This Week in Google 284 (Transcript)
Leo Laporte: It's time for TWiG, This Week in Google. Jeff Jarvis
is at Davos, so we're going to give him the day off. But, hey, don't worry.
Matthew Ingram and Kevin Marks are here. We'll talk about the latest news from
Google. Why don't we spend some time talking about the latest news from
Microsoft, too? It's all coming up next on TWiG.
Voiceover: Netcasts you love from
people you trust. This is TWiT! Bandwidth for This
Week in Google is provided by Cachefly.com.
Leo: This is TWiG, This Week in
Google, episode 284, recorded January 21, 2015.
Legislating
Unicorns
This Week in Google is brought to you by Casper, an
online retailer of premium mattresses for a fraction of the price. Because everyone deserves a great night's sleep. Get $50 off
any mattress purchase by visiting casper.com/twig and don't forget to use the
promo code TWIG.
And by lynda.com. Invest in yourself for 2015, lynda.com has thousands of courses to help you learn new tech, business and
creative skills. For a free ten-day trial, visit lynda.com/twig.
And by LegalZoom. Visit legalzoom.com to save on your legal needs and
gain access to a network of legal plan attorneys for guidance. LegalZoom's not
a law firm, but provides self-help services in your specific direction. Visit
legalzoom.com and use the offer code TWIG to receive $10 off at checkout.
It's time for TWiG, This Week
in Google. The show where we cover not just Google, but the
Cloud, Facebook, Twitter, anything happening up there in the sky. Kevin
Marks is down to earth but he is in his garden with the lovely sky behind him. Hi, Kevin.
Kevin Marks: Hi there. Not many clouds, though, I'm afraid. It's
clear blue sky here.
Leo: Glad to have you. Yes, and if it starts raining, just,
you know. I don't know. You have an umbrella, I think
you'll be all right.
Kevin: Not much chance for that here.
Leo: I love the orange tree behind you, though. Also with us from Canada, where it's not quite so balmy, Mr.
Matthew Ingram, GigaOm.com.
Matthew Ingram: Yes, no orange trees in my backyard.
Leo: No orange trees in your backyard. But there's always a
place for you if you want to come down here. Matthew writes for GigaOm and both of us have been talking before the show
about what I think is the biggest story of the day, which has nothing – it has
little to do with Google. Actually, in some ways it might have a lot to do with
Google, which is Microsoft's Windows 10 announcement, which we've been covering
live most of the day. Just the short hand version of this, Microsoft has
announced that it will be giving away Windows 10 when it ships later this year,
probably in fall, to anybody running Windows 7, Windows 8 or Windows phone and
that, in effect, what they've been talking about for some time, one Windows, is
going to somewhat come true. There will be universal apps that will run across
all platforms. There will be a form of – Microsoft calls it “continuum” but
it's really like Apple's continuity that allows you to run apps across all the
different platforms and data the same way.
They also announced, and I think this will get a lot of
attention, a holographic – they call it the HoloLens,
which I guess is designed to be used by hollow men, with apologies to T.S.
Eliot. The HoloLens is a heads up display which
offers 3D holographic images in your field of view as you look around. It's
being used, apparently, now by JPO at NASA to help scientists who are designing
the – you know, plotting the plans for Curiosity, the Mars rover. Literally,
walk around. Mark Rocks says, “Hey, go get that.” Here's what it looks like.
You know, some pretty amazing uses. That's going to
get a lot of attention.
But I think really, the big story is kind of a
revitalized Microsoft under its new CEO Satya Nadella. They did – there is a
Google angle here because very early on, at the very beginning, Microsoft – it
was Terry, I want to say, Richardson. He's the Senior Vice President for
Operating Systems. He said, at Microsoft, you're not the product. I think
that's a pretty clear shot at Google, Facebook too.
Matthew: They said they were canceling the Screwgle nonsense.
Leo: They did cancel Screwgled. I
think this is really the new Screwgled, not to go
after Google in advertising but with snark but to go
after them with better products. What an idea, what a concept. They showed off
Cortana, which of course, is directly aimed at Siri and Google's Google Now. It
does many of the same things Google Now does and again, they emphasized, “But
it's in a trustworthy way. Your privacy is protected.”
Cortana was very conversational. I know – Matthew, you
didn't see the video either?
Matthew: No, I didn't.
Leo: No, neither of you saw the stream. But the demo was
pretty impressive, very conversational. Of course, it's always hard with demos.
Matthew: Yes.
Leo: But it seems to me that Microsoft is looking at Google
as a serious competitor and doing everything they can to bolster their product
line in every way. They mentioned a new browser. They're going to – not exactly
a redesign of Internet Explorer. The same engine running –
the same rendering engine running it. But Spartan will look very
different, much cleaner. In fact, it looks a lot like Chrome. I imagine
Microsoft loses some market share to Chrome with new Windows users.
Matthew: I guess the thing that sort of bugs me about – I mean,
you're right. Microsoft is, you know, offering a lot of interesting things and
the free Windows is a great way to get people on to the platform. I feel like
you could say this about Google too; I suppose you have to – it feels like you
have to pick someone to get married to, you know, to basically be absorbed into
a platform and use everything that platform has to offer you. Have to have the
mobile device, have the desktop software, have to have the – they have to suck
in all the data about your life before things like Cortana can become useful.
So you really have to choose who you're going to sort of get married to because
a lot of these things are not going to work unless they have all your stuff,
and all your objects and all your devices.
Leo: Yes. Apple invented this, didn't they? The whole idea of a complete ecosystem. Google has done
their best with Android and Chromebooks to do the same, although they don't
really have a desktop operating system. They have Chrome, which is certainly a
wedge into the desktop operating system. It's very clear that today, Microsoft
plans to do the same thing. Satya Nadella said, “We plan to be cross platform,
everywhere our customers are, because we're a customer-centric company.” He
even said, “Instead of having to use Windows, we want you to love to use
Windows.” But he said the best experience will always be within Windows 10. He
also told developers, and this is a really important message, they have been
shedding developers. The fact that Windows Phone didn't take off, I think you
can ascribe it a great degree to the lack of apps, the lack of developers for
it. He said, “We're going to have a universal platform and you'll be able to
write for all the devices.” Then the pitch to the consumer is Cortana and all
these other things work better the more time you spend with Windows devices.
Everything will work better. So you're exactly right. I think all three
companies, Apple, Microsoft and Google are positioning themselves to be the
universe. Microsoft has – go ahead.
Matthew: I guess, I mean the main reason I stopped being so
exclusively Apple was that started to bug me. Sort of, the fact that you had to
use everything Apple and that certain things just
wouldn't work if you didn't use Apple. Plus, just the sheer
expense of buying all of these Apple products, because they're not exactly
cheap. I wanted something that was more – you know, where I could hack
together networks out of different things as opposed to having to have
everything Apple. They are actually getting better than they used to be, but
that was the thing that sort of pushed me away. It felt like Android was a much
more open ecosystem.
Leo: Is it?
Matthew: Good question. I mean, is it? I don't know. Certainly,
when it comes to things like Google Now, I use Gmail and I have a Google device, all that stuff. So everything – I use Google Drive,
and Google Plus and Google Photos. So I am sort of – you know, I've bought into
their ecosystem. So is that different? I don't know.
Leo: Your original premise was, this was – do we add Amazon
to this too? Isn't Amazon trying to build that same sort of total ecosystem now
that they do hardware?
Matthew: The only thing I could say about Google is they do
tend to be better at letting you get access to your data and take it with you.
I don't know if that's changed or not, but that was a pretty core principle for
them.
Leo: Absolutely, yes. Go ahead, Kevin. You're a little low, I don't know if that's us?
Kevin: I'm not sure.
Leo: Oh, we hear you now. There you go.
Kevin: Okay, maybe I was -
Jason: There was a little bit of typing, so I was riding
fader. Sorry about that.
Kevin: Oh, I'm sorry. I'll repoint the mic. So what I was
saying is, that's the thing. Each of them is having to build the whole set because they've stopped trying
to inter-operate, which is kind of sad. So everyone's building their own Cloud
services, their own hardware, their own operating system, their own desktop
applications and things like that. It's all a bit odd, because we've had the
Web as the unifying platform for a while. But they've all seemed to have said,
“Oh, no, we can't actually work with each other anymore,” which is odd. Because it leaves this sort of gap in the middle for someone like
Dropbox to be the Cloud that works on all the Clouds, which is kind of strange.
Matthew: Yes, in fact, Dropbox is a great example. I mean, I
just said the other day, I've got a Macbook Air. I'm running Linux Mint on my desktop and I've
got Windows devices in the house as well. I was trying to share folders on my
network, you know, which I control. It made me want to shoot myself. It's so –
even a simple thing like that is so hard. They've made it hard. Apple's
actually made it harder than it used to be. So Dropbox becomes like a godsend.
You literally just put stuff in there and it's everywhere. It's as simple as
sort of cross-platform networking and file sharing should be, or could be.
Leo: You can't really fault them. I mean, the best way to
profit is to force people to use your stuff exclusively. Although, it is
encouraging that a company like Dropbox can make its way by offering a true
cross-platform solution.
Matthew: I mean, Apple's argument has always been, you know.
It's similar to the one you described that Microsoft is making, “By doing this,
we make it better for you, the user. So your experience is better, things work
properly, you don't have to worry about anything getting damaged or your data
flowing to other people because we control the whole thing from soup to nuts,
from one end to the other.” That's why the walled garden is such a great
metaphor, because it's a beautiful garden. The flowers are amazing and it's
really well – it's clean. No one mugs you. You can barely even see the walls
until you get up close to them and even then, you think, “Ah, this is such a
great garden, man. These flowers are amazing.” But, you know, the minute you
want to do something outside the garden, it's like you can't do that.
Leo: I have to say, though, I look at Dropbox and then I look
at Microsoft offering unlimited storage on one Drive. Prices
dropping at Google. I wonder how long Dropbox's business plan as a
storage company alone can survive. I mean, I wonder how much longer they're –
Kevin. You're an advocate for open – you're an advocate for exactly the
opposite of what we're seeing here.
Kevin: Right, I mean, that's the thing. I suspect that we're
at one end of the pendulum swing and we can have another swing back to the
other end. The Web is still there, it's still that infrastructure for all of
this. It's just that these different companies are just approaching it from
slightly different angles. Google was the one who was approaching it from that
angle, but because as they grew, they got more locked out by the others – to some
extent, they had to build their own operating system as a defensive move. They
built them as a source operating system so that other people can use them too.
So the biggest OS in the world may well be AOSP rather than Google Android some time this year. That line looks like it's going to
cross.
Leo: Explain what the difference is.
Kevin: So, Android is two things. Android is an open source
project that's an operating system for phones and it's also a set of Cloud
services that Google provides. So if you have an Android phone with Google on
it somewhere which – oh, in this case, I have Nexus. So it just says Nexus,
it's implied. That means that they've signed the deal with Google, they get to
use Google Cloud services. There are certain – which means that without that, you can't run Google's apps that rely on their service. So
you can't run Gmail, or Google Maps, or Drive or the App Store and so on. So a big chunk of the Android functionality.
AOSP is Android Open Source Project, which is all the bits that you can freely create and build your
own stuff. Google has been moving bits back and forth between the two. So, you
know, one of the reasons on a non-Nexus phone, you end up with two copies of
everything, is you have an AOSP-derived email client and Gmail. You have an
AOSP-derived browser and Chrome, and so on. This is, if you buy a Samsung
phone, confusing. You end up with two of everything, because Samsung has
decided, “Okay, we don't want to be fully dependent on Google. We want to make
sure we have got -” [crosstalk]
Leo: You know what's interesting, I just saw a browser benchmark with SunSpider only,
which I think is a javascript benchmark. It showed
the Samsung browser on the Note 4 to be literally twice as fast as the Google
browser, the Chrome. So it is possible for Samsung to say, “We're going to make
better apps.” Not just Samsung apps, but better apps.
Matthew: Right, because they're plugged into the hardware,
right? They can use all sorts of things that Google can't.
Leo: You know, Apple has chosen to
make its solutions barely inter-operable. I mean, really, Apple is Apple.
You're not going to try to use iCloud on any other platform.
Matthew: Nor would you want to.
Kevin: That is a change.
Leo: Nor would you want to.
Kevin: But that is a change, you know. At the point where
Apple weren't the ones who are in the league -
Leo: They couldn't do that.
Kevin: They had iDisk. And with iDisk, you could use on Windows and you could use other
places because it implemented web app. That was the standard way of using it.
It was a web server, and it was great and I miss it. Especially because I paid
them money and they deleted all my files. That's not very nice.
Leo: That's not nice.
Kevin: But no, they've gone the other way and said, “Okay,
we're going to build something that works for us.” Apple has a lot of good -
work as well, there's not news. But the companies tend to do that work at the
point where they're not on the largest end of the market. So one of the things
Apple did to get itself out of its trough was to adopt a freebie estate call to
the operating system and a bunch of standard stuff for that layer, and built a
lot of very good networking protocols and things like that, right? MDNS and all
the stuff that makes up Bonjour is a set of internet protocols that they helps
establish. There's a huge amount to establish Wi-Fi. You forget that now, but
it was Apple who decided, “Yes, this is really important and we're going to
make the base station a - anybody else's base station to ensure that people
have lots of base stations. We'll put it in by default on all our machines so
there's a reason to buy a base station and boost it way up.” That created the ability for everyone else to follow along and join
in. So Apple is very good at making those choices, making the integration good
enough that you understand why they want it and then ramping the industry forward. You can say the same thing about USB, to some extent.
Matthew: USB, exactly. But I think you're right that they did a
lot more of that when they were relatively smaller, not the world's most
valuable company. They felt like they had to try harder.
Leo: As soon as you have market share, you go, “Screw the
other guy.”
Matthew: Right, why wouldn't you?
Leo: As J-DoggNunan [?] says in
our chat room, “Microsoft has walls. They're just not as high as Apple's
walls.”
Matthew: That's true.
Kevin: Well, thinking now, Microsoft is in that situation.
Matthew: I want to point out, Kevin also has walls. Kevin is in
a walled garden right now.
Leo: Kevin is literally in a walled garden.
Kevin: It's a fence, actually.
Matthew: They're very nice.
Kevin: Nice wooden fences, only about seven foot high. Yes.
Matthew: Beautiful garden.
Kevin: But the houses aren't very high here, so we aren't
overlooked, which is nice. So Microsoft is now in the situation that Apple was
then.
Leo: In a way, they've flip-flopped.
Kevin: And where Google was a few years ago, where they are
the outsider trying to get people to come back to their platform that they think
is great. So they're going to have to do things differently, and they are.
That's one of the things that have been refreshing about the last year or so
with Microsoft is that they've been doing things like, “Oh, yes, of course we
support running our software on Linux.” And you're like, “Wait, what?”
“Oh, yes. Our cloud will support Linux images as well
as Windows images,” and you're like, “Hm. This is not the Microsoft I
remember.” It's encouraging. I'm hoping that we'll see more of that and, you
know, obviously they've restarted browser development with IE and they've been
doing good work with standards there as well. So I'm hoping we'll see more of
that and with the other shoe to drop there, they've been releasing versions of
Office and so on for Android and iOS.
Leo: Even before they released it for Windows Phone and
Windows Desktop. The touch-first Office came out on iPad, is coming out on
Android, still is not available for Windows or Windows
Phone. But that's what Nadella was talking about. He says, “We're going to be
cross-platform. We're going to focus on what the customer wants.” That's nice.
“But the best experience will be in Windows.” Is that the best we can hope for,
really, in this modern world? Is there a chance, Kevin, of this kind of wonderful
idea of open?
Kevin: Google tried doing that the other way around. Google
tried very hard to put stuff out on iOS for quite a long time and then at the
point where Apple started running interference on them and cancelling apps, and
they got into a huge fight about it and called in the FTC. At that point they
said, “Okay, if we've got to spend six months getting this into your app store,
we're not going to do iOS first anymore.” I think they swam back a bit from
that again and they've -
Leo: But that was a market – ultimately, yes. I know there
were court cases. But ultimately, that was a market share decision. Google did
that and I remember, this was Andy Rubin was pissed off about this. But it was
ultimately a market share decision. “Yes, we've got Android. But where are all
the users? They're on iPhone. We need users.” But as soon as you have market
share on Android – in fact, look how Google has treated Windows Phone. I mean,
really, it only exists from Google's point of view.
Kevin: Yes, and that's a reasonable decision. That's like,
“Well, there are barely any users of Windows Phone.”
Leo: I guess that's why my question is, is there any hope
of true openness?
Kevin: Well, the interesting thing is going to be how Google
is with AOSP and I think they're going to have to start supporting writing
their stuff in AOSP. But they could lose a lot of market share which is – you
know, to some extent, they're already turning a blind eye to that in China in
the same way that Microsoft used to turn a blind eye to Windows being prided in
China.
Leo: Did you say – it seems to me, everyone will be a bully
the minute they can be the biggest kid in class.
Matthew: I mean, that's always been
the case. I mean, if you look at -
Leo: Well, I guess it has.
Matthew: I can't remember if it was Tim Woo or -
Leo: But didn't you see My Bodyguard? The big kid turned –
never mind. That's a movie.
Matthew: But I mean, if you're the
underdog, it's in your interest to be open. Because you're trying to afford -
Leo: Right, only as long as you're the underdog.
Matthew: Exactly. As soon as you stop – and in some cases,
Google is very open with all sorts of things. But it's not open at all when it
comes to Search and the algorithm because that's its core. You know, it has to build
a moat around that and keep people away from it. But it'll be open with other
things that aren't as relevant. The thing that I'm sort of wondering about is,
it was one thing when we were talking about hardware, so a phone, you know, a laptop. Does your phone work with your laptop?
Can you – even OneDrive or those sorts of things, that's one thing. But if
you're talking about things like Cortana, and Siri and Google Now, they're
basically ingesting all the data about you that they possibly can from every
point of view. Where your phone is, are you in traffic, have you turned on
voice recognition, are they reading your email, are they reading your
voicemail? So once they get access to all of that, I think that the idea that
it's a silo – [crosstalk]
Leo: “And we're going to protect you.”
Matthew: - locked into them, that becomes more of an issue than if it's just a piece of hardware.
Leo: That's an important thing to take away from this,
which is, these companies may currently say, “Oh, we'll keep it private. Trust
us.” But there really isn't a financial incentive for them to do that in the
long run. At some point, you can assume they'll give up.
Kevin: But there is. There is – if they have a huge breach,
it's going to be very bad for them.
Leo: Well, there's a breach and there's them saying, “We've
changed our privacy policy.”
Matthew: I would love to see Google say something like, “You
can take all the data we have from Google Now and from – [crosstalk]
Leo: And delete it from our servers, yes.
Matthew: “Or you can take it from us. We'll package it and give
it to you to go to a different service. They'll ingest it.”
Leo: Wouldn't that be great? But is there any incentive for
them to do that? There's none.
Kevin: You can get a lot of the data out with take out, but
what you don't get is the inferences.
Leo: You can't then apply it to Cortana or Siri.
Matthew: I can't get all my GPS coordinates from using Ways for
the last two years.
Leo: There's no standard for that, anyway.
Kevin: You can get all your – if you
turned on Google Location History, you can get all of that. I downloaded and
it's huge.
Leo: I think it's a KML file.
Kevin: I'll say, it was like a gig.
Leo: Then what do you do with it? It's not like Siri or Cortana could make
sense of it.
Matthew: It's not for free – [crosstalk]
Leo: So I guess my question is, Kevin, is there any way to
convince these companies who are acting in their best interests financially –
is there any way to convince them that it's in their best interest to keep it
open? Isn't it always the case that from an economic point of
view, once you're big enough, create a silo? Create a walled garden?
Kevin: That's just it. The challenge is them realizing that
isn't true.
Leo: It's not true?
Kevin: It isn't true. If you do create the silo, then you
create fragility and you're storing up massive, massive problems. You don't see
it until suddenly, the site goes -
Matthew: It's too late.
Kevin: Yes, and this is the site death's problem. “Oh, this
is great. They're looking after my stuff. I don't have to think about it. Wait.
Where did it go?” Which is exactly what happened to me with iDisk. It was like, great. Apple is a big company, I'm paying them $100 a year to keep my stuff on
their servers. I'll be fine. I can make links to it over the web and do that.
Then one day, they're like, “Oh, we're not doing that any more. You can
download it if you're quick and all those links are now dead. Homepage.mac.com
doesn't exist anymore and screw you.” That isn't some little Friday night
startup. That's Apple. Apple, not for lack of money, just
from corporate decision making.
Matthew: I think it was Anil Dash said his iCloud, all of his
photos, his wedding, his child, all those photos,
gone. So iPhoto upgrade, library not found.
Leo: Horrible. That's just bad customer service. There's no
– that's not what we're talking about here. A sensible company would not cheese
off its most devoted users by deleting their -
Kevin: This is the problem that in the long term argument for
open versus closed -
Leo: Yes, because you take it with you. Yes. You own it.
Kevin: The fragility of something is not just, is it well
designed? It's also the corporate structures that own it and contain it. So
when, if you build something that's on the platform owned by a large company
that they can take away, then you're building it on sand because they can take
it away. They've all done this. They've all deleted things. Microsoft is
actually better at this than most of them because they have been a developer
platform for a long time and they tend to keep things going for longer than
you'd expect. But it's, you know. Google has deleted all kinds of things that
we've got stuff on. Apple has taken things away, too. I'm sure Amazon at the
moment hasn't done that with AWS at all. But its business is that, there's the
potential that they could do that. They have done things like delete stuff for
political reasons, like they took the WikiLeaks stuff down, for example.
Leo: Is it conceivable that the open source community could
come up with a response to this and create an alternative to this?
Matthew: I think they're trying. There's
certainly open Cloud initiatives.
Leo: I think it's interesting that you use Linux on your
desktop, for instance, Matthew.
Matthew: Yes, and one of the reasons I wanted to do that is I
wanted to explore as much as possible of a more open community and a sort of
more open ecosystem. The problem is, it's still so
much work. You spend a lot of time – you know, it's like building your own Kit
car or something. It's great but you spend all this time maintaining it because
no one could figure out how it works.
Leo: 2015 will be the year for desktop Linux, I'm
convinced. No, I'm joking. I think that ship has sailed.
Matthew: I mean, I switched to Mint
because Ubuntu was becoming such a crapped up piece of garbage.
Leo: Mint is loaded with all the stuff, you know. The drivers.
Kevin: We have mobile Linux. Android is Linux. It's hidden
but it actually is the biggest operating system in the world now.
Matthew: I mean, Mac is Linux.
Kevin: Chrome is Ubuntu underneath, isn't it? So there is a
sizable set of computers out there running desktop Linux. It's just that they
call it – [crosstalk]
Leo: Well, what does that mean from the point of view of
what I was saying? Can the open source community create a credible alternative?
Android is not a credible alternative to the walled garden.
Matthew: The problem is that users in general do not care
whether something is open.
Kevin: And AOSP is a credible alternative to Windows Phone.
Matthew: Right. So the vast majority of users just want
something that works. They don't really care who's ecosystem, they don't really care what standard it operates on. So if you can
provide a totally siloed ecosystem experience and it
works great, which is what Apple has done, lots of people are going to say,
“Fine, I don't care that it's not open.” So how do you make them care? You
can't, really.
Leo: Well, maybe they care when their data gets deleted and
next time, they care.
Matthew: If something bad happens, and I think if you look at
companies – [crosstalk]
Kevin: [crosstalk] – keep a list of all the things that have
gone wrong, right?
Leo: But there are a lot of people – maybe it's Stockholm Syndrome. But there are a lot of Apple users who have been
bitten by Apple's crap Cloud services for years who still – most still embrace
it, use it and say, “I'm loving this.” So again, I
raise the question. Is there any way that – I mean, maybe we should raise the white
flag on the idea of a cross-platform nobody owns it, non-proprietary solution.
I'm not talking Cloud, just for everything as an alternative to the walled
gardens offered by Microsoft, Apple and Google can survive.
Matthew: The thing is, I mean, Kevin knows better than anyone
that history of sort of open things like that, open ID, and open this, and open
Cloud and they often get a certain amount of traction. But they don't get sort
of mass acceptance. They just don't.
Leo: So forget it. We should give up.
Kevin: Well, no. The thing is that they -
Leo: Go read a book.
Kevin: They do and they don't. The successful ones disappear
and you forget, you don't think about them anymore. We don't think about, oh,
open Web. But we're running open source browsers on open standards, HTML and
etc.
Matthew: That's the best example.
Kevin: Open source servers as well, primarily. So there's a
huge layer of stuff that's gone there and the point is – the point I was making
is that the original thing about open source was, it's
great for programmers. You don't have to write the codes that somebody else has
already written and then there was this interesting interaction which was that
if you managed to convince your company to open source software, then it was
great for you as a programmer. Because you could leave the
company and still work on it. It was like, “Great, wow. That's a real
advantage for me as a programmer.” So programmers like that because it meant
they didn't lose their code libraries and things, their knowledge base, when
they leave companies. That helped contribute to the – [crosstalk]
Leo: That's interesting, yes.
Kevin: But then the thing that the companies at first were
like, “Wait, we're losing all this precious IP.” Then they realized, “Wait.
You're leaving the company and you're still working on this code that we're
running but we don't have to pay you anymore.”
Leo: For free!
Kevin: “Oh, that works for us too.”
Matthew: “What a great idea.”
Kevin: “What a great idea. Let's do more of that.” To the
point where we end up with stuff like Heartbleed because everybody's running
the same code base and no one's done a security order on it in ten years
because the thing is all outsourced to people who were – [crosstalk]
Matthew: They forgot they were even running it.
Leo: “Somebody already wrote this, so let's just use it.”
Kevin: Right, but you know, that's
usually a good thing. When it becomes a monoculture risk, it can be a problem.
But it's generally less of a monoculture risk than everybody running
Microsoft's code, or everybody running Apple's code or everybody running
Google's code.
Matthew: It feels to me like the web was almost an accident.
You know, that no one realized -
Leo: A brief utopian period.
Matthew: Well, no one realized how important it was going to be
so no one cared.
Leo: Isn't that what Tim Woo says? You mentioned Tim Woo,
the author of The Master Switch.
That's kind of what he says, is that in the nascent period of any of these
technologies, and he starts with the long distance system, it is free, and open
and magical because nobody's making any money on it. As soon as the big
companies come along and they take it over, it goes from proprietary to closed. Now, he says that it swings, right?
Kevin: Also, they do – [crosstalk]
Matthew: It's always a pendulum.
Kevin: - is takeover as well. The other thing that he
describes is that part where they go and get the government to mandate them and
probably competitors as well.
Leo: It becomes the incumbents.
Kevin: And that creates the innovation pressure for the new
things to occur.
Leo: I guess that's what I was asking. Is there space for
an innovative – maybe it's open source, but maybe it's somewhere else. An
innovative solution that isn't a walled garden, that does allow users to take
their stuff and move it somewhere else, too.
Matthew: I think there always is.
Leo: I hope so.
Matthew: In fact, what we may be seeing – I think what I took
away from Tim's book, anyway, is that the pendulum often swings or shifts when something new happens. So the Web is a good
example. Is there something new enough? Is it going to be virtual reality? Is
it going to be the Cloud? Is it going to be automated assistance like Siri and
so on that will kind of – someone will come along at the same time as people
are starting to feel locked into these silos and the two will find each other
the need or the desire to have your stuff somewhere else or to be outside this
walled garden will meet someone who has an offering that's powerful enough.
Leo: We're talking about Google, the Cloud, everything to
do with up there in the sky. Matthew Ingram of GigaOm, Kevin Marks of Kevin Marks. You note that
Jeff Jarvis isn't here. He is in Davos enjoying a lovely dinner right about now
and he'll be back next week. But, hey, we've got two great people to take his
place this week. Somebody – where did we find this great Hyperland video?
Kevin: Oh, I threw that in.
Leo: That is lovely.
Kevin: So I wanted to make the point, at one part of
Microsoft's thing was, they said, “Look, and NASA is
using our VR stuff to explore Mars.” Then, Hyperland is a documentary made by colleagues of mine at BBC in 1989 and – in 1990. It's
written by Douglas Adams and he's the guy from NASA doing this VR thing.
Leo: So Douglas Adams of the Hitchhiker's Guide – and you
may recognize Doctor Who.
Kevin: Yes, Tom Baker who was 81 yesterday appears much more sprightly in this as it's -
Leo: The fourth Doctor Who. So this is from what year?
Kevin: '89.
Leo: Wow. This is NASA using virtual reality.
[video plays]
One for each eye. I've worn these. I remember going to SIGGRAPH.
Remember fighting the dragon at SIGGRAPH with that helmet on?
[video continues]
Oh, jeez. It weighs about eight pounds.
Matthew: I tried one in -
Kevin: He says it's light.
Matthew: I played a sort of 8-bit video game and just about
threw up.
Leo: But this was Jaron Lanier. This was the early days of
VR, yes.
Kevin: Oh, and W Industries, you remember that? The pterodactyl game?
Leo: Yes, the pterodactyl. That's not the dragon, the
pterodactyl.
Kevin: The founder of that is a friend of mine, he's here
now. I saw him over Christmas. He said, “Oh, we're still doing interesting
things in the space. Keep your eyes open for VR and AR.” So I suspect he's been
doing a lot of work on the optics of it. One of the challenges is not having
these big opaque screens that are across your face, but doing the thing that
Microsoft showed up today where you've got some glass in front of your face
that is about as opaque as the sunglasses I'm wearing but is able to inject
light from multiple directions. So you can get a realistic overlay on the
outside world.
Leo: I always thought that was the best way to do it.
Kevin: But there's a bunch of really hard things to do with
that because if you do the classic headset thing like the NASA guy is wearing
there, or like Oculus, what you're doing is putting screens there and using
lenses to put them at infinity and then your eyes have to relax and look at
infinity. So you're creating a tension between your eyes sense of where it
should be looking, because it assumes that it should be focusing close up at
things when it's converging and so on. So your eye ends up with getting
muscular strain as it's trying to resolve those things. You often end up with a
headache from that.
Leo: Well, if Oculus Rift makes you nauseous because of its
poor frame rate, this thing. The frame rate on this is horrific.
Kevin: Oh, this stuff was very slow. But the other thing was
that what they're able to do now with optics is they're able to buy having
multiple lenses and multiple light sources from actually producing light that
is produced at different distances. So it's like the – what's the camera called
that takes the photograph that's always in focus? It's that same technology of
light pipes and having lots of different distances. They're starting to be able
to construct glasses that can actually give you an overlay that isn't going to
give you power lapse errors because of that problem. So I'm not sure how much
of that Microsoft had in what they showed today, but I think they had some of
it. It's some very subtle and deep optic stuff going on there. So there's -
Leo: Here's the fourth – oh no, that's not Tom Baker.
There, is that him?
Kevin: Yes.
Leo: The fourth Doctor Who.
Kevin: Tom Baker in this was the Cortana. He's your
intelligent assistant. If this is up on Vimeo, just search for Hyperland. It's worth watching because it gets huge amounts
– part of it is a history lesson of how hypertext happened and they interviewed
Ted Nelson and show a bunch of stuff. But a chunk of it then comes into
speculation of what might happen and it's fascinating, because the Web isn't in
it. It's all these things that aren't quite the Web. They didn't find Tim
Berners-Lee just before the Web happened with all the parallel universes around
there. It ends with a lot of VR speculation which is now, as we come to
reality, the stuff catches up. So if you've got an hour, I would recommend
watching Hyperland.
Leo: I remember seeing it way back when. This is what's so
interesting about technology is kind of the discontinuity of inventions. You
can predict things like VR but you cannot tell when it will happen. And you can
say, “Oh, hyperlinking might make a lot of sense.” But you can't see how
exactly it will emerge. This movie was made exactly as Tim Berners-Lee starts
to create the first World Wide Web.
Matthew: Exactly, and I remember I was writing about business
and the stock marker for a newspaper here and I was writing about the big cable
companies trying to build their own internets. So they saw the whole thing, you
know, people buying TV shows, and people downloading movies and people
listening to music. But they each wanted to build their own system. So, like
the one in France, where you'd have to have a specific terminal and it would be
run by your cable company. So the exact opposite of sort of
the open internet. But they were building them. They were spending
billions of dollars to build these things and luckily, you know, the web came
along and it got traction before those things did or we'd be right where we are
now. Except it would be, you know, CompuServe instead of Google Cloud.
Leo: There are many, many universes all happening at the
same time and we're just picking one to be in today. Matthew
Ingram from GigaOm, Kevin Marks. More to come with This Week in Google. Our show today
brought to you by my mattress. Hey, you know. When you get a great mattress,
you got to tell the people. You got to tell the world. My
mattress, Casper. If you don't believe it's my mattress, we've got a
video to prove it. Casper is an online retailer of premium mattresses for a
fraction of the costs because they don't bother with show rooms. They ship you
the mattress. I was talking to someone the other day, so actually, we had a
king mattress in our new apartment. We couldn't get it in the door, it was too
big. So we had to get two twins, but at Casper, this is a queen size mattress
in the box the size of I don't know. It's like one of those
little dorm room refrigerators. You open it up, you take it out, it goes
(whoosh) and it's amazing. It is the best, most comfortable mattress you've
ever had by marrying two technologies, latex and memory foam. You get a
mattress with just the right sink and bounce. It's firm but soft. I don't know
how they do it, it's kind of magic. Plus, you could buy it online easily.
Now you could say, “Wait a minute, how can I lie on it
first?” They actually had a booth at CES but you don't have to go to CES.
Mattresses, five minutes on a mattress in a show room is not going to tell you
whether that mattress is going to be good for a long time. Casper offers free
delivery and painless returns. They'll come and get it and you have 100 days to
decide. Not five minutes on a showroom floor but 100 days so you can figure out
if this is the right mattress for you and they'll take it back, no questions ask. They'll come and get it, you don't pay for
anything. Casper mattress, try sleeping on one today. Just visit casper.com/twig. About $500 for a twin, $950 for a king size and you
save an additional $50 if you use the promo code TWIG on checkout. $50 off any mattress at casper.com/twig.
Yet one more company that's changing the way we shop by
offering a great service online, and eliminating the showroom and saving you
lots of money. Why pay for that showroom? You don't need to, casper.com/twig.
Don't forget to use TWIG at checkout.
Matthew: Was that you jumping on the mattress?
Leo: That's me. I sleep on a Casper. I love my Casper. In
fact, I liked it so much I got it for my son, who, you know, he's in a dorm room. Actually, they rent a house but there's like upstairs and
he needed a mattress and I said, “Well, wait a minute. I'll just have Casper
send you one.” He loves it.
Matthew: I noticed you were wearing the fancy pajamas, not the
SpongeBob pants.
Leo: Oh, you saw me in the SpongeBob jammies? Next time
I'll use those. “Yes, those are my Christian Howell the Third jammies.” I
actually got those – I don't know if it's worth telling the story. But I got
those. I did a commercial many, many years ago for Clark's Shoes, a TV
commercial, in which I had to hang from a bar and drop into a bed. They were
showing how comfortable Clark Shoes were. They shot it with – it was still shooting
in video, that's how long ago it was, in super slo-mo
camera and then the bed would collapse under me and they had all this smoke and
dust would come up. I did about 20 takes but they got me special jammies, that's them. Still got the jammies. Some day I'll find that Clark's shoe ad. Thanks to
YouTube, it probably exists.
Moving on. We'll talk more about Microsoft. I think there's a lot
more to say. It was a very, very significant, I think, announcement today from
Microsoft. But of course, it's very easy with these to get kind of sucked into
the reality distortion field. So Paul and Mary Jo will join us just a little
later on this afternoon. They're in the briefings right now and will have more
to tell us about Microsoft's plans for the future.
President Obama gave his State of the Union last night
and there was a lot of concern about what he would say about hacking. They had
telegraphed that they were going to strengthen the hacking laws using the RICO
Act and racketeering laws to make it really problematic for hackers. He
actually did talk about cyber security in the State of the Union. “If we don't
act,” he said, “We'll leave our nation and economy vulnerable. If we do, we can
continue to protect the technologies that have unleashed untold opportunities,
blah, blah, blah.” New cyber security legislation,
which at this point, I think after the Sony attack, might be easy to push
through. Maybe that's why the White House and the FBI were so anxious to pin it
on North Korea. It makes it a lot easier for them to get this legislation
through. “No foreign nation, no hacker, should be able to shut down our
networks, steal our trade secrets or invade the privacy of American families!”
This is when you know this is BS, “Especially our kids.”
Matthew: Right? As soon as they mention the kids, you know.
Leo: “Consider the children! Hacking takes such a toll on
our children.” That's when you know that they're trying to pull the wool over
your eyes. So they do want to raise the punishment for hacking. They do want to
bring racketeering into this, which means that you could, for instance, have
all of your personal effects confiscated by law enforcement even without being
charged of a crime simply because you say, “We're in a hacking chat room.” I
think nobody's going to oppose the idea of protecting American soil against
the, you know, evil hackers but this will be interesting to see what the
response is to this. Did you have any thoughts? None at all,
huh? Just me, sitting here. Consider the
children.
Matthew: I think Kevin's talking.
Leo: Kevin's talking but nothing's coming out of his mouth.
Kevin: - because, yes, I'm playing whatever again. Sorry. So
this is more of the – they want two opposing things which is the problem. This
is the problem of all security.
Leo: It's like Cameron.
Kevin: It's exactly like Cameron except Cameron – to some
extent, both Obama and Cameron are electioneering now. Obama can't really get
much passed so he's able to go back to campaigning, which he's very good at and
throwing out ideas, saying, “Well, here's these great things to do but these
guys will stop you doing it.” So he's able to go back to proposing things that
are slightly less practical but he is still pushing some of these, you know, security state agenda things. Cameron – [crosstalk]
Leo: Well, CISPA, for instance -
Kevin: - has got an election coming up in May. CISPA – hang
on. Let me finish.
Leo: Go ahead, do Cameron. You've got a British accent, you
should do Cameron.
Kevin: So Cameron is -
Leo: He's your Prime Minister, not mine!
Matthew: So to speak.
Kevin: He's campaigning to his base. So he is – the
conservatives campaign on law and order so they're saying, “We should not let
these terrorists have any ways of communicating that we cannot intercept,” even
though he is being briefed by people who understand this properly and know
what's going on. But it doesn't matter too much if he can write the song by it.
Matthew: I don't even think he believes half the stuff he says.
Kevin: I'm not sure that's true. I think he believes some of
that stuff but also, he's saying stuff that will play well to the conservative
base, which is law and order, terrible foreigners and
things. There's a sort of set of tropes that come out there. Whereas, Obama's
doing the opposite and doing the things that appeal to his base and wind up his
opposition, so there's a bit of politicking there but there is some underlying,
you know, problems here that are actually very hard, which is, how do you
decide between the interests of the people who want to keep stuff private and
the interests of law and order in trying to see what you have? The answer is
you have due process. You have the notion of the poisoned tree. You have the
Fourth Amendment. You have the Bill of Rights. In the UK, there's the whole
European Human Rights doing all this. This upsets the professional spies
because they're – you know, they believe they are above the law because
historically, they have been. So there's this tension where you get strange
things being defended by groups because they're being briefed by the spies
saying, “We have to do these terrible things to keep us safe.” We have to
intercept communications. You know, they can point to, you know, Turing in the
Imitation Game and say, “Look, if we hadn't cracked the codes, we wouldn't have
won the war,” but that in some ways – and that's broadly true. But in some
ways, that's also like this sort of scary mainstream propagandizing for torture
that we've had in US media for the last ten years, 15 years, as we have programs
like 24 and – well, too many to mention. At some point, they use the plot
device of, “This person knows something. We have to torture him until he tells
us it.”
Matthew: Zero Dark 30.
Kevin: Yes, but it's just like Alias that did it. It's not
just the dark ones. It's like this is a mainstream trope in police
investigations stuff as well. So there is this problem that it is very easy to
write these good and evil narratives and say, “If they're the bad guys, then
anything we do is justified,” and that is what draws this in. It worries me
when the politicians are dancing along that line as well but in particular, it
worries me for the encryption stuff. Because in those cases,
it's less clear cut than the torture stuff. Torture is fairly obviously
appalling and to some extent, they understand they're defending something
completely appalling and therefore, they have to assume the lack of humanity in
the other to do this. So the American Sniper debate at the moment is literally
about Clint Eastwood making a film that presents all Arab people as subhuman
and therefore, you think this guy is justified in shooting them on site. That's
pretty much the message of the film. It's a well made film and lots of people like it. That is not that different a message from mainstream,
you know, war film, police film, whatever. You have the good guys who are
somehow impervious to bullets but they can kill people with one shot. You don't
think about that because they're the bad guys. It is a – you know, a video game
worldview that plays out as a media trope as well. That is a more worrying one.
The encryption one is harder to explain to people
because it doesn't map to their experience as much. People do understand that
they want their own stuff to be secure but they do want the police to be able
to keep an eye on the terrorists. That is going to be very hard to explain
this, particularly as we get to the point where they're saying, “You have to
provide a way for us to read these messages.” The companies say, “We've
designed it so we don't have any way to read these messages. That was the whole
point of that software update. That's why What's App
and iMessage can no longer read the contents because
we were sick of having to give them to you lot.” It's doing the exact thing we
were talking about in the first strike, which is damaging the customers'
ability to have faith in that the company is looking after the stuff.
Leo: I got to admit, if I'm in the NSA and all of the
sudden, the terrorists are using What's App to communicate and I can't see it,
that's of concern, right? I can understand why they'd be worried about that.
Admittedly, I'm sure the people at the NSA understand the implications of what
Cameron called for. But admittedly, it doesn't seem an unreasonable thing to say
that we should be able to see into What's App and, if
you said, there's due process. So maybe I have to get a search warrant and, you
know, go to the FISA court and say, “We think that there are two people having
a conversation, planning an attack. They're using What's App. I want to be able to see what they're doing.” What's App maintains a
special key in escrow, something like Skipjack where given due process, given a
warrant, they can hand that key over to government who can then read those
transactions. Is that unreasonable?
Matthew: Part of the problem is that the FISA court and all
those deliberations are completely secret. They apparently
routinely rubber stamp things or whatever. Is that due process?
Leo: But they have to be secret or the terrorists will know
what we're doing. By the way, if you haven't figured this
out, playing devil's advocate here. Go ahead.
Matthew: I think there is a pretty good argument against sort
of mass collection of data and surveillance that is not targeted at anyone in
particular. Because there's some evidence, at least, to show that is almost
completely useless.
Leo: Great article in the New Yorker this week along this
line called “The Whole Haystack” by Matthias Schwarz, in which he makes a very
strong case that in fact, there is no value to mass
collection of data. You cannot yet use that data in any way that will help you.
What he uses as an example is the fact that we have already known almost every
major terrorist attack in the last 15 years. It's been committed by people we
already know about and are already watching.
Matthew: In fact, Paris is a good example. Those people were
known to police, known to the authorities and yet, still, it couldn't be
prevented.
Leo: The size of – the lack of data is not the problem.
Matthew: Exactly. It's filtering it and understanding it. We
just don't have the tools. I think what the NSA is hoping is that they can just
accumulate this data and then eventually, they'll figure out how to actually go
through it and find the bad guys. But at least to this point, there is no
evidence that has occurred.
Leo: Right. You'll hear a quote often from Congress or
perhaps law enforcement saying, “There have been 54 cases – at least 50 threats
that have been averted because of this mass collection of data.” President
Obama said that at a press conference. He says lives have been saved but in
this article, Matthias goes through those 54 cases and he quotes Patrick Leahy,
senator for Vermont who says that number is plainly wrong. These weren't all
plots and they weren't all thwarted. According to a deputy of – CIA director
Alexander, there's really only one example of a case where bulk phone records
stopped terrorist activity. One. So that was -
Kevin: Well, also there's found that most of these stopped
terrorist plots in the last few years have actually been organized by the FBI.
Leo: Not the NSA.
Kevin: But no, there was an Intercept article about this
which basically said, “Yes, they've stopped all these terrorist plots except
that these terrorist plots involved an FBI informant finding some loner and
offering him a quarter of a million dollars to go and blow something up, and
then arresting him before he blew anything up and saying, “Look, we stopped a
plot.” It was like, I think you actually started a plot and then stopped it as
well.
Matthew: See, that's progress.
Kevin: This is what happens when you get credit for stopping
plots. Like we haven't had enough plots? We'd better find some.
Leo: Let's find some plots and stop them.
Matthew: I just don't see how – even if there were a dozen
cases, which there are not. So let's say there's five.
Even if there were, does that justify the kind of routine surveillance and
breach of -
Leo: Billions.
Matthew: - human rights that we're talking about.
Leo: Billions.
Matthew: On a scale that's unprecedented.
Leo: General Alexander is the head of the NSA – or former
head of the NSA, by the way, not the CIA> He called it, and this is the name
of the article, “The whole haystack,” which is basically all electronic
conversations going on around – not just in the United States, everywhere, at
all times, forever.
Matthew: Every time I think about the NSA's bulk collection, I
think of that scene in Indiana Jones when they take the -
Leo: The warehouse, yes.
Matthew: Yes, and they put it in a box and they hide it in a
government warehouse where no one will ever find it.
Leo: That's where the data is. Legislating
for unicorns. I don't know what this is about. I've got to read this
here real quickly. Oh, this is the idea of – this is good. Should we play this
video? You must have put this in here, Kevin Marks. This is from Parliament.
[video plays]
Oops, sorry. Let me get out of … Wait, let's get that.
Those are the big four, by the way. Let's go once more and listen to those.
Which countries are doing this spanning encryption?
Matthew: Did he say Moldova?
Leo: I think he did.
Matthew: Boulder. Is that a country?
[video continues]
Leo: I love Parliament. That fellow, by the way, Julian Hupert, is a member of Parliament
from I don't know where.
Matthew: I thought that was Louis CK for a second.
Kevin: Brighton, I think.
Matthew: Well, that explains a lot.
Leo: And that woman is Theresa May, she's the home
secretary. The woman who, in fact, would according to
David Cameron, be signing the warrants that would allow them to crack into
encryption.
Matthew: I mean, the problem is, when she says, “There will be
no safe place for terrorists to communicate,” that means there will be no safe
place for anyone.
Leo: For you and me to communicate!
Matthew: Because how do you differentiate?
Leo: Go ahead then, Kevin.
Kevin: Right. That came from the post that described this as
legislating for unicorns. They said, “Well, you can say there will be no safe
place for terrorists but that doesn't mean you can actually write legislation
that does this. You can say, 'we will introduce legislation that will make sure
we have plenty of unicorns in the UK,' but it doesn't mean you can actually
make it happen.” That's effectually what they're promising. What they – so
they're being a little cynical about this. They've come up with this nice
formula of, “There can be no safe space for terrorists.”
Leo: They know, they must know.
Kevin: But if you actually want that to be true, you need to
do a lot of that through physical surveillance as well as through interception.
To say that we can just – to say that they can do it purely through
interception is nonsense. But what you have to do then is follow that up with
detailed surveillance and, you know, planting cameras on people and those kinds
of things, which is also obviously intrusive and requires warrants and things
like that. Part of the sort of reason the NSA and GTSQ [?] have such enormous
budgets is they've represented the politicians and they don't need to do that any more because they can sniff all this traffic and do
this without doing that. It should be cheaper if they just invest all this
money in building these giant haystacks so they can sniff for needles.
Leo: What about President Obama's proposals? Is that also
legislating for unicorns?
Kevin: I haven't – has he introduced a legislation proposal?
I hadn't seen one.
Leo: Well, of course, CISPA's back, whether CISPA has a
chance of passing is unknown, but that's the bill that has failed Congress
several times that proposes a sharing of data from the NSA with companies like
Google and Microsoft so that Google and Microsoft could get right in there and
help them fight terrorism. That has been turned down several times due to lack
of privacy protection.
Kevin: Right and the equation has been called the “snooper's
charter” -
Leo: Much better name.
Kevin: That's not the official name but that's what it's been
successfully labeled the last five times they've tried introducing it to
Parliament over two different governments.
Leo: It did sound like Obama was endorsing it. He endorsed
enabling better communication between the government and companies about cyber
threats. That sounds like CISPA. He had said before, “I will veto this if it
passes.” It never did pass but perhaps he's changed his tune.
Kevin: The cyber threats – there's a bit of a dance around
that one. Because the cyber threat stuff, the better
communications in government about cyber threats was a separate program where
they were talking about data breaches and information pooling about that. That was where that stuff came from. So they may be blurring lines there
deliberately to make that, use that formula. I've been to a few meetings and
things about that and the idea there was, there was this problem that companies have information about data breaches and advanced
system threats and those kinds of things. It would be useful to share this with
other companies, but they didn't want to share with competitors. Can the
government by the broker for relaying information back and forth? Which actually ignores how security really works, which is, there
is a network which shares that which is all the security ops guys at your
company who know all the security ops guys at the other companies. When
some of that stuff shows up, they call their mates and say, “Look out for this
one.” [crosstalk]
Leo: I think it's just fascinating that Obama -
Kevin: [crosstalk] - won't actually work and that's, you
know. The way the internet has been secured reasonably well so far has been
through those kinds of back channels and security communities and risks,
account risks and all those. The security community that we
know and love. It's very hard to reconstruct that in relationship to
reform. The government people know that too. They employ people in that group
of people as well. So there's this set of people who understand that and a set
of people who are trying to build a management structure on top of it that
makes it – you know, historically it's been very hard to do it successfully. If
you start trying to tell your security people how to share things with people,
you're going to have a bad day because you will lose the better security
engineers if you tell them they can't talk with people outside the company.
Leo: What's really amazing is this reminds me of 1984's –
what do they call it? True speak, where truth is lies, war is peace. Obama says, “In order to protect the privacy
of American families, we have to snoop.” It's just like, “Oh, no, that's the
opposite!” It's like nobody's listening.
Matthew: We've always been at war with Eurasia.
Leo: Exactly! “In order to protect, no foreign nation, no
hacker, should be able to shut down our networks, steal our trade secrets or
invade the privacy of American families.”
Matthew: “We're going to do that.”
Leo: “That's our job.”
Matthew: Right. If anybody's going to invade anybody's privacy,
it's going to be us.
Leo: I just – (squeaks)
Kevin: Also, the other thing is the shout out about American
persons – it's not American persons. There's all the legislative stuff that
says, “Oh, we must protect, because of the Fourth Amendment, we can't snoop on
American persons.” Which, you know, has a slightly generous
definition. It includes me despite my lack of citizenship because I live
here and so on. But they interpreted that to say, “If we can assume that more
than half of this conversation is not on American persons, we are allowed to
snoop on it. Then we'll clear the whole thing anyway.” So that was how they
danced their way through that at the NSA.
Matthew: And they're also allowed to use a sort of best guess
as to whether someone is a foreign citizen. So if you mention Saudi Arabia,
boom.
Leo: Boom, you must be a foreign citizen. Well, I'm sitting
here with two foreign citizens. This entire show is being recorded by the NSA
for future reference because clearly, I'm communicating with foreigners.
Matthew: Luckily, the NSA is collecting the non-US part and the
Canadian and British governments are collecting the non-US part. So it all
works out.
Kevin: Right, and then they have an information sharing
agreement, if that's true. The US can collect information on non-US citizens,
and the Britains can collect on non-British citizens
at an age that British citizens do because they don't have a Fourth Amendment
there. But they have an information sharing agreement so they can share the
information that they have on each other's citizens with each other and join
the dots up. So that's how they circumvent that particular -
Leo: It reminds me of a great novel and later Sean Connery
movie called The Anderson Tapes. Sean Connery creates a plan to rob every
apartment in a large, wealthy apartment building and as soon as the robbery
happens, all the law enforcement agencies that have been tapping every phone
immediately erases all records of this thing being
planned. “I don't even know. We were listening but we don't want to have
anything to do with it.”
I should point out that there's an even more sinister nexus, that between Hollywood and content creators who would
also like to use these laws to protect copyright. There's been no attempt by
the federal government to separate the interest of fighting terrorism with the
interest of fighting copyright. When they talk about hackers,
that's just as much people stealing HBO Go passwords.
Matthew: And we've seen that from SOPA and PIPA and even
before. Copyright maximalists and sort of content – massive content empires are
more than happy to cozy up to whatever it takes to get sort of strong
legislation passed. If they want to kind of – fear about terrorists and so on,
because that's what it takes, they're more than happy to do that.
Leo: They want to add racketeering, any act which is
indictable on any of the following provisions, etc., etc. It means that, you
know, if you share your HBO password with somebody else, you can get ten years
and they can take your house. It merely relies – at this point, we're merely relying on the goodwill, on the kindness of the court system, to
keep that from happening.
Matthew: And if you breach the Terms of Use of a website, you
know, that's effectively hacking.
Leo: Right. “Such persons' interest in a property, real or
personal, that was used or intended to be used to commit or facilitate the
commission of such a violation, any property, real or personal, constituted or
derived from any gross proceeds, all this goes to law enforcement.” In fact,
even without an indictment if you're a co-conspirator. Unbelievable,
unbelievable. So this is the proposals the White House has put forward
to protect us and our privacy against them foreigners. While we're at it, let's
just shut down all the bit torrents too.
Matthew: It's going to be the same as before, it was if you had
tools that would allow you to break encryption on a CD or DVD,
that was a crime. If you have encryption tools that will allow you to
protect your computer from snooping by the government, that will be a crime.
Kevin: I think that is a crime. That's section 12.01. Cory
Doctorow just announced that he's going back to the EFF to try to get that
repealed. That was on today, saw that.
Leo: Good, good man.
Kevin: So that's encouraging because he's -
Leo: He was EFF's international ambassador for goodwill or whatever, I don't know what his title is.
Kevin: He basically would go to all these DRM meetings and
understand what was going on, and write it down and publish it. Also, gather
all these drafts and things that they were trying not to publish and publish
them. So that sort of was a big part of stopping this stuff being much worse
than it is.
Leo: “Cory Doctorow and the EFF aim to eradicate DRM in our
lifetime.”
Matthew: That's a good goal.
Leo: “We pledge, before this decade is out, to eradicate
DRM.” In fact, they even use a rocket ship, Apollo 12.01. They clearly
understand the reference. A mission to eradicate DRM in our
lifetime. Well, good on you, Cory. Cory is so well spoken, so smart and
I agree with him, absolutely, 100%. Although I have to admire, you know, I have
to point out that Larry Lessig, who for a long time,
this was his call, realized that nothing was every going to happen unless we fixed how we financed elections. He's focused
everything, all of his efforts now, on campaign finance reform, even creating a
PAC – this is brilliant. We've talked about this before, a PAC, political
action campaign, to raise money to lobby members of Congress to eliminate
lobbying and PACs. You got it, you know? It's the only way to do it.
Matthew: I sort of feel like Larry maybe fighting that one sort
of dragon was almost too easy so he decided to take on an even bigger one
that's even harder to kill.
Leo: It's not enough just to get rid of DRM. Let's get rid
of corruption entirely. But he does make a point, you know, as long as
Hollywood is spending money on these members of Congress, you're not going to
get anybody to do anything. It's great if you can add the children and
terrorism, everybody wins. All right, we're going to call this episode Legislating for Unicorns, I think. Let's take a break and
when we come back, more. We also have a mini change log, some new stuff from –
not much.
Jason: Not much and I scoured.
Leo: A short change log. I have a change log, but I'll save
that for my tool of the week. We're talking right now about Google, the Cloud
and the Google-verse with Kevin Marks, kevinmarks.com. On the Twitter, he's @kevinmarks – oh, he's a wonderful voice, a passionate voice
for open standards. And Matthew Ingram, who covers all of
this stuff so beautifully on gigaom.com. Always great
to have both of you guys on the show. Jeff Jarvis will be back next
week. He's in Davos, probably snoozing right about now. This is This Week in
Google.
Our show today brought to you by lynda.com, a nice way
to learn the stuff you need to know to get as smart as these guys. Millions of
people around the world use this, over 4500 courses with topics like web
development, photography, visual design, business, as well as software training
like Excel, Wordpress and Photoshop. All their
courses are not only taught by experts in the field but great teachers. It's a
nice combination. They're shot beautifully, they're
not like junky YouTube videos. They're really great. Plus, they have full
transcriptions so if you're looking at a course at lynda.com, you can see what
each chapter holds. You can jump right to a portion of the content, which makes
it very easy to get exactly the information you need. You're going to love it.
You can learn everywhere, too, on their iPad app – they just got a new iPad and
iPhone app. The Android app, it works with Chromecast, the Kindle app, so
really, there's no limitation. You can watch it everywhere you go. I'm taking a
course on the Swift programming language, the new Apple language.
That's the other thing about lynda.com, they're fast.
They come up with new stuff all the time. Up and running with the Google Cloud
platform, that's a course. Simple Android development tools, that's a course.
There's a weekly series called “Small Business Secrets,” that's Dave Krenshaw's great – he's like a small business coach with a
great course on accounting, marketing, management, things like that. I love
lynda.com and I think it's a great deal because you pay once for each month and
get everything, the run of the place, all 4500 courses and there are new
courses added each and every week. Hadoop Fundamentals, I wonder if Mary Jo
Foley has studied that.
I have to say, we've got a special deal that's pretty
good. We're going to give you ten days free when you go to lynda.com/twig.
You'll get unlimited access not only to every course on Lynda, including access
on your iOS and Android devices, but you'll also get new courses as they're
added each week. They're always adding new stuff. Lynda.com, you can learn it!
How much can you – there's Lynda. Oh no, I'm sorry. That's Lynn Langit, that's the Hadoop course. I wonder if they have a
course in micro formats. Lynda.com/twig free for ten days. There's no better
way to learn than lynda.com. We love her and we thank Lynda for their support.
Google wants to become – well, we know they want to
become an ISP. They've gotten a little more serious about Google Fiber in more
communities. Now, Google wants to become a wireless carrier. Exclusive
story from the Information. In fact, let me log in because I pay for the
Information so I can read the whole thing instead of just an excerpt.
Kevin: The second paragraph.
Leo: Yes, exactly. It's worth paying for if you're not
already doing that. Jessica Lessin and her team do a
great job. I asked her, said, “Aren't people going to do what I'm about to do,
which is subscribe and read the article out loud?” She said, “That's all right.
People will subscribe because they want to get access to it.” So, by the way,
Amir Afrati, who works at the Information, has such
great contacts at Google. He's often breaking scoops.
“Google plans to shake up the US wireless market,” he
writes, “By selling Voice plans to consumers directly from Google through
T-Mobile and Sprint.” So I guess to become an NVNO. It's the latest example on
how Google tries to prod incumbents to change their business to benefit Google.
That's what they do with Google Fiber to show you that it's possible to have a
gigabit ISP and make money on it. Google's expected to reach deals to buy
wholesale access to the carrier's mobile, voice and data networks. It's code named “Nova.” Nick Fox is leading it, a long time
Google executive and Amir says that launch this year seems likely.
Matthew: It makes sense in a lot of ways. I mean, I'm kind of
surprised they haven't tried it before now.
Leo: Disney's an NVNO. I mean, Virgin. There's so many NVNOs. It's an easy thing to do. One of our advertisers, Ting, is a
Sprint NVNO. It's an easy thing to do.
Kevin: I think there's a timing issue here because they
couldn't - at the point that they were trying to launch Android across all
carriers, they couldn't do this. Now where the carriers aren't going to drop
Android, they can do this.
Matthew: Yes, good point.
Leo: It's interesting that Sprint and T-Mobile are -
Kevin: [crosstalk] – with these carriers already, right?
Leo: Explain to me why Sprint and T-Mobile are willing to
have these third parties lease wholesale data and sell below their retail
prices, offer better plans.
Matthew: I think both of them are struggling. I don't know that
but it feels like it.
Leo: I guess they need to, huh?
Kevin: So there's a fixed cost and a variable cost. So the
fixed cost is maintaining the network and growing it, and they want to keep
doing that. Then there's a variable cost, which is running all the stores, and
signing people up, recruiting them and so on. So if they can offload some of
that variable cost stuff to somebody else, they're happy to do that because it
helps them with the fixed costs. That's basically how NVNO stuff works.
The other thing is, the reason
that it's those two is that they're the GSM ones. So they're the ones that's actually straightforward to operate with as opposed
to Verizon, which is weird as anything, and AT&T, which is not quite as
weird but still slightly odd. Whereas, GSM is like the global
standard. The reason you can force things out of phones and put
different ones in and have them work broadly is because they use GSM. Yes,
there are different bands and you have to have different radio frequencies and
this is actually slight – [crosstalk]
Leo: Sprint's not, though. Sprint's a CDMA carrier, so they
would have on GSM and one CDMA carrier.
Kevin: Oh, I thought Sprint was GSM.
Leo: No, that's what's interesting is that it's both. I
think that's probably why it is both, because I think that they want to be able
to say, “Whatever kind of phone you have, you can bring it over.”
Kevin: That makes some sense, yes.
Matthew: The risk, I mean, as a mere mention, Sprint and T-Mobile,
their networks are not as robust. They don't have as good coverage.
Leo: Yes, they're the also-ran. That's why they're willing
to do it.
Matthew: Exactly.
Leo: The reason that Google has held back on this, Amir
writes, is because they didn't want to upset wireless carriers. They needed
them to carry Android phones. Now it's not such a big deal because -
Kevin: I just said that.
Leo: You just said that? Sorry. I'll add – did you say
this? The other reason is anti-trust. I think if Google were to come along and
say, “Why don't we just buy T-Mobile?” They could do that but the likelihood of
government approval, I don't know. It might be an unknown. So this is a great
way to do it and as you point out, cleverly, in advance of my insight, that now
they don't have to worry about pissing off the other guy. You just said it so
eloquently I didn't understand you. “They didn't wanna piss off the other guys.” You know, I think I'd be tempted. It wouldn't even
have to be less expensive if I thought Google was going to deal with my more
fairly and have better customer service. But Google's not exactly known for
customer service, are they?
Matthew: No.
Kevin: Right, but it's a tradeoff, right? So I had a problem
with – I use T-Mobile because I think their customer service, their deals are
pretty good and their service is good. But I had a problem last week where my
phone stopped working on Wednesday, just as I was getting on the train to San
Francisco. So when I got to San Francisco on Wi-Fi, I'm tweeting them saying,
“My phone stopped working. What's wrong with it?” They're like, “Oh, try
turning it on and off again.” It's like, “Yes? I've tried that. I've also tried
taking the SIM out and putting it back in again.” It's not a physical flaw.
It's a network flaw. Then, after doing that with the customer service guy, they
say, “Oh, you're right. I'm going to have to transfer this to engineering.
They'll get back to you within 72 hours.” I was like, (gasps). I knew it. I
wasn't actually going to San Francisco the rest of the week, so I let it sit
for 72 hours, but that got me to Saturday. Saturday, it was like, “They still
haven't fixed this and they haven't responded any further with updates.” So I
went to the shop and said, “Give me a new SIM now, please.” And that's
straightforward. Then it worked. So clearly, it was not necessarily the SIM but
the network's knowledge of the SIM. But it's like, “You should have told me to
go to the shop. You have retail shops for this exact purpose. Why did you tell
me to wait 72 hours for engineering to look at it?” So it's one of those, like,
(growls).
But the thing is, Google
doesn't have retail shops. So if that happened to me with Google, they would
have to mail me one. But I suppose if - [crosstalk]
Leo: Erica G. in our chat room says she has Google Fiber
and the service is excellent on Google Fiber. So Google has figured out, in a
consumer-facing business, at least how to do it right. Presumably, that would
apply to an NVNO.
Matthew: I wonder whether Google – would they prepared to help
Sprint and T-Mobile with their networks? I mean if, say, Google has a lot of
people who understand networks, lots of people who've
got tons of Fiber, maybe they can actually help improve those services.
Leo: Intriguing.
Matthew: Then buy them.
Leo: I think that's got to be the ultimate goal is just
buying them.
Kevin: See, the head of Deutsche telecomm was flying a kite
about selling T-Mobile this week, wasn't he? I saw something about that.
Leo: They've been trying to sell it for ages. Remember
Softbank was going to – or was that Sprint? Wasn't Softbank going to buy them
and then Sprint/T-Mobile merger and … But historically, the US government has
not looked too kindly on these things. They don't care if Comcast and Time
Warner merge, that's fine. Apparently, teleco lobbyists are not as effective.
Kevin: But Google buying T-Mobile would be less of an issue
because it wouldn't reduce competition.
Leo: They're not in the same business. Yes, they're not in
the same business.
Kevin: I mean, it would create
vertical integration, which is another kind of problem. But it's – they're less
worried about it these days, I think.
Leo: I'd be all for it. Sprint, not so much. Not so crazy
about Sprint.
Kevin: I haven't tried Sprint. I've been on the other three
and T-Mobile has generally been really good.
Leo: I like T-Mobile.
Kevin: They don't bill you for stuff you don't want. They
don't force you to buy a plan and at the point, you can add any gadget to it
you like. You don't get overcharges for the data. You can actually get
unlimited data, though they do throttle you a little bit.
Matthew: Don't they have a global roaming plan?
Leo: They do. It's not very fast but it works.
Kevin: It's limited to 128 kilobits, which is good enough for
normal textual stuff. But you're supposed to, again, be able to upgrade it. So
last time I was in the UK, which was actually quite a while ago, middle of last
year, I thought, “Oh, okay, let's try these upgrade parts.” So
I quickly upgraded and it doesn't work, doesn't resolve properly, no way to
know who I am or connect to it properly. So, you know, I think to some
extent they realize that these billing systems are actually gigantic, creaking
bits of crap. So if they simplify them, then the customers are happier because
they don't have to interact with a crappy billing system. They don't have to
employ people to be on the phone with them. It works out better for them to do
flat rates.
Leo: You know what would be interesting? Google already
does something kind of cool with the Android device manager. You've used that
on All About Android, right, Jason? Where -
Jason: Well, we actually used it when I was hosting TWiG while you were at -
Leo: Oh, you did it on TWiG.
Jason: Yes.
Leo: Yes, I saw. I mean, it's very cool. So Android device
manager is only available on Nexus phones right now but presumably, if Google
did do an NVNO, they'd be offering – you'd think they would sell it just
through the Play Store or would they have brick and mortar store fronts? How would
they do it? I don't know. But the device manager, what's cool is it's kind of
like the Fire phone where you can get help, a live person, right?
Jason: Yes, you just hit a button and it asks you a couple
questions. Then connects you, within a minute.
Leo: Can't do that on AT&T.
Jason: It was pretty seamless, very impressive.
Leo: Actually, you can do it on AT&T but – never mind.
Matthew: But there's no one there.
Leo: I'm on AT&T and it still works. You have to have a
Nexus phone, it's just on the internet. That's kind of
cool.
Matthew: They have internet on phones now.
Leo: They have the internet on phones. It's an amazing,
amazing thing. It's tiny, but it's there. The little
internet. Apparently, Google makes its self-driving cars in Livonia. Is
that near Moldova and Kazakhstan, I don't know what -
Kevin: Is that a place or a city?
Jason: I believe it's Michigan,
right?
Leo: It's near Detroit. So I think that's kind of cool.
Kevin: Well, that means they can actually build and test them
in snow, now. So that's really good.
Leo: They didn't work in snow, did they? The autonomous
vehicles don't work except in perfect conditions. Only in
California. The goal is to have driverless cars on the market within
five years says Chris Urmson, director of
self-driving cars for Google. This is an article from the Detroit Free Press.
Apparently, if you live in Detroit, you don't have to be explained where
Livonia is. It sounds like Fredonia, the fictional country where the Marx
brothers lived. But apparently, it's near Detroit. ]
Matthew: Or Elbownia. That's where
all the programmers worked in Dilbert.
Leo: Within five years. Are you ready? That's 2020. I'm
ready.
Matthew: I'm ready.
Leo: Really, would you get in one of these? There's no
steering wheel. You just get in the car and you say, “Take me home.”
Kevin: You say, “Okay, Google, oh no!”
Leo: We've all seen that Silicon Valley episode where -
Kevin: I just realized this could be the problem. You're
listening to TWiG and Leo starts directing the car.
Leo: Okay, Google, take me to Livonia. Take me home, is
Livonia. Right now, it's legal – driverless cars are legal in Nevada, Florida,
Washington DC, California and Michigan. I think, though, I may be wrong. But I
seem to remember it was for research purposes, not just, you could take it out
on the road. You can't just get in your autonomous vehicle and drive around.
Matthew: I think Google's going to have to come up with a –
like, the one they have now is cute and everything but I think they're going to
have to come up with something a little more Jetson style or Tesla style for a bunch of people to bother getting in.
Kevin: That is Jetson style. It's a
little – [crosstalk]
Matthew: It is a little, yes.
Kevin: But I mean, somebody, I think it was Megan Mcardle who said it's not so much a driverless car as a
driverless golf cart.
Leo: They asked Urmson about
that. He said, “When you first see a driverless vehicle in your neighborhood,
do you want it to be a big black SUV? Do you want it to be a Hummer?” No. He
says, “We chose that design because it looks safe, and friendly and
unintimidating.”
Matthew: It does.
Kevin: Also that golf carts have different rules about them
than cars, so you know, there are categories of cars, right? I suspect that they're
hedging their bets because you can drive golf carts all over the place and you
can drive them on roads in most jurisdictions. But they have different
constraints on them than cars, trucks and so on.
Leo: I think these are also urban, primarily urban,
vehicles, right? They're really -
Matthew: Definitely.
Leo: So they need to be small for parking for getting
around.
Kevin: Yes. I think these have a smaller range than the ones
we were using before. With that, you know, the other half of that thing is
Tesla, isn't it? So Tesla is building the electric cars that people actually
want because they're very nice cars and they're talking about driverless Teslas and projection – as well. The thing is, this isn't going to be like a sudden thing. The people who
bought Tesla Roses don't want driverless cars. They want to feel control and do
0 to 60 in three seconds or whatever the hell it is.
Matthew: But I'll bet you $10 Uber is going to buy some
driverless cars.
Leo: Uber should.
Matthew: So the first place you're going to see them is stuff
like that, I think, in cabs, delivery vehicles and then eventually they'll
start becoming more consumer level.
Leo: That's kind of key, too, because the autonomous
vehicles that Google showed us all had steering. They were regular – they were
retro-fitted regular cars. They had steering wheels, and brakes and
accelerators. These new little pod-like ones, they don't have a steering wheel.
Kevin: Well, that was the point. In order for them to be able
to drive them on the road, they had to have someone sitting behind the wheel
and have a wheel, have a big red override button.
Leo: Just in case.
Kevin: So this iteration is rebuilding them without them but
with little cuddly golf carts that won't hurt anybody if they hit them.
Leo: What about the person inside? I just think there's
psychologically this notion that I am at the mercy of this computer, whatever
it wants to do, it's going to do, even if it's drive off a cliff. There's
nothing I can do to stop it.
Matthew: I still like what you, and I and Jeff were talking
about, the question about whether the Google smart car AI would make a decision
to drive you off a cliff if the only alternative was, you were going to smash into a bus.
Leo: We had some visitors yesterday from Switzerland. Their
two sons were studying in the United States. I said, “What are they studying?”
They said, “Philosophy.” Oh, boy. I said, “No, wait a minute. That is the new
thing to study because we're in a world now where ethicists, people who are deciding
what artificial intelligence can and cannot do, it is a realm of philosophers
and ethicists. That's a perfect thing to study right now. There's going to be a
huge demand for people to solve things like this.
Well, okay, you're writing the software. The car could
avoid collision in two ways, one, by steering into a bus full of school
children, two, by driving the person in the car off a cliff.
Matthew: The occupant.
Leo: The occupant. So, I mean, the ethical, moral, right
thing to do would of course be to drive off the cliff, save children's lives.
Matthew: But that's prevented by the Third Law of Robotics.
Leo: I'm the occupant. What do you do?
Kevin: Well, you know, these throw the fat guy under the
train things tend to be done by philosophers that don't understand physics. A
fat guy would not stop a train and a Google Car would not stop a school bus.
Leo: That's why they make them so small. The occupant dies
no matter what.
Kevin: If you search for “car hits school bus” on Google
Photos, you'll see what a car looks like when it hits a school bus. There's a
reason school buses don't have seatbelts and weight five tons. It's because if
they hit anything, the thing suffers.
Leo: It's a giant battering ram. The cost could be an
issue. The lidar that these things use is $70
thousand.
Kevin: Yes, but that's – you know, you can scale that. That's
because they don't make many at the moment. All that is, you know, is optics
and senses and okay, there's a mechanical thing that makes it rotate.
Leo: That's actually what this guy said. He said, “It's not
made of Ubobtainium.”
Matthew: It's hilarious, if you read the piece, one of the only
accidents that a Google test car has had is when a driver manually moving one
rear ended a Prius.
Leo: That's really the truth and the answer to that ethical
question is, it's pretty clear these will save lives in net because, you know -
Matthew: Hugely.
Leo: Yes. Six thousand people die each year while texting.
Kevin: But that was the point I was trying to make with the Teslas and we've already seen this with cars, is that this
isn't a once-for-all change for drivers in cars. It's an incremental change.
The cars do more – these cars are down a lot because the cars have more and
more stuff built in, now. They have lane following. They have overtaking lidar warnings. They have a whole set of, you know, ABS
systems and things that can detect what's going on. There's more and more
computation going into the cars to make them safer and the driverless thing is
just an extrapolation point on that curve, where at some point you're like,
“Okay, keep following the lane. Okay, maybe I'll take over.”
Oh, parking. Parking is one that you can just press a
button and the cars will do that now. So there's a whole set of cars that have
pieces of this, where they've solved a small part of the problem space. The
question is, can you solve the entire problem space?
At the moment, it's hard and it requires, basically, Google to scan the entire
road in 3D three times and decide which bits are permanent and which bits are
temporary. But that may change over time as they get better at this. But the
thing is, of all the things to try to top, it's not that high a bar. There's so
many people who are really bad drivers and crash cars into things so many times
that it shouldn't be that hard for them to do a bit better.
Matthew: To do better, yes. But I think there's going to be,
I'll bet you, in cultural terms, there will be a huge upswing in people driving
giant, gas-guzzling Chargers and GTOs and stuff. Because they
want to maintain control over their vehicle.
Leo: “I own my destiny.”
Matthew: Right, so they'll pay people to remove all the
automatic driving things so that they can drive wherever they want.
Leo: Yes. I actually have been in the self-parking Ford.
You would think that might be a little creepy. It's not, it's actually kind of really cool. The wheels spin and it does the whole thing
and does it perfectly. Admittedly, that's not the same as driving down the road
but I think we can get used to this idea that the car drives itself.
We might even like it.
“Google is now a more trusted source of news than the
websites it aggregates.”
Matthew: That doesn't surprise me at all.
Leo: This is something that the delegates at the World
Economic Forum in Davos will see. This is from the 2015 trust barometer. Search
engines are now more trusted than news sources. Traditional
media on this graph, the blue. Red is online search engines and the line
has crossed. By the way, online search engines don't just beat traditional
media, they beat whatever hybrid media is, social media and owned media. I
don't know what these other things are.
Matthew: I actually had a friend who used to be in the
newspaper business say that he didn't understand these results. He said, “How
can the aggregator of something be more trusted than the thing that produced
the content?” I said, it makes perfect sense to me. Because if you go to a single news outlet, you get their news and
their version of the news. If you go to an aggregator, you get news from
all over and presumably, you get a broader sample. It's for objective –
[crosstalk]
Kevin: I think it depends how you ask it. Because if you ask,
“Do you trust the media?” You get a, “No, I trust Google more.” But do you trust,
you know, this particular media that you're a big fan of, whether it's Fox or
MSNBC, wherever it is. Then you'll get a different answer because people still
do like – people still buy the media or ascribe to the media that reinforces
their worldview and they will like those ones more.
Matthew: Actually, Emily Bellin, I
had this conversation about when trust came up before. She said, you know,
“Some of the best newspapers probably wouldn't score high on a trust ranking
because lots of people hate them because they are continually coming up with
investigative stories.”
Leo: Right, they keep goring my ox.
Matthew: [crosstalk] – because the countries are corrupt and
whatnot. So of course people would say they don't trust them and yet they're
clearly better than some of the other ones that fans of Fox news would say,
“Oh, yes, I totally trust Fox news.” Because it tells them what they want to
hear.
Leo: That's a good point. If you view them as an
aggregator, you're going to see them all. I think we all use aggregators, now. I use aggregators. I don't go straight to the New
York Times site to read New York Times articles.
Matthew: In fact, that, whether it's Twitter, or Facebook -
Leo: Or Flipbook, Flipboard, Prismatic, yes.
Matthew: All these tools. The default for lots of people is, “I
want to see information from more sources.” So the idea for newspapers, at
least, that your app is going to be so good that people will just download that
and then they'll just pay for the stuff that comes from your news organization
is just farcical People aren't just going to choose that, I don't think. Or not enough of them.
Kevin: Well, what they're doing is saying, you know, “It's
time to say, we need to build loyalty and we'll give you things and make you
feel special for being part of our club.” That's something Jeff's been
promoting to the -
Leo: Yes, and that's kind of what we do here. I mean, we're
not the only source of tech news by any means and actually, in a way, we are an
aggregator, aren't we? We aggregate news from a variety of sources into this,
the commentaries that we add.
Matthew: Actually, New York Times has had a fair bit of success
with its app, NYT Now, in which they have content from other places that is
selected by their editors. That's a form of aggregation, something that they
used to despise.
Leo: That's where value is added. I presume the value we
add, which is analysis, commentary.
Matthew: In fact, newspapers were the original aggregators.
Let's face it, they took stuff from news wires and all sort of places, and
packaged it up for you and gave it to you.
Leo: Right, right. Quick Google change
log, some quick things from Google. Actually, there's only a couple.
First, Google Chrome version 40 is out. Chrome 40. I'm on the beta, so my beta is 40.0.221485. Hand
off now available on the OS 10, material design as well. The new Google Chrome
is here. The new Google Chrome is here.
As far as change logs go, a change could be a negative
change. The death of Google Glass. Google Glass sales
are halted. That's it, it's over. But Google says, “We're still committing to
launching a consumer product but we're not going to make the Glass Explorer
edition anymore.” That's good, because it cost $1500 and was launched two years
ago, almost two entire years. The Glass team will move out of the Google X
division, that's their RND blue sky division and will have its own department. According to Tony Fidel, the chief executive who was part of the
acquisition of Next. So that's kind of interesting. I guess what this
says is, Google's making a devices division.
Matthew: It's interesting that they're sort of de-emphasizing
Glass and Microsoft is coming out with this very sort of Glass-like thing that
they're clearly betting hugely on.
Kevin: Well, I suspect this isn't a coincidence. I suspect
Google got some rumor of that and were like, “Okay, let's get this out of the
way so that Microsoft does that.” Obviously, Google Glass is a very imperfect
AR in that it's not doing the complex problem of overlaying things directly on
the world. It's just giving you a little screen in the corner. That's a
sub-feature that the Microsoft's does, but what you can actually do is say,
“That thing over there, draw a red ring around it.” It'll let you go and get
it. That's a much harder problem and you know that Google is working on that
too. The reason that you're here in this harbor is they're probably going to be
releasing something new like that. But they're possibly being cagier about it
than Microsoft is being with this.
Matthew: Well, there's the Magic Leap, right? Isn't that the
thing? Their sort of Oculus style -
Kevin: Yes, they bought Magic Leap, yes.
Leo: It makes sense that Google would separate all of these
into device division.
Kevin: I think Magic Leap is more like the stuff that
Microsoft showed. The Microsoft stuff was built by the people who did Kinect
and a lot of that is about the 3D sensing and getting that stuff right. Magic
Leap was that piece of it, not the 3D rendering side of it.
Leo: Magic Leap was a $2 billion startup that was basically
in stealth. Raised a lot of money and it turned out they were building, kind
of, augmented reality glasses, right? How close are these to what Microsoft
showed off? I mean, I see the finger commands.
Matthew: Yes, they feel very similar.
Kevin: I mean, you know, given the video I showed from 25
years ago, these aren't that novel of ideas. In fact, the patents
on a lot of this stuff has actually expired now. Even the data glove
there that that NASA guy was wearing in the show.
Matthew: It did have wires coming out of it.
Leo: Right. No wires, it's a mesh.
Kevin: The difference is with this stuff is that the Kinect
thing and the Magic Leap thing are both doing lidar type things of saying, “Okay, your hand is there, your hand is there, your hand
is there. I can model where your hand is and understand what you're doing with
it and therefore use that as the gesture without you having to actually strap
yourself in.”
Leo: Do you think – it looks so weird. So they're showing
somebody typing on a virtual keyboard and so, really, it just – you're standing
there going like this. Do you think people will accept that notion and just get
kind of used to that?
Matthew: People accept a lot of weird things.
Kevin: They accepted typing on Glass, which everyone said
they wouldn't.
Leo: Yes, but there's a physical thing you're typing on.
Kevin: Yes, but it doesn't have any resistance.
Leo: Well, I'm thinking about people looking at you,
thinking you're such a dork.
Kevin: You would probably do it on a table surface or
something -
Leo: Yes, and you'd do it in private, I think. Is this
better? I guess it's better because it's programmable and -
Matthew: What's that finger thing? Go up – what's that? Is
someone wearing a keyboard on their fingers?
Leo: “I've got productivity on my middle finger and I'm not
afraid to use it.” I don't …
Matthew: That looks painful. That doesn't actually look …
Leo: I guess you stick your hand out and then you can use
your other hand to -
Matthew: Oh, oh.
Kevin: Oh, it's showing you buttons on your hand for you to
click.
Matthew: Okay. So you have to remember which part of your
finger has the -
Leo: No, you don't, because you're seeing it.
Matthew: Oh, it shows you.
Kevin: One of the overlays, yes.
Leo: I don't know. I'm glad I'm getting old. I probably
won't have to worry about this.
Kevin: You'll be grumpy like me because – [crosstalk]
Leo: “Where's the mouse? There's no mouse!”
Matthew: “You damn kids.”
Leo: But it is exactly – I mean, this reminds me so much of
so many, you know, fantasy videos we've seen. Apple had the Knowledge Navigator
thing. This is all kind of in that. Microsoft's been talking about this for a
while, actually. “You already have a room opened, would you like to close
existing room and auto-map contents of room? Would you like to auto-map
contents of room in existing room or would you like to cancel?” I would like to
take a nap, please.
Matthew: “I would like to delete all the contents of every room
you've ever created on accident.”
Leo: “In his living room, this man has pulled up the pod
interface, what the applications call a mini work station for the user that
ensconces their entire view.”
Matthew: Sounds lovely.
Leo: Yikes. Yikes. This is Magic Leap. So did you say that
Google bought them?
Matthew: They did, even though nobody really knows what they're
doing.
Leo: Here's the Hula Hoop interface. That's what they call
this thing that's all around you, the Hula Hoop interface. But it's available
even on the couch.
Matthew: Perfect.
Kevin: I guess you wonder why you'd be on the couch if you
could just lie down in bed.
Leo: That's exactly right. I don't want to have to hold my
arm out, either.
Matthew: Then it's Wall-E, right? You're floating on a
levitating couch with it over it.
Leo: Yes. I just want to think it and it happens. I don't
want to have to move around.
Matthew: I think that – the first thing I thought of when I saw
the HoloLens was a project that I think MIT was
working on years ago because I think they were using Quake, or Doom or
something like that as a model. They were trying to get VR or AR goggles that
would allow you to play Quake, except in the real world. So you'd be able to
run around and see monsters, basically, hallucinate and then shoot them with a
sort of virtual weapon. I would pay money for that.
Leo: You'd get sick, though, very quickly. I can promise
you. How about socialization? It says, “Oh, don't worry. You'll never do yoga
at home again when virtual people can appear beside you.”
Matthew: Nice.
Leo: You know, we were reviewing
yesterday the Samsung VR glasses that are based in Oculus Rift and use a Galaxy
Note 4. One of the toys in there, you can watch a movie. You can watch it, just
on the screen, but you also can be in a movie theater. You know, there's chairs. You can look behind you, there's the
projectionist.
Jason: It sounds so cheesy but it's strangely like -
Leo: It's comforting.
Jason: Yes. Because there's also light
effects. Like they kind of thought about the fact when you're watching a
movie on a big screen, it reflects on the seats beside you. So the action on
the screen actually illuminates and darkens everything that surrounds you. It's
very normal … strange.
Matthew: [crosstalk] – for talking -
Jason: Yes, a cell phone ringing in the corner.
Leo: Here's something I can get behind. You're sitting on
your riding mower, you've got acres to go before you
sleep. You're riding the mower, you're boring, (mower sound). Magic Leap allows
you to butcher virtual gophers instead. You can turn it into a game, as you're
mowing, and the gopher says, “Splat.” You can mow it. Wow. I'd like to have
been at the meeting where they brainstormed this crap.
Matthew: Yes.
Leo: Chopping veggies? Not without a bonus multiplier! So everything will game-ified. Everything. You're playing “Chop the Cucumber.” This is
really – I don't know, I'm not against this. I'm not against this.
Kevin: Have you guys seen Black Mirror?
Matthew: No, read about it.
Kevin: Have you seen Black Mirror, Leo?
Leo: What is it? What is it?
Kevin: It's a British TV show which is kind of like the
Twilight Zone but modern world. They just released it on Netflix in the US. It
originally came out in 2011 in the UK and then they did a second series last
year, I think. They just did a Christmas special this year.
But it's a series of standalone episodes that show
little science fiction stories where they've involved technology in some way.
So the second one is the completely game-ified world
where everyone is going around owning credits for things they didn't have to
spend on media, and other things and so on. That's the moral of the second one.
The third one is that you have a little AR thing in your head which records
everything you've ever seen and done, and lets you thread up on – for people to
see and how that effects people's relationships and so
on. So there is a different – it's worth seeing. It's dark as hell. It has the
same darkness that the original Twilight Zone has but it also has – it's the
presumption is that, you know, projected forward technologies are the kind of
things that's being built.
Leo: I love this. These are not – they're all standalone
episodes, you don't have to watch in sequence.
Kevin: You don't have to watch in sequence.
Leo: Oh, you do have to watch in sequence?
Kevin: You don't.
Matthew: The only one that I heard about was packers
threatening the Prime Minister and forcing him to have sex with a pig or
something.
Kevin: Yes. That's the first episode and that tends to be a
very dividing episode. People either like that or hate it.
Matthew: Or they hate it, right.
Kevin: It's put a lot of people off the series, which – the
premise is farcical, but the point is that it's farce that's well acted, which
is played very straight, which is a very British tradition. The actors in it
are brilliant, you know. It's Rory Kinnear, who won like the Best Actor Award
last year.
Leo: Is Jon Hamm in it?
Kevin: Jon Hamm is in the Christmas special.
Leo: Mr. Mad Men.
Matthew: I have to admit, the first one, the description put me
off, so I have not watched it. But the description you gave sounded
fascinating.
Kevin: If the description puts you off, then don't watch that
one, necessarily. It may put you off the whole series. Because the point of it
is, it's about – actually, if you listen to on the media this week, at
Gladstone Interviews, Charlie Brucker who created it
explains the invention of the idea, which came from the 2010 election where
Gordon Brown went out and talked to a voter, and then got in the car with his
mugger friend and – [crosstalk]
Leo: Oh, I remember that.
Kevin: - said, “Why did you put me in front of that bigoted
woman?” Then he had to go back and apologize to her. Brucker was surprised that the Prime Minister of the country was so driven by the media
that he was forced to do these things he didn't want to do. He says, “Okay, how
do I extend that into farce?” He came up with this concept for that. So if you
know that context, it makes a bit more sense. But it is very much a – that one
is in that sort of uncomfortable – foe territory. The others are more – that
one's not directly about technology, really. That one's about media and people
– I mean, it is about technology.
Matthew: Society.
Kevin: It sort of shows the Twitter mob encouraging him to do
it, effectively. It's that thing that goes on. But that's the sort of sense
that it's about technology. But the second or third ones are actually very
strong dramas and then they keep that going for the second series as well. They
all – I don't know what the word is. They are exaggerated in the same way that
Twilight Zones are. They're in that, “What if we took this thing and pushed it
too far?” But they're also acted and directed well enough that they can sustain
quite a lot of it. So it's definitely worth watching.
Leo: On Netflix in the US, Black Mirror.
Kevin: Yes, I think they – they might not have the Christmas
special yet. You may have to find other sources for that.
Leo: I went to Channel 4's site and it looks like it's
blocked. I can't get any video to play, so you can see at least the first
season on Netflix. “Miss Piggy and a Politico, Together at
Last.” Wow.
Oh, is there one more thing in the Google change log? I
forgot. Let me look.
Kevin: There was the – I think three of these are the same.
So they did a new Google knowledge graph and there's pluses and minuses to this. So they're trying to give you more useful
information. So they're giving you stuff like marking upcoming events and
letting you book tickets from it. It provides that directly in search results
and sending them outside. They're doing a big push on events, which is good,
but the way they're doing it technically is a bit weird in that instead of
marking up the HTML, they're saying, “Inject this giant blob of JASON into the
top of your thing that only you can read. Read that and do this thing.” So
they're basically – it reminds me of -
Leo: And this is the thing. This is that bar on the right,
the knowledge graph on the right. So I just searched for Beyonce,
and my God, I mean. You get songs. You get her social media profile. You get
recent posts. If she were on tour, which apparently she's not, you could even
buy concert tickets from this.
Kevin: Yes. I don't think she's actually on tour at the moment.
Leo: Who's on tour?
Kevin: Midge Yur is on tour but he
may not be big enough.
Matthew: Midge Yur, wow.
Kevin: He played San Francisco last weekend.
Matthew: There's a flashback.
Kevin: I know that he's on tour. I was like, “Should I go to
this?” I'm like, “Yes, I can't go to San Francisco that much.”
Matthew: Grateful Dead are coming back, too.
Leo: Oh, Grateful Dead is reuniting, yes, even though one
of them's dead.
Kevin: [crosstalk] – for upcoming events. So he's playing
Halifax, Canada, in Connecticut, and some of -
Leo: Also controversial is this appearance of the social
media profiles, as well, for whoever you search for. So even if it's a brand,
so if I search for – not “Pixer,” but Pixar, this is
Danny Sullivan's example on Search Engine Land. Not “Pixer,”
but Pixar, I will get their Facebook, YouTube and Twitter profiles. But isn't
this what people want? I don't have a problem with this.
Matthew: Why would someone have a problem with it?
Leo: It's making Search more useful. I don't know. Do
people? I thought you said some people had a problem with it. Let's see if I
can find some tickets – oh. You can also get maps. If you search for Starbucks,
you'll not only get a map of Starbucks but their social profiles. Same thing with Google. So, now this is important if you are
a brand. Google has added some markup for this. So you want to, on your – if
you're a brand, you want to on your Pages make sure you use the markup so you
will get these social profiles embedded. Structured markup and -
Kevin: So this is – markup is one of my hobbies.
Leo: Yes, I know.
Kevin: The way they're doing this is really odd. So, you
know, I've not going to get too much in the weeds here. But the normal way you
do markup is you do HTML, so you have human-facing visible stuff in your HTML.
Then you have invisible stuff that's not for the humans as attributes on that.
Leo: So author equals would be an example, right? The
author -
Kevin: Real author or something like that.
Leo: So Google's deprecated this, alas, but they used to –
you'd put it on your page, real author equals Leo Laporte and then if people search for my name, it would display it more nicely.
Matthew: I'd love to know why they deprecated it.
Leo: Yes. It's weird, isn't it? Because it used to be -
Matthew: I thought it was a great feature.
Leo: You'd have your picture next to your blog. You'd know,
“Hey, that's Leo. There's his picture.”
Kevin: I think we all liked that and they decided it didn't
help people click through to things that they wanted them to click through.
Leo: So now they're not using HTML, this is JASON. It's Javascript, is that what it is? What is it?
Kevin: It's not quite Javascript,
even. This is weird. They're using JASON LD, which is – JASON link data. It's a
JASON variant that has some pre-process directives in it that says, “Things
that start with that is a special kind of thing that says how to interpret this
and translate it.”
Leo: So, for instance, if you're an organization, you would
have @type:organization and
then it would know, “Okay, this is not a human.”
Kevin: Right. So this is the schema.org stuff. But it's the
way they're doing it - so schema.org has its own problems which I can go into
at length if you want. But broadly, it just defines inheritance graph rather
than a set of properties, which means that everything is a member of a thing,
is a member of a thing, is a member of a thing. So you have a place, which has
a subclause of something, which has a subclause of something and so on. So it has that sort of
data architect overkill thing going on. So if you look up, “volcanoes” on
schema.org, they have fax numbers because they inherited that from places. So
it's one of those – this kind of almost makes sense. It doesn't quite. So it's
a little overdone and it's also not defined by a very open process. It was
basically written up by Fiat, by Google and then they have a main link where
you can send suggestions. But it's very unclear how you add stuff to it.
Leo: If the big search engine says, “This is what we want,”
they're going to win. You don't have to use JASON LD, you can also, it looks
like, use Microdata.
Kevin: Yes, but the weird thing about JASON LD is that their
putting JASON into the page but they're not assigning it to anything. So
they're putting a script type here but they're just putting instruction there
that isn't actually assigned to a variable. So the page can't access this.
Leo: Oh, wow.
Kevin: It just declares a bunch of stuff that hasn't got a
variable name. So it's completely for their purposes.
Leo: Only Google can access.
Kevin: The page would have to pass it – no, the page would
have to, like -
Leo: You'd have to have some more script that would say,
“Look for some -”
Kevin: You'd have to run some script to say, “Look for a
script with tied application OD plus JASON and then get the consensus, then
pause it.” It's like, this seems …
Leo: Surely you can embed it in a larger script, couldn't
you?
Kevin: No, because it's got a different type.
Leo: It has to be its own – it has to be its own
application, LD plus JASON? Huh.
Kevin: So it's this very weird thing of, like – this is,
hint, there's a long blissful debate about this. But this comes down to the –
is the stuff that you're adding to this designed to make sense to the people who
are looking at the page or is it just for the search engines? The drawback of
doing stuff just for the search engines is that you can say different things to
the search engine than you can to the page.
Leo: Oh, that's a big deal.
Kevin: Which is a bit tricky and, you know, they're assuming
that they're smart enough to penalize people who do bad things with this. But
what that tends to mean is, the search engine has to decide whether you're a
trusted source or not. So if you look at Twitter, it's kind of like this. Their
system is not quite the same syntax but it's quite a bit similar. You dump a
bunch of crap into your head with metatags that say,
“I'm a Twitter card, this is my URL, I'm this kind,” and so on. Then Twitter
can process it and give you, if someone pastes that link to a tweet, they'll
get a better preview. Except that, because that's letting you inject stuff into
Twitter, they'll only let you do that if they've approved you. So it becomes a
way for them to exercise control over the rest of the Web. To some extent,
isn't that same trend showing up in Google? Historically, Google has been,
“We'll make sense of the whole Web and present the results as best we can and crawl things, apply – in certain areas.” This is more of the, “We will tell you how to markup your page and then
decide whether you're worthy to be included in the knowledge graph or not.”
Leo: The ticketing links are done exactly the same way. So
it's not that Google is creating links and then giving you a place to buy, but
in fact, the artist – here's an example with Ariana Grande. The artist would
use special markup to show a card that included on-sale date availability and
the link goes to your preferred ticket site.
Kevin: Yes, but the point is, the
artist would already have that on the page. So what they're saying is, “Okay,
we gave out -” [crosstalk]
Leo: You don't have to go to the page. It's right in the
search results.
Kevin: If you buy into the artist, the artist is happy with
that. The point is that they're saying, “We're not going to try and make sense
of your page. In order for you to appear in this, you have to add this special
magic markup that will then make it appear in Google. If you want to appear in
Facebook, you have to add this different special magic markup that'll make you
appear in Facebook. You want to appear in Twitter? We've got a third set.”
Leo: There's even a comedy event markup.
Kevin: Oh, this is – you would not believe schema.org.
Leo: So comedy is a different category than music,
apparently. Wow.
Kevin: They've got this sub-class mentality. So they're
trying to – they define an abstract base class and they define sub classes as
the things that they actually see in the real world. It's a bit of a – it's a
bit weird. So I can see what they're trying to do, but it
reminds me of Google Base, which you may remember from a decade ago. It
was a similar idea, which was like, “Okay, parsing the Web is hard. How about
you just upload your stuff to Google and we'll use that to reference, to create
better data here.” So they've – and they cancelled that a few years afterwards
because they realized the only people who do that are the spammers. Then they
had this giant database full of things that didn't matter that much.
Now, the way this works is where, in these kinds of
examples, when you're given the clear incentive, it's like, “If you want to sell tickets, we'll make you able to sell tickets from
the front page of Google before they even click through.” People will dash to
do that. So that's probably a benefit to both parties. But what it's doing is
putting slightly more control over the presentation of the Web to Google in the
same way that we were discussing at the top of the show, where we haven't been
having this problem with the different silos doing different things.
Matthew: I think that actually plays into some of the people's
criticisms about where Google is going with Search, is – not just competitors,
but even users. Where they're trying to pull more stuff into their box from
wherever to give you the stuff you want, which is good. But you know, they
choose what information goes in there and they choose who it comes from. Those
choices sometimes benefit Google. Sometimes they benefit someone else. It sort
of moves beyond the kind of purely, you know, objective search that people
thought they were getting.
Leo: There's always the risk, as there was with the author
tag, of it being deprecated and you spent all this effort doing this and
suddenly Google says, “Yes, we're not going to show that anymore.” Since it has
no place on your own page, it's completely lost now.
Kevin: That's the thing. So, you know, the Microformat way of doing that – my favorite that I helped
found, so I'm biased here. Microformats works on the
basis that what you want to do is annotate the webpage with the smallest
possible change to give the computer's a hint how to
parse the content that's already there. So real author is an
example of that. Real author says, “This particular -” You know, there's
a way of marking up a page to say that it has an address book entry in it, an H
card. You can mark that up to say, “This is the person; this is the URL; this
is their photo, address,” whatever else you want to do. But how do you know
which one's which? Well, if you put real author on the link, then you're
saying, “This one's the author of the page.” Point it to the other page as the
author of this page. That's a hint that should be obvious to the layout of the
page to the human but isn't obvious to the computer.
So when you have an article on the web, it'll have the
title in the bylines somewhere on it. The bylines might be the top, might be
the bottom. But it's usually clear what's mean by that. It'll say, “bi-whatever.” By putting the real author, you can say, “This
is the real author of the page, give a hint to the computer and they can parse
it, and display the author in that way.” Where that becomes
useful – if you go to – a concrete example. If you go to kevinmarks.com,
my site, and scroll down to the bottom, there's a bunch of comments there that
have been created by parsing other people's sites that have an H card in them.
So when they present the comment, it'll show their face next to the comment.
That's the indie Web way of doing this. So we're trying to mark the stuff in a
way that isn't just defined by one company but is defined by a common practice.
Leo: I like that. So instead of having something like Gravitars, using a third party service, it's just in their
page markups. Yes.
Kevin: You know, we looked at them. One of those common things on the web is little
people, little icon plus name, you know. Face, plus name, plus link. So the
minimal way you do that is, you just put, “Class=h-card” on the element that
contains that and if it finds a photo, it's the photo it finds. Sometimes it's
the name. If it finds the URL, it's the link. So you can do that by adding one
value. But if you want to add more detailed ones, if there are six photos in
there, you want to say which on is which? You can put
explicit classes on the elements that represent those.
So the point is that there's two facets. One is that it's not designed to be, “Here's a data structure that
you have to populate, probably into the system.” It's like, “Here's something
you're already doing. Here's some extra hints to make
it easier to parse.” But the second thing is that the information in it is the
information that's already being presented to the people looking at the site.
So you're not going to forget to update it. So the other problem with this
hidden mediator stuff is that when you go to change the site, you'll change all
the stuff that people can see but -
Leo: But you might forget.
Kevin: You might forget about them or maybe you're updated by
another process and you can't tell that's wrong until someone searches you on
Google and sees a picture of the other customer service that's doing it. Because your IDs got crossed or something. So you're better
off – there's this basic principle called DRY – don't repeat yourself. Which
is, if there's information about something that is being presented, present it
once rather than twice. Because if you present it twice, you don't know which
one is correct. The problem we're getting now with
this kind of markup is that if you want things to be presented correctly by Google,
and Facebook and Twitter, you've got to do three different sets of proprietary
markup in there, put the title in three times, put a link to the photo in –
they have different rules about what size photos they accept, so you often have
to make a different photo as well.
Leo: I like it.
Kevin: If your photo isn't big enough, Facebook won't show it
and if it's the wrong shape for Twitter, it'll crop the middle out and make it
look weird. So there's -
Matthew: Real author's a great example. Like, Google went to a
lot of publications, including ours, and said, “You should implement this
thing. Here's how to do it. You have to do it because if you tag the author and
connect it to their posts, they'll show up higher in Google. There will be a
little picture of the author and everything. It will help your rankings, blah blah blah.” You know, because
everyone wants to help their rankings and no one knows what helps or what
doesn't. So if Google specifically says, “This will help.” Then you better do
it. So then lots of people did it and then Google was like, “Well, this isn't
achieving what we want it to, which is to get more people to click or whatever.
So now we won't support it anymore.”
Leo: Well, that's the Google change log. That was good.
That was good. I'm glad we got into that, although I think Microformats is a dim, dark area of the Web.
Kevin: Well, we're trying to make it -
Leo: I wish you all the best. I think it's obviously
brilliant.
Kevin: It's fairly widespread.
Leo: Does Google not acknowledge it? When I search for Microformats, the first thing I find is a Google page about Microformats.
Kevin: Google does acknowledge Microformats,
yes. But it indexes – they haven't updated. We've made some changes to the Microformats markup as we learned stuff and they haven't
caught up with that yet. So they're indexing Microformats 1. Microformats 2 has less arbitrary declarations in it. So it's easier to
write parses for but Google hasn't updated their list of – [crosstalk]
Leo: Is that what Chris Messina was doing at Google? Was he
trying to – no?
Kevin: No. Chris was somewhere else. Chris was inside G+
trying to make it sensible.
Leo: I know Chris Debona is
there, promoting open source. They need an advocate – is that what you were
doing at Google? That's what you were doing.
Kevin: Yes. I was doing some of that stuff, yes.
Leo: You were advocating Microformats.
Kevin: I was doing Microformats,
yes. I was doing open social a lot. I did get Microformats into Blogger, that was one thing I did. I did
encourage the Search people to index it at the time and they have done well
with it.
Leo: Is that how Google kind of works, is it needs to be
somebody in there who's advocating?
Kevin: That's like any company. You know, it's a mixture.
They will find stuff from the outside and use that as a thing to look at, as
well. But if there isn't someone making a case for it, it's unlikely to get
down.
Matthew: “Open social ...”
Kevin: The other thing they will do is look at the – you
know, is this stuff widespread? So Wordpress' default
templates have Microformat's markup, which make it
easier to make sense of Wordpress when you're parsing
it. So there is a large chunk of the web that has this on, that means that's a
useful thing to do that also parses Microformats from
that. But there's also, now, a large part of the Web that has Facebook markup
on because Facebook has encouraged people to put the stuff in so they show up
in Facebook. So the – don't know what Google is. Did they index the Facebook
markup as well and thereby cement the Facebook markup as a way of doing it, or
do they invent their own thing?
Leo: It's so difficult when you're the big search engine,
to know what to do.
Kevin: You know, I suspect they'll do both. They'll try and
make sense of it.
Leo: Challenging, though.
Kevin: Yes. It's an enormously hard problem. The point of
this – part of Microformats was us coming up with ways of doing this that
didn't involve a separate file. That was the original motivation. But the other
thing was, I was running a search engine when we first
started doing this, because I was doing technorati. It was making my life
easier, if we could do literally this thing of, “If you put this markup in,
then your face will show up in technorati.” It was one of our use cases for
that. That was, you know, very like the real author thing that you were talking
about. The point of that was trying to do it in a way that was literally the
smallest possible change to what people were already doing so that the thing
that e found is, the more complex that you put in this, the less likely people
are going to get it right and the better chance you've got of presenting their
search result with some strange noise in it because they accidentally put the
wrong thing in the wrong place.
Leo: Elon Musk is going to compete with Google but not on
the ground. He wants to build an internet in space. What a character.
Matthew: He's Iron Man.
Leo: He's Stark. Elon Musk, at a SpaceX event in Seattle a
few days ago says that he's going to spend $10 billion – of course, of other
people's money. He doesn't ever spend his own money. Building
a space internet. The hyper loop, by the way, is not his either. He's
just saying, “We should do this.” So hundreds of satellites. The idea is, and by the way, Bill Gates tried this. A lot of people tried it.
Hundreds of satellites would orbit around 750 miles above the Earth. That's
actually closer than geosynchronous orbit, which is up to 22 thousand miles
above earth. So lower satellites mean less latency, of
course. It would give you fiber optic speeds, he says.
Matthew: Well.
Leo: Depends on what kind of fiber optic, I guess.
Kevin: So it was – I don't see the director announcement with
Musk. I'd like to hear what he actually said but I think I read on GigaOm – I
usually rely on them for this kind of thing, right? His point was that the
speed of light in air is higher than the speed of light in fiber by a factor of
two-thirds. Well, one and a half, whichever way you look at it. So even though
they're 1200 miles up, they should be able to get pretty good latency out of
this compared to Fiber. The speed of light in copper is even slower, it's a factor of three, so it's half the speed again. So in principle, the
signals through the air should be faster. And that's true as long as there
aren't too many hubs. Honestly, it's only actually true if you're getting to a
server that's far enough away. So it'll probably make sense that they have
ground stations near data centers or something.
Leo: So this is kind of like Project Loon. It's the same
idea, but instead of balloons, you use satellites. They're a little higher but
not much.
Kevin: [crosstalk] – it's like 1200 miles or 12 miles.
Matthew: If you read that story, down lower it says there's
another competing effort from a guy named Greg Wyler called One Web. They
actually know each other. They're working on two competing multi-billion dollar
space satellite projects. Yes, it just makes no sense of me.
Leo: What was that, DRY? Is that the acronym?
Kevin: Don't repeat yourself. But this is different. You
actually want redundant satellite networks so that if one runs redundant, the
choices are -
Leo: Greg Wyler was a satellite chief at Google, Virgin and
Qualcom are involved in this. So they've got some big money. But is Elon Musk
proposing doing it all by himself? I doubt it.
Kevin: No, it would be in consortium. But the point is, you
know, whoever's doing this, they'd have to launch all those satellites. That's
good for him, because he's trying to make launches anyway.
Matthew: Well, and Richard Branson said that he doesn't have –
Elon Musk doesn't have the rights to the spectrum that he'll need, but Greg
does. One Web does.
Leo: Interesting.
Kevin: Spectrum rise is tricky when you're trying to do
global satellites, is that right? That was part of the problem with Iridium.
The other problem with Iridium is that is was conceived way too early and it
was thinking about voice. But the other -
Leo: Iridium exists, by the way, and went back to the
course. I think it was only 60-some satellites. We're talking hundreds of
satellites in this one.
Matthew: Every time I think of satellites, I think of Sirius
and XM. You know, they both spent billions and billions of dollars, blew their
brains out creating two competing networks. That was a really bad idea.
Leo: Right. You know, I feel like, though, the notion of
ubiquitous global internet is certainly an idea who's time has come. If there's a way to implement it, it would be great to do that.
Matthew: I think it'll happen. I just don't know why we'd want
two companies spending billions to do it.
Kevin: Well, it depends, I mean -
Matthew: Competition is good, but.
Kevin: You know, we've got one GPS
system. You know, there's a secondary one that's not quite there yet but we
have one GPS system because it was built by the US government for its purposes
and then realized it had civilian applications. I think there is, potentially,
value in more than one competing satellite network in the same way there's
value in more than one competing ISP. The question is, you know – then it
becomes a business decision whether they can roam to each other or not, unless
the technology is radically different. But I think there's a difference in
thinking about, is this something you put a satellite dish on your house for or
is this something that talks to your phone? Those are actually very different
use cases.
Iridium was designed for a phone. They were to be very
mobile and therefore, you had to have this thing that looked like a 1980s cell
phone to be able to pick up the signal and not fry ahead. Whereas, if you're
doing something that's designed to have a dish on your house like a satellite
TV dish, that's a different proposition and that's more viable because you can
get a lot more signal through that.
Leo: Let's do that and then we'll use mesh to get it
mobile, and we'll be done. Then you and I can go home and use our Hula Hoop
keyboards.
Kevin: Well, way back in – Christ, what was it? '96? Astra,
who runs the satellite networks for Skye over Europe.
They added data capacity to their satellites and I was doing the new media
stuff in the UK at the time. So we got to go and talk to them about – I think
we were helping them with a presentation and then talking about what are
possible applications for this? At the time, the applications were really crap
because they didn't have much bandwidth left after they used most of it for
television and comms that they were already doing. So they had this weird
ability that you could basically multiplex a packet to the whole of Europe
simultaneously. In fact, that's what they had to do because they had all these
satellites that were pointing down and these dishes that were receiving all the signal. Then they had to de-multiplex that and
decide who gets which packet.
So it didn't make much sense for actual, you know,
sending audience and traffic over that, because you're massively wastefully
sending packets tot he whole of Europe.
Leo: It would be great for concerts.
Kevin: Well, except that's exactly what they're doing with
the satellites anyway.
Leo: Oh, that's right. Oh, yes.
Kevin: They've got a TV channel for that. But it'll be great
for, like, a CDN where you really do want to send the same image to everyone,
where you want to send -
Leo: Microsoft could have used that today for their Windows
10 announcement.
Kevin: Right, or they could use it
for, you know, if you have a software patch you want everyone to have, those
kinds of things. It would make sense but it would be very hard to actually set
the infrastructure up for that. We spent a long time trying to think about what could you do, if you could send a data packet to
everyone in Europe who is attached to this thing at the same time and couldn't
come up with any good ones. It's actually quite hard.
Leo: I'm sure you should have asked – I'm sure if you'd
have asked GCHQ, I'm sure they would come up with something you could do with
that. We're going to take a break. When we come back, our
tips of the week. Matthew Ingram from GigaOm, Kevin Marks, Leo Laporte.
Our show today brought to you by LegalZoom. You may
say, “I need a lawyer because I want to make a will, I want to form a
corporation, I want to get my trademarks. I want to get my life organized.”
But, you know, you don't need a law firm. You need LegalZoom. They're not a law
firm, but they provide self help services in your specific direction. A great place to start protecting your family with a will or a
living trust. LegalZoom can help. They make it easy. They make it
affordable. Getting your life organized can mean taking control of your
financial fears too, so starting that business or – you know, we were talking
on Monday. We had a great Triangulation with one of the three founders of
Apple, the lesser known. There was Jobs and Wozniak. Then there was the guy
named Ron Wayne. When Apple first started, they formed as a company and Ron
said, “I had to get out because I was liable. It wasn't an LLC, or an S-corp or
C-corp, it wasn't a corporation.” So because they didn't have the sense to
create a limited liability company in the earliest days of Apple, they lost one
of their three founders and Ron lost a fairly large fortune as a result.
Make that LLC. It's easy. It's affordable. You can do
it. That's how I did it ten years ago. I'll tell you, it was great advice. I
got it from Cameron Rose. He said, “When you start TWiT, make it an LLC.” At Legalzoom.com. For more than ten years, LegalZoom has
helped millions of people get the personalized attention they need and if you
do want more help, they have a fantastic – they call it the LegalZoom legal
plan where you can actually talk to an independent attorney in almost every
state. These are attorneys they've vetted. You can read their profiles online
and you can even read the unedited customer reviews, pick the attorney that's
right for you at pre-negotiated, very low, low rate. So if you're just starting
out a corporation and you say, “Well, should I incorporate in my state or in
Delaware?” You can find out what the pros and cons are on each.
So LegalZoom, again, not a law firm but better,
frankly. Self help services at your specific direction. Don't let another year
pass before you get your life organized. For legal help you can count on for
your family or your small business, legalzoom.com. By the way, if you tell them
you heard it on TWiG, use the offer code TWIG and it'll take $10 off at
checkout. We'll get a little credit. Legalzoom.com, we thank you for your
support of This Week in Google. I used them. I cannot recommend them more highly.
This is the time in the show where we like to give you
some tips, or tricks or tools. I don't know if, Kevin, you've got something.
The last time you were here, maybe a few times before that, you showed us a
game that I couldn't stop playing. So be kind. Oops, you're muted. Unmute.
Matthew: What was the game?
Leo: Was it Freeze? Or was it -
Kevin: It was 2048.
Leo: 2048, that's right.
Kevin: Remember that?
Leo: Oh, God. Don't do that again.
Kevin: Okay, I wasn't going to show you that. Have you seen
Linkbubble? Okay, so I might have to change some settings for you to see this.
But it's an Android browser and the fun thing about it is that you know the
bubble heads in Facebook that it shows for the chat? It does that for the
browser.
Leo: What? Why would I want to do that? So I have a little
– oh, look at that. You get a little pop-up for the
page you want.
Kevin: So if you're browsing Twitter, or Facebook or the Web,
whatever, and there's a bunch of links, you click that and a little bubble appears
on the side. Then, when you want to read them, you click that and get a little
set of bubbles so you can go back to them.
Leo: Oh, neat-o. Look at that.
Kevin: It's kind of cool. Then you can take the bubble and
throw it. You can either throw it to Share or to Instant Paper – [crosstalk]
Leo: Oh, I'm going to get this. That would be useful
because we do this all the time going through stories for other shows.
Kevin: Yes. So it's basically the embedded web browser view.
So it behaves mostly the same as Chrome except that it adds this little UI
change to the browser that makes it quite fun and useful. So I recommend that.
I think it's free if you only want one bubble at a time and there's a Pro
version which is $3 or something, maybe $5. That lets you have access to
multiple bubbles. I tried it out, I thought, “This is really useful.”
Particularly for that use case of – it's quite hard to do the “read later”
thing reliably inside another app. Because this is where it can hook into the
URL thing and give you an aside, you've got the secondary thing that you can
throw them out to the Read Later thing if you want to. So it gives you a way to
triage the links as you click on them.
Leo: I've already downloaded and installed it.
Jason: I should also add on that, Chris Lacey's the
developer. He also released for another dollar an app called Tap Path.
Basically, what it allows you to do is you can program either
one tap, two taps or three taps on a link to do different things. So one
tap would go to your browser, two taps would go to the Linkbubble, three taps
would give a share or something like that. So they play really well together.
Kevin: I haven't played with that one.
Leo: Probably already featured on Android App Arena.
Jason: I had Chris Lacey on All About Android and it was featured on there.
Leo: There you go. This is why, if you watch All About Android, you're way ahead of the game. This is, of
course, not only my producer, Jason Howell but also the host of All About Android and Android App Arena. Tap Path, browser
helper by Chris Lacey which works in conjunction with Linkbubble.
Jason: And other apps, but it works really well with Linkbubble.
Leo: Cool. I just bought both. This is getting to be a
pricey show, here. How about you, Matthew Ingram? Anything you'd like to give
us a plug for?
Matthew: Yes, actually, you've probably all heard of, or used
or talked about this before, but that reminded me of one of the tools I really
like on my Nexus. That's Andmade Share, it's called. It replaces the Share menu
on an Android device and you can create your own sort of custom Share menu,
reorder the things in your Share menu and then you can – what I do a lot is I'm
looking at a webpage and it's something I want to save to Instant Paper, I want
to share it on Twitter, I want to share it on Google+, I want to post it to
Facebook, want to post it to Tumblr. You can do all of those things by just
clicking checkboxes. So take this item, post it to Twitter, post it to Google,
save it to my Instant Paper and put it on my Flipboard magazine. It's a
massive, massive time saver.
Leo: Sounds fabulous. And it's free. Like Handmade, without
the H. Andmade Share. I've been dying for something like this. I wonder if it'll work on – you know, it's really annoying. The Samsung
Share menu, of course, puts all the Samsung stuff on top. If that would work
with that, it would make it a lot more – I would take one of the many ping
points away from the Note 4. But now that I'm using a Nexus, who cares? To heck with that.
My tip actually is something new. What's App is a
messenger service. We mentioned it earlier because, like a couple of other
messenger services, they've added encrypted transactions, which is fantastic.
You can use it as a messenger. A lot of people worldwide use it who are trying to avoid spending money on costly SMS plans. But
one of the drawbacks of What's App, it was available
on every platform including iPhone, Android, Blackberry, Nokia 60 phones and
Windows Phones. But you couldn't use it in your browser until now. They've
actually done this very cleverly. They've created a What's App client that will work with Chrome. So if you go to web.whatsapp.com and
I've set it up, it looks just like What's App. I
actually don't have a chat going on but you would see it. You link it with a QR
code. So you'll have to have the newest What's App
version. I just installed What's App. It worked fine
on Android with the latest version, which was pushed out today. I was just
checking on my Windows for my Lumia 1520 and it works with that too. In fact,
the only thing it doesn't work with is iPhone.
Matthew: iOS, yes.
Leo: What's App says it's because of restrictions Apple
places on the iOS platform.
Matthew: I just set it up, too. It was super easy to get the QR
code.
Leo: Super easy. It was a clever way to do it, I thought,
because you're very much tied to your phone number in What's App. What's App uses your phone number to identify you. So this was just a
clever way of kind of saying to the Chrome browser – it's not an extension as
far as I can tell how they're doing this. It's just a web page. But it's a good
way to tell the web page, “This is the phone number I'm going to use with
this.” I have What's App on all my phones, which means
I have it on many phone numbers.
Matthew: I actually had a friend tell me the other day, he said
he was texting with someone and I said, “Why don't you use What's App?” He said, “Because I can't.” Whatever he was using had a web client and he
said, “I can't use it on my desktop.”
Leo: That's one of the reasons I use Hangouts for my
texting. Or Push Bullet, you could use. There's lots of – Mighty Text. Lots of
way to do that but now What's App will do that as
well. It doesn't – see, I asked my kids and their friends. They don't use What's App. Nobody in the states, I think, really, has
picked up on What's App. But I started installing it because it's secure. I
think it's really great that they built encryption into it. Strong
encryption into it. I think they're using the – what's the technology
the Red Phone guys created? They're using that. So it's open source publicly crypted.
Matthew: We actually have our whole family – I've got three
daughters who all have phones. Some of them are iPhones, some of them are
Android. I'm Android and my wife is iPhone so we all use What's App.
Leo: It's cross-platform.
Matthew: We use it for group chat.
Leo: That's pretty cool. Actually, I don't know if that's
what What's App is using now. I don't know what kind
of technology they're using. But I'm pretty sure it was open source publicly
crypted. Of course, once it's in the What's App app,
you don't know what it is. So I don't know if it's – closed source is always a
little tricky with encryption. Open whisper systems is what they're using. Whisper systems.
Matthew, great to have you, as always. Matthew Ingram. He scribes at gigaom.com. We really
appreciate the time you give us and we hope to have you back soon. Also, Kevin
Marks in his garden with his birds. Kevinmarks.com. What are you up to, anything you want to tell the world?
Kevin: Well, the indie stuff with Microformat is kind of
interesting. So it's the tenth anniversary of technorati tags.
Leo: Wow.
Kevin: I was like, “Wow, that's a long time ago.” So that was
one of the first examples of this where we got people to markup the Web –
markup the numbers with tags in them so that we could parse them and present
them back to them. That syntax is still quite widely used. It's used by
plugging platforms originally to the tank pages. Though, the hashtag has become an even shorter shorthand for that. Now, people can
just hashtag, the hash and the tag inside the post. But one thing that we've
added recently in the indie Web stuff is how to markup a tag of a person
because you – this is a thing you can do in Twitter where you app reply to
somebody or where you – in Facebook, where you can mention somebody and it'll
do a link to them. Google+, if you plus them, it'll do the thing and link to
them in the page. But there wasn't a distributor way of doing this reliably. So
what we did was we just used the H card thing we talked about. I'll give you a
link to this. I'll just stick it in the chat, find the chat window – hang on.
What it does is, it does the H card thing we were
talking about earlier where we were using it for not showing the author of a
post, but if I want to show somebody else in a post, then I can do the same
thing of linking to them. I put H card on it to say that this is a person and
then I can – if I put classic – category, I'm saying, “This is a person tag,
I'm intending this is a tag of that person.” I did that on my last set of home
to website club notes so they're all marked up in that way. Which means, when
people get mentions from that, they can know that I was tagging them as a
person. So at the point when I post it, it'll send mentions to the people I've
marked up in the post and they can know that I've been talking about them.
Leo: Of course, as always with Kevin, I have no idea what
he just said.
Kevin: I'm sure I can take another pass through it.
Leo: No, you're just too smart for me, Kevin. Don't even
bother. It's indiewebcamp.com/person-tag.
Matthew: I thought he put it well. It's like @-mentioning or plussing someone on Google+.
Leo: But it's open.
Matthew: Not restricted.
Kevin: It's a combination – the nice thing is a combination
of a set of building blocks we already had. So we already had the building
block for marking up a person. We already had the building block called Web
mention for saying, “I've already linked to this page, here's the page I linked
to you from.” So the combination of the two means that if someone links to me
on the Web and puts this in and then can send a Web mention to me, then I will
have it linked to me. I can see it.
Leo: Does that always work everywhere?
Kevin: It'll work – well, you need to be listening. So for
them to be making the mention, you need to be listening to them. So if they're
using indie Web stuff, then -
Leo: So I could do this in my leoville.net, you know.
Kevin: But no one will send their web mentions. The
difference is, you have to handle the web mentions on
you homepage as well as on host pages. I'm not sure Known does that yet. But
I've got a script on my personal site that uses another third party service
we've mentioned, webmention_.com that listens for the web mentions for my site
and then I can then embed some javascript that embeds them at the bottom. So if
you go to kevinmarks.com, you see the little comments, and notes and things
that people have sent me in those web mentions. Yes, that's the webpage from
webmentions_.com is all there.
So the point about this is, there's a lot of these little bits and pieces we've been building. They each
have to make sense as a standalone thing. So I've got to markup the page to
say, “I'm mentioning this person.” That's great on its own. I just made that
page more sensible. But it's only useful if someone is going there and looking.
So web mention is a way of saying, “I've mentioned this page. Here you are, go
and have a look.” Then the problem is, how do I set up
something that listens for that? Well, this Webmention.com is it all – that you
can set up to listen to them for you and then it will feed them out. It'll let
you get them back later so you don't actually have to have code running
yourself to do it.
So it's a series
of these pieces where we build a little piece, and then another little piece
and go, “Oh, we can combine those two pieces. Then I can mention people on the
Web and I don't have to be using Twitter or Facebook to mention them inside
their world.”
Leo: I like having Matthew Ingram and Kevin Marks on. It
makes me feel like Barbie. “Math is hard. My head's gonna 'splode.” All right. I guess I'll have to read this, and figure it out
and something. I don't know what I'm going to do with it but I like the idea.
Let's put it that way.
You guys are great and we really love having you on. So
thank you for being here, both of you. We do This Week in Google every
Wednesday, 1 PM Pacific, 4 PM Eastern time, 2100 UTC.
You can watch live. Jeff Jarvis will be back next week. We're getting close to
the one month. I think we can call Gina Trapani and see -
Jason: I believe I already have.
Leo: She said she could be on every month.
Jason: I have to find the schedule page but I believe I
already have her booked.
Matthew: Time's up!
Jason: February 11th, put it on your calendars.
Leo: Time's up on her sabbatical. Maybe she can explain web
mentions to me.
Jason: Just when she thought she was out, they pull her back
in.
Leo: If you can't watch live, on demand versions available
always with all our shows at TwiT.tv, that's our website, in this case,
TwiT.tv/twig. You get every show there or subscribe. That's the best thing to
do, that way you get them each and every time. They download on to your mobile
device or your thing, whatever thing you use to listen to shows like this.
We're glad you do. Thanks for joining us, we'll see
you next time on This Week in Google. Bye, bye!