text or an image can be altered by
the medium or substrate it exists on, and of course the ground you build a house on
or grow your crops in can impact both a lot too.
This is vastly more true for something dynamic and ever-changing like our minds, which are
heavily impacted not only by the format and function of our brain, but also by our bodies
and our hormones.
However, it's not just obvious stuff like getting nutrients and energy to run a brain
on from our body, or even how our hormones can alter our thinking.
Your height for instance, as most of us can recall from our childhood, alters your perspective
on the world in both a literal and metaphorical way.
So too, a person with very good vision sees the world very differently than someone with
impaired vision or who is blind.
Our bodies impact our thinking a lot, and whether you tilt toward nature or nurture,
hereditary or environment, as the dominant factor in a person's worldview, both are
inseparable from that body.
So when we talk about replacing it with something cybernetic or uploading the mind to a computer,
we have to take into account problems which would arise either by leaving that body out,
or by not emulating it well enough.
Failure to do so could cause serious issues, total failure, or rampant insanity in the
uploaded mind.
Your computer runs an operating system, like Windows, MacOS or Linux, which has been written
by programmers from the ground up.
To complicate matters, our minds aren't like that.
Instead, they are a product of our evolution.
They have been created and refined in the crucible of selection over billions of years.
So, if a mutation meant, for example, that we got better at remembering where our food
was stored, how to trick our prey, react faster to a predator that was trying to eat our ancestors,
or used less energy to think, they thrived that little bit better than the other individuals
who didn't have those mutations.
Those with the beneficial mutation survived to pass their genes onto the following generation.
A lot of the makeup of our minds are a product of things we might never understand, a reason
buried deep in our genetic past that gave our ancestors an edge over whatever problem
or calamity was facing them at the time.
Evolutionary leaps also added new functionality to our brains, producing reptilian, mammalian,
and primate layers which must communicate with each other.
Much of our individual personalities may be a result of how these areas interact.
Our minds don't come with instruction manuals or documented code.
Even so, our minds have made us the successful apex species we are today, but they are more
like Heath Robinson contraptions, products of a mad scientist, than the relatively well
ordered, documented computer programs that run on our computers.
Emulation is another term to add in here, as we often talk about emulating the mind,
or emulating the whole body, and it often accidentally gets used as basically meaning
to run or operate, when emulation is a bit more specific.
You are attempting to replicate the operating conditions with something else when emulating,
and this often means a loss in efficiency or an addition of complexity in order to do
that.
You can emulate a screwdriver with a butter knife, or vice-versa, but it's kind of clunky.
You can emulate a book with a Kindle or other e-book reader, but while that lets you carry
around thousands of novels, clearly better, there is a different feel to it, and the originals
were often poor substitutes.
So often the benefits far outweigh the downsides, and those downsides can often be minimized
or removed, but a device being emulated by another device is often not as efficient as
the real device.
Odds are that a mind based on neurons but running on neurons being emulated by microchips
would not function as well as a synthetic mind designed to run on those microchips in
the first place.
Basically, the human mind is a seemingly random collection of junk assembled at random by
evolution to operate.
This is why in spite of still being slightly more powerful than our best modern supercomputers,
even our math savants can't keep up with a cheap calculator with less complexity than
an insect's brain.
The flip-side is true too, though.
Our computers are not the same as our brains and we cannot simply run our minds on silicon.
To do that, we first need to mimic the brain itself on the substrate in the computer.
Emulation is a type of bridge, something that allows us to mimic our brains while running
on a computer substrate.
That bridge needs to be quite accurate, though how accurate is hard to say.
There may be a lot of wiggle room.
On the other hand, since we often talk about enhancing a mind once its uploaded, it's
potentially possible even a small tweak to improve performance in a minor area could
cause insanity, something we talked about more in the Cyborgs episode.
Emulations can be touchy too, you can often improve the graphics of an old game by emulation
on a more powerful modern computer, but it often results in system crashes or odd behavior
from unexpected problems.
If your goal is specifically to emulate a human mind, then efficiency is secondary,
but if you want an AI that is doing some task at a human level, you can probably do it with
less processing power because we don't need to emulate the brain substrate in the computer.
Many mental tasks, like physical processes, may not require a full human and may actually
function better when built to that task.
Much as it's cheaper to make a machine to turn a crank than it is to raise a child to
be a crank-turner, and this may apply to more generalized and abstract things like problem
solving or creative art too.
If you want to upload and emulate a brain, it's assumed you need to actually replicate
the neurons and their connections.
You craft an emulation of a neuron that completely emulates all its necessary functions for thought,
do that 100 billion times, once for each neuron, and now you load a mind to run on those, copying
the states of the mind's real neurons and the synaptic connections between them.
A lot of folks have tried to estimate how much processing power that would take, many
of them familiar names here at SFIA like Landauer, Kurzweil, Bostrom, and Sandberg, and the major
approach is the Multiplicative Method, where you figure out how much computing you need
to properly emulate a neuron, in terms of how it contributes to thought, then multiply
that by 100 billion.
"How it contributes to thought" is debatable, your average neuron is composed of a billion,
billion atoms, but you presumably do not need to model each one of those, same as you don't
need to model each atom in a concrete road or a steel truck chassis to model vehicle
traffic patterns accurately.
Choosing which ones to emulate, though, is quite a trick but we can make some educated
guesses.
In 1999, Kurzweil put the figure at 200 bits per second per connection, 1000 connections,
and 100 billion neurons, or 20 quintillion bits per second.
Or 20 million GigaFLOPS, 20 PetaFLOPS.
That's the one we usually use here but many argue it's way too low, and Thagard estimated
it 5 million times higher in 2002, at 10^23 FLOPS, reasoning that the number of basic
computational elements was not the 100 billion neurons, but rather the proteins in those
neurons and that you'd need to model those for an accurate emulation.
In 2006, Tuszynski estimated it even higher, way up at 10^28 FLOPS, though while I find
Thagard's reasoning for including proteins as the base computational unit fairly reasonable,
I think this last estimate is too high.
Most discussions of transhumanism you will encounter, even here at SFIA, tend to use
Kurzweill's lower figure.
Though it isn't the lowest nor the only approach.
I am going to attach a link to Sandberg and Bostrom's 2008 paper, "Whole Brain Emulation:
A Roadmap" in the video description, that's a very good but long paper at 130 pages, but
in Appendix A they include a table of many of the various estimates with appropriate
citation as to who did it and when and how, if you want to see the range of values and
the reasoning given to judge for yourself.
You'll see many values given in MIPS instead of FLOPS and I won't discuss the difference
between those two here, but they do there, lower in the appendix.
Needless to say the paper is a decade old and like most from that period makes some
fairly optimistic claims about improvements in processing speed.
Right now you can buy a teraflop processor for a couple thousand bucks, and if Kurzeweil's
figure is right than 20,000 of those would let you emulate a human mind for 40 million
bucks, plus all the additional hardware, plus an electric bill that would make the Sierra
Club weep at a couple thousand bucks per hour.
Of course that's all chump change by modern computer research budgets so you're probably
wondering why no one has done this yet, but they've certainly been trying, and getting
pretty close as these things go.
IBM managed to simulate 500 billion neurons and 100 trillion synapses a few years back.
The hardware appears to be just about there now, Moore's Law might be staggering or
even dead in recent years but we are just about fast enough now and in a decade or so
should be there There's more to it of course, beginning
with capturing an image of an actual brain to genuinely render, you can have the computational
power but still need that basic program, as it were, same as you can make a 500-page blank
book fairly quickly but filling that up by writing a novel takes much longer, and until
we've done one and have some digital person to talk to, we can't say if that produced
a solid mind emulation.
Indeed you'd have to chat with the original donor whose mind was copied to see if the
emulation was really matching up, not just an AI, the point of such a thing is not just
to produce a mind able to act human, but to be a true copy of a specific mind, and if
it is acting very differently than the original, ignoring the shock to perception and worldview
of being uploaded, you didn't succeed.
It's also important not to think of the computer as particularly necessary to this
or a black box.
The substrate matters, but so long as it is emulating neurons, it shouldn't matter too
much.
If we upload someone's mind, call him Bob, to a computer simulating neurons, if done
correctly it should act the same as the version of Bob running on real neurons or some other
substrate.
For instance, assume someone has integrated a device to convert what you say to the signal
Bob's emulated neurons for speech receive, and vice-versa for him to reply, you are still
talking to Bob whether you are talking to a cloned copy of his brain sitting in a vat,
one running on a supercomputer, one being handled by a giant hive of ants or vat of
algae, or one running on people.
You could be strolling around a crowded planet-city, an Ecumenopolis, of a trillion or so folks,
most of whose job was to take shifts throughout the day to play neurons for Bob, stopping
at restaurant to eat meal and talk to Bob on the phone, while effectively sitting inside
his brain, surrounded by folks who are busy doing some tiny piece of Bob's emulation,
or on a lunch break from doing so, and have their own, totally independent brain that
Bob can't see, and you are still talking to Bob.
Indeed on a big enough Ecumenopolis there could be many different distributed minds
running on the place and that restaurant might have one or two folks from each mind there,
and they might be assigned new scripts for different neurons, even of different minds,
every shift.
So it's hard to even think of it as being inside Bob's brain.
His mind still exists though, by any logic that permits a brain emulation on a computer,
it's just as valid and you can talk to Bob.
Or a version of Bob anyway.
Much as I can handwrite a book, the first edition, and photocopy it, and that book will
still exist if I torch the original, we can have many Bobs at once, a legion of them.
Now, the mind is constantly shifting and these will diverge too, over time, as they get different
experiences.
They can also diverge very quickly if the copying process isn't quite perfect or if
the shock of the process alters their worldview, as suddenly being uploaded to a computer is
presumably a life-changing event at least on par with something like a heart attack
that makes you revise your diet and exercise routine and decide you should focus more on
family than work.
Nothing new there, but something that gets overlooked occasionally is that simply being
a copy, and knowing you are, can make you act differently.
In Taylor's prior trilogy, the protagonist, Bob, has each copy pick a new name and being
a bit of geek, Bob and his copies tend to borrow them from science fiction, and as more
exist, other popular fiction.
The thing is, many of them put an extra focus on being a bit like that person.
If copy number 100 of me took on the name James T. Kirk, 101 might be Picard, and 102
might pick Sisko.
I'm very fond of all three of those characters and by my very nature will want to stand out
as an individual, even among copies of myself, indeed probably more so.
I'm also fond of D&D and role-playing games so acting out the part a bit for fun would
come easily enough.
I'd still be me, at the core, but I could easily see Kirk-100 being more brash and daring,
Picard-101 trying to be thoughtful and diplomatic, and Sisko-102 being more confrontational and
ethically-gray, right from the outset, just because I value individuality and diversity
of opinion, and so not only is that character influencing me a bit as a theme, but I'm
going to try to let that part of me that resembles them have more of the lead.
So that divergence of behavior can happen really fast, even in perfect copies, if they
know they're copies.
A person who wants to be a doctor or an astronaut, but also has heavy family obligations that
interfere with that, could copy themselves twice and have each follow those paths.
When they come back they will be changed by that, but they will also be different from
the moment of that assignment.
If you've been awoken and told you are both free of all family and social obligations,
and also told you aren't too welcome to continue those relationships because another
copy of you is handling that, that mix of liberation and exile is going seriously change
you far more than a typical day will.
Speaking of astronauts, a common concept nowadays is to colonize the galaxy by sending uploaded
minds.
They can be packed on compact hard drives and later transferred to biological bodies
to help survive the rigors of the journey, and lower costs, and can be beamed at light
speed to distant colonies.
In our post-scarcity civilizations discussion, we've mentioned the dilemma that it might
be hard to find colonists if life back home is a Utopia.
I don't find that too plausible, but you only need one copy of a colony, a couple of
hundred volunteers.
Heck, you only technically need one.
They will all diverge pretty quickly, there's years of lag time between star systems and
each will also start at a different time and have a different lag time getting news and
tech updates from back home on Earth, so each colony will be pretty different right from
the outset.
What's more, even if you only had a 1,000 volunteers and needed 500 to fill all the
roles for a new colony, you ought to have a huge number of permutations of specific
colony composition even considering that you might need 2 doctors at each and only have
7 folks qualified to hold either slot, you'd have 21 unique medical teams, more if one
was the boss and the other was the deputy at random, which would give 42.
If you had a commander for each shift, 3 shifts, selected from 20 decent candidates, you'd
have 1,140 possible trios, but 6,840 possible duty rosters, and that matters, because whether
you are on the morning shift or the evening shift alters who you are interacting with.
Two completely identical colonies, same 500 people, could turn out very differently.
In one, Jessica got the morning watch, and becomes fast friends with her comm officer
Jane, and they share a love of pottery and start a club that does Ceramics on Wednesday
evenings, and then since the colony wants a unique name rather than a number, and polls
its crew, that club suggests Kiln and it wins the name.
The planet goes on to take pride in pottery and trades in unique hand-crafted pots a lot,
which many folks back on Earth collect.
In the other, Jessica get the night shift, and her Comm Officer is Jake, and they hit
it off too but romantically.
It doesn't work out though, and they have a bad break up, and Jessica is stressed and
snappish and makes some enemies and Jake, being a bit bitter, helps lead those to stage
a coup.
Two very different histories from the tiniest change.
This shouldn't be surprising though, for all that we talk about diverse human genetics,
we are all basically copies of each other with very small changes to DNA, we just focus
on those changes a lot.
Your DNA contains about a billion bytes of information, the equivalent of around 10,000
books, a good sized library.
Trying to hunt down those differences between you and another human would be like hunting
through that whole library to find typically about one word being different every other
page, and a lot of those are pretty trivial, like replacing the word sphere with ball or
globe, or large with big or sizable.
The significant ones are much fewer.
That divergence is enough to create unique people, even before we include environmental
differences of experience.
I don't think you'd have any problems coming up with a thousand colonists, especially
since they don't have to leave, just include a copy for use, but that's a near infinity
of options when we toss in obvious differences in destinations and challenges too, with years
of divergence to work with.
I'd bet I could get the full thousand right out of my own audience with room to spare.
Even if you only got one, though, you could just do your copying here, make a few thousand,
and give them a few years to diverge as they all train separately in each needed specialty
and randomly pick the crew from those and ship them off.
By making copies of the copies for future use, you only add to those changes and diversity.
Another option, as a way to deal with declining populations if birth rates tumble because
people don't want to raise kids, is you just copy people old enough to care for themselves.
Notice by the way that we are not talking about artificial intelligence.
Someone like this might be able to match one, since they could add more processors or be
sped up to experience subjective centuries in minutes, as we've discussed in our Transhumanism
and Cyborg episodes, but they aren't an artificial intelligence.
Okay, the last major concern on this is always about the upload itself and the copying issue,
along with the idea that you might slowly replace a person's neurons with tiny computers
instead, the gradual approach.
As I've mentioned before, this is copying, I wouldn't consider an upload of me to be
me, and of course neither would that upload since it shares my views, at least initially.
He would not regard himself just as a copy since I wouldn't, we're just two different
Isaacs, but not everyone shares this view and if someone walked in and shot original
me I would expect my copy to act as original Isaac since we both have duties and obligations
important to us, so he can't take off to Tau Ceti without at least duplicating himself
again.
Other folks might feel differently and there's no real way to prove or disprove a lot of
the varying opinions on this topic.
My copy acts in accordance with my opinions on the matter and in my own case, if it was
just me and him, we'd break down all the things we had to do by what each of us might
have an easier time handling, then flip a coin on the rest, trading them around a bit
to ensure that we were setup to be functional as we diverged, and we'd always introduce
a random element to decision making.
Alternatively, if I were gradually replacing neurons with computer chips, just having little
robots scan recently dead or dying neurons and then replacing them, I would consider
that still to be original me, and that seems to be the majority view in my experience.
There's no real copying going on that doesn't happen all the time by biology, the replacements
just happen to be of a different substrate.
So I expect this one to be the more common approach for civilization in the future.
Also, our bodies and brains are constantly being refreshed and replaced.
You would be hard pressed to find atoms in your brain that you have had since birth,
so the concept of completely replacing neurons with machines is hardly much different.
That's to a point anyway.
I think most folks would start that way and after a while decide they were still themselves
and order a final rapid conversion of all remaining cells, just to get it done with.
When contemplating this philosophically, whether the gradual change takes one century or one
minute, it's hard to argue that anything different has occurred.
Of course if some super-powerful laser is scanning your brain so hard and accurately
it is vaporizing a layer of brain at each step, and uploading that to a computer, it's
very hard to argue that's functional different.
Continuity of consciousness is arguably interrupted, which it is not if you are just replacing
a neuron at a time, but by that argument if I freeze you and thaw you out a century later,
or shut down a computer emulating a mind for a night and turn it back on, consciousness
was interrupted too.
For that matter consciousness is interrupted every night you fall asleep, your brain keeps
going and doing stuff, but you are literally unconscious, and I don't think we can just
toss that aside as semantics owing to these words and terms predating modern science.
I can't offer any answers on continuity or identity, incidentally.
That's been a furious discussion since the Ship of Theseus thought problem in Ancient
Greece, about if replacing components of an object makes it cease being that object anymore.
I consider these to be the sort of things you probably can't prove or disprove.
Even my own opinions on them mostly represent me trying to be pragmatic when contemplating
these things.
I can't prove I have free will either, but I take it as a given I do, not because I can
in anyway prove that, but because I consider the alternative, and its implications, to
have no value.
In the same way, I can't prove this reality is real, not a dream, but there's no value
to me in believing it is a dream, even if it is.
So I come down on whichever side of these discussions seems best suited to avoid me
losing my mind or sense of identity and purpose, but I don't confuse that preference with
proof.
I think we've got enough food for thought for today.
We could spend whole episodes discussing the mechanics or philosophical issues or applications
of mind-uploading, and we probably will revisit the topic and we have looked at some of those
in other episodes, we've also got some great books on the topic we've recommended before
and we did recently get around to finally compiling an almost complete list of all the
books I've recommended over our 137 official episodes and some specials, I'll link that
in the episode description too and I would appreciate anyone letting me know if there
any missing or listed wrong.
However, one of the most fun of those for me was the Bobiverse Trilogy by Dennis E.
Taylor, and his newest novel, the Singularity Trap, takes an interesting direction on the
gradual replacement approach too, along with several of the other topics we look at here.
A novel doesn't have to be hard science fiction for me to enjoy it, but I always like
seeing more books where the author takes their science and research seriously, without dumbing
it down either, and still have a good time with it too.
However instead of telling you more about the book, I'll turn it over to someone who
knows it better.
Singularity Trap touches on things like the Fermi Paradox, Uploading, and the questions
of consciousness and individuality.
All timely topics in science fiction but a long way from what I grew up with.
As science fiction evolves it is getting simultaneously easier and more difficult to write.
I grew up reading Golden Age science fiction with Captain Future and gravitometers and
Podkayne of Mars and the lush jungles of Venus.
Those days of swashbuckling spare opera that plays fast and loose with science aren't
necessarily gone—witness the Star Wars franchise—but it can be a lot harder sell now.
This doesn't mean there's a lack of subject matter, though.
Now we have multiverses, Fermi Paradoxes, alien biologies, Quantum Weirdness, terraforming,
megastructures, various forms of 'punk', post-biological intelligences, and anything
with "Einstein" in the name.
But the SF readership is much more educated and discerning these days, thanks in part
to resources like Isaac's SFIA channel.
So authors have to do their homework or face the possibility of a good old fashion flaming.
Not that this is in any way a bad thing.
SF has always been about positing "What if this were true" and then running with
it.
We are in as much a Golden Age as the days of Jack W Campbell and Robert Heinlein.
Only the specific subject matter has changed, not the sense of wonder.
And Dennis does a great job covering a lot of those topics in his novels.
The newest, The Singularity Trap, is exclusively available on Audible right now, and you can
get a free copy today, just use my link in this episode's description, Audible.com/Isaac
or text Isaac to 500-500 to get a free book and 30 day free trial, and that book is yours
to keep, whether you stay on with Audible or not.
Next week we will be returning to the Outward Bound series for a look at Colonizing Mercury,
and we'll discuss some pretty interesting ways we can make that barren, sun-scorched
rock a nice place to live.
For alerts when that and other episodes come out, make sure to subscribe to the channel,
and if you enjoyed this episode, hit the like button and share it with others.
Until next time, thanks for watching, and have a great week!
For more infomation >> マツダ「コスモ」が復活する？レンジエクステンダー搭載で2027年に登場するとの噂 - Duration: 6:16. 
For more infomation >> ル・マンをモチーフにしたポルシェ「911RSR」が公開。これでレースに参戦だ！ - Duration: 4:09. 
For more infomation >> 對柯P高唱「分手吧」 民進黨選對會決自推人選北市長 - Duration: 4:33. 
For more infomation >> 집사부일체 첫 사부 전인권, 63년 한 집에서 살며 지켜온 아내 정혜영과 딸, 아들의 가족사 - Duration: 7:41. 
For more infomation >> Disney A Wrinkle in Time
For more infomation >> Suzuki SX4 1.6 SHOGUN NED.AUTO, 1STE EIGENAAR, AIRCO, ELEKT. R+S, LM-VELGEN, ZEER NETJES - Duration: 1:05. 
For more infomation >> Free admission to Vermont state parks - Duration: 1:15. 
For more infomation >> Dilemas sociais - Quando olhar só para o próprio umbigo atrapalha todo mundo - Duration: 4:23. 
For more infomation >> The Surprising Truth About Email Marketing in 2018 - Duration: 6:24. 
For more infomation >> D'Youville Graduate Commencement 2018 - Duration: 2:04:31. 
For more infomation >> Release Highlights - June 2018 Version 10.8.2 - Administrator - Duration: 6:23. 
For more infomation >> Laura Smet et la « grosse raclée » de Laeticia Hallyday après un message sur le cancer de Johnny - Duration: 2:15. 
For more infomation >> Sylha & Mon'éthique - Lancement le 13 Juin à la Belle Electrique - Duration: 0:37. 
For more infomation >> Octopath Traveler - Cesta rituálu a výzkumu - Duration: 3:52. 
For more infomation >> Audi A1 Sportback 1.0 TFSI 95pk Adrenalin S tronic | Navi | Cruise Contr - Duration: 0:53. 
For more infomation >> JOURNÉE RALLYE GENDARMERIE 2018 EDSR 34 + Trajectoire de Sécurité Moto #MOTOVLOG 23 - Duration: 8:35. 
For more infomation >> Live in the D: New season of family fun - Duration: 4:38. 
For more infomation >> Live in the D: Need something to do on the weekend? - Duration: 4:51. 
For more infomation >> Live in the D: Motor city brew tour - Duration: 2:58. 
For more infomation >> Live in the D: Keep an eye on this guy - Duration: 5:25. 
For more infomation >> MINI Clubman 1.6 D Cooper/ Panoramadak/ Mini navigatie/ Climate control/ Cruise contro/ Lmv - Duration: 1:05. 
For more infomation >> Uomini e Donne gossip: il gesto d'amore di Sara Affi Fella per Luigi Mastroianni - Duration: 3:07. 
For more infomation >> Comique Strip' de Marion Lebègue - Duration: 1:40. 
For more infomation >> Physical Disability and Labor Market Discrimination: Evidence from a Field Experiment (français) - Duration: 4:42. 
For more infomation >> Rapha – Riding is the Answer – Bianca Pettinicchi - Duration: 1:01. 
For more infomation >> Groups - Create a Group - Instructor - Duration: 4:18. 
No comments:
Post a Comment