This is the worst day in commercially available operating system security in my lifetime. There have been bad operating systems in the past (pre come-to-Jesus Microsoft, like 98 and XP...), but rarely has anyone taken a great security product (ChromeOS) and merged it with the worst currently-shipping security product (Android).
Most of ChromeOS's safety "features" come at the expense of not being able to do anything; ChromeOS can't run arbitrary malicious code because it can't really run ANY arbitrary code. Yes, yes, Emscripten, the birth and death of Javascript, etc. -- the fact is that security through uselessness isn't a recipe for adoption.
ChromeOS can do most of the tasks that an everyday Internet user might want. There are lots of people who are content with its restrictions. It's far from useless.
The update system on Chrome is wonderful, in that it just happens. No strange dialog boxes, popup messages or confusing options for the computer-illiterate. It just goes ahead and does it. A power user might not like that, but it is a great system for a huge userbase.
How many insecure ChromeOS devices are there out there? Very few, I'd expect. Not many other operating systems can claim this.
> ChromeOS can do most of the tasks that an everyday Internet user might want.
That's damning with faint praise, isn't it? It can, at this point, do MOST of what an everyday Internet user might want? It's that whole 80/20 thing that everyone thought would let somebody overthrow Microsoft Office -- 80% of users only use 20% of the features. Well that's true and all, but it's not the same 20% for everyone, and you only have to miss one feature to make a Chromebook an untenable alternative to a Windows machine. There was a brief window where Chromebooks were significantly cheaper than a comparable Windows laptop but Intel took the ARM threat seriously, Microsoft took the licensing seriously and now the base Chromebook price from a name manufacturer isn't significantly cheaper than the HP Stream or what have you. "Most" is not going to cut it now that the price advantage is gone.
> There are lots of people who are content with its restrictions.
This is really only true in that we live on a planet with many, many lots of people, and so you can find lots of people for which almost anything holds true if you funnel it down. ChromeOS market share is a rounding error at this point, and if there were any signs that this was turning around, Google apparently decided they weren't enough for them to continue.
Agreed. About a month and a half ago, my grandfather asked me for recommendations on a new computer. I asked him what he needed to do on his computer, and it was looking like a Chromebook (or in his case, more likely a Chromebox) would be perfect, until he mentioned that he runs Quicken. And my mother (an accountant) was emphatic that Quicken Online was no substitute.
Interesting point. I switched to a Mac 10 years ago, and to this day the only software I still fire up a Windows virtual machine for is Quicken. (Quicken Australia never made a Mac version.)
I don't know how old your grandfather is, but Chrome OS is a much better choice than Windows for an elderly person. The sheer power of Windows combined with the most malware of all OSes and unnecessarily frequent UI changes make Windows the very worst OS for the elderly. OS X is better than Windows in this regard, and iOS is better still. Chrome OS is just great for old folks and Chrome OS devices are extremely cost competitive (Chromebooks are cheaper!) with Apple devices.
The comment to which I replied said Quicken Online was no substitute according to the mother. Maybe it's no substitute in sheer power, but still a passable substitute? The elderly benefit from giving up some power in exchange for less system maintenance.
Having an elderly family member struggle to stay up on Windows is super sad to watch and try to help with. I personally definitely wish that I had been more proactive migrating my grandfather to a more locked-down system when he was slightly younger and had a better capacity to adapt.
If the grandfather really is elderly, like older than 75 or 80, I would push in the direction of asking the question "Does he really need Quicken anyway, or maybe he's at a point where he could give up that responsibility, or compromise with Quicken Online?"
I've never met anyone who said it was, so I doubt it. Stop being a blind advocate and realize that some people do need Windows because they need apps that only run on Windows.
It's a trade-off. In my grandpa's case, he would love now to give up Photoshop in exchange for freedom from a litany of Windows nuisances (viruses, registry issues, major UI changes - things many of us forget about that use OS X or Linux). Unfortunately, that ship has sailed; he's 93 now and while not necessarily senile, he has real trouble grokking the cloud and therefore Chrome OS is prohibitively confusing.
When I'm in a pessimistic mood, I can see a day where we return to $4000+ "developer" machines. I think Doctorow wrote about the war on General Purpose Computing. It turns out that developers and power users were casualties in the race to protect users from themselves.
But if you needed to (and had lots of patience…), you could get your whole development stack on a $250 notebook, because there's $250 notebooks with Windows, or where you can install Linux etc.
Granted, it's not as productive (trackpoint or riot!), but it works.
But with the increasing tivoization of computing? Once below-1000€-laptops/desktops disappear in favour of oversized tablets, how are children going to learn coding? People in developing countries? Poor people in developed countries? Not everyone can afford a 1000€ price tag. Should these people really be excluded from everything that isn't mindless passive consumption?
> Once below-1000€-laptops/desktops disappear in favour of oversized tablets, how are children going to learn coding?
Simple enough; they'll code on tablets.
I'm actually currently working on a prototype for how a simple, useful development environment might look on a phone or tablet. It's a stack-based concatenative array language with a zoomable user interface. Instead of representing code as lines of text, it takes a more Smalltalk-like approach of a live environment. The benefit of being a concatenative language is that it naturally lends itself to a tree-like format where each word can be viewed and edited on a phone screen. Since semantically it's closer to a weird mashup of J and Perl(!), it can be concise (hopefully still readable) and the vector aspect makes drawing graphics pretty straightforward.
The downside, of course, is that it can be quite hard to reason in. Truthfully, I'm not really sure how to work around that—and I don't think most children will easily think in a function composition/matrix manipulation way. That said, I'd like to find something since I'm fairly sold on the idea that tacit programming with arrays is key to making coding on a tablet work (I think it's fairly clear that imperative or even conventional function languages would be a royal pain to use in such an environment). Maybe something Lisp-like where the user zooms around the AST would be an alternative.
> Poor people in developed countries?
Much cheaper phones and tablets are already vastly more popular than laptops or desktops among poor people. The solution is to move coding forward on mobile, not simply keep rather trashy cheap laptops around.
That's what I was thinking. Off the top of my head (not 100% checking compatability, but with current Newegg prices), to replace my current machine, not counting the 24 inch 1080p monitor (which in my experience last a long time as long as a tornado doesn't throw rocks into them (really)) and keyboard, mouse, etc.:
Supermicro XnSAE workstation motherboard: 215
3.3GHz 4 core Xeon E3-1226 CPU w/HDMI out: 217
32 GiB DDR3 1600 ECC Kingston memory: 240
Low end 1/3 w/year for 5 years data center 80GB Intel SATA SSD: 100
Seagate "enterprise" <550/TB/year 4 TB hard disk: 210
982 USD plus shipping plus whatever enclosure, power supply, high quality fans etc. you put into it, which ought not go over $200 new. E.g. after discovering a Lian-Li enclosure I love I bought 3 more and cycle through them.
Plus backup, however you do that (make sure critical stuff is offsite!). And, say, a man-week of your time to configure, order, build, install and configure Linux. The result will be very fast and rock solid (well, if put behind a good UPS) workstation class machine with a 5 year design life.
I think it's just a matter of time for that trend of exclusively mobile/dumb devices to start reversing especially as phones/tablets increase in processing power. Canonical already showed us it is possible to run both Android and a full desktop OS on the Android kernel years ago[1], MS Surface tablets are basically laptops without a fixed hinge between the keyboard and screen.
I see at least one clear profit motive for the industry here, let's say you have a "Desktop enabled, smart" smartphone, you'll probably want to buy a dock, screen, keyboard, mouse, productivity software, etc.
While maybe this won't always be the case, the current chromeOS structure allows for arbitrary code, you just have to enable a development mode. The process for doing that is well documented. I don't have a problem with a computer that comes restricted as long as there is a documented process for un-restricting it. Same reason I don't find SIP in OS X or Secure Boot a problem. Security features are good as long as they are configurable.
Right, that's a problem. I fully agree with you on that. I just think it's important that fight for the right thing. Secure Boot isn't the problem. In fact, most people probably should leave it on because it's a good security feature. It is absolutely critical that these features work for users and not against them though.
Funny you should mention Microsoft Windows. As I understand it, the phone OS is now unified with tablet and desktop - but you still can't run anything but windows on a windows phone. I don't expect them to offer up the driver source code (all though that would be nice) - but an open boot loader would be a start.
I suppose it's no surprise. It goes something like this: a) underwrite the device (not necessarily lose money on every sale, just lower the margins), b) introduce an app store, c) take a cut of every transaction.
As long as users and developers need your OS, your device - your appstore - you will make money.
> Once below-1000€-laptops/desktops disappear in favour of oversized tablets, how are children going to learn coding?
There's plenty of free, on-device code editor, compiler/interpreter, etc. apps for Android, including ones that will allow you to run code in the app, or build, package, and install (given that you've enable non-Store installation) Android apps right on the device.
I think there is a difference in not wanting to buy a cheap laptop because of build/hardware quality, and not being able to buy a cheap laptop because it won't let you run arbitrary code. One's a choice you make because you have the money to afford a nicer laptop. The other makes it even harder for the less fortunate to understand computers.
How is a locked down tablet/laptop in any way similar to a computer with a built-in compiler/interpreter and a full hw schematic available in the manual?
I don't remember getting any compiler or interpreter for Amiga, Atari or Apple Mac. You had to pay for them and the schematics were part of the OS SDK, also commercial.
Also going besides the ROM BASIC or Forth meant buying a compiler/interpreter.
I know the Amiga manuals (Amiga 2000) was very well documented, a friend fixed ours by measuring the various documented points on the main board, finding a poorly soldered component or fried transistor (I forget the details).
AFAIK Arexx was part of Worbench? But yes, you did need to get a compiler/interpreter for that. I thought you were talking about stuff like the C64 generation of computers.
I remember there were GNU tools available for the Amiga, and some magazines came with various development tools -- eg: Blitz Basic. But most were certainly commercial (including Blitz Basic) - but one must consider that even with aminet - there was nothing like the essentially free distribution of today (eg: push to github).
According to this site, Amiga Basic was actually bundled with the machine (note-section: "bundled basic language interpreter (free with machine)"):
> > There are lots of people who are content with its restrictions.
Much of the use of ChromeOS and ChromeBooks is in education (where they compete with Macs and iPads), where the buyers are not the users, and the restrictions are actually part of the appeal. It's a decent niche I guess, it's just that but when I see some of the newer Chromebooks with Core-i5 processors, I can't but feel it's a waste of horsepower.
What you're saying flies in the face of the growing evidence that many people are happy with a computing device (phone, tablet) that can't do everything, but can do a lot.
I bought my mom a Chromebook and she loves it. I think the group of users who are willing and able to do all their computing tasks inside a browser window is larger than you think.
Nobody I know kept their chromebook as their main computer for very long because it wasn't able to do the things they wanted to do with their computer. The only people I know who have kept using it wiped it and installed Ubuntu.
Representative sample concerns aside... I've had one for 2 years and am still very actively using it and considering chromeboxes to replace aging PCs at my place and for some of my relatives (come being on the tech illiterate side of the spectrum, while others ran IT departments in their days). I use it for my kids (3 and 6 yo at the moment) and they use it way more than the tablets (not useful enough for them) or the "real" PCs (yet too unapproachable for them without supervision except for basic workflows).
It's not my "main" computer though, but that was never the problem raised in this thread. The fact is, chromebooks and Chrom{e,ium}OS are very decent platforms for all kinds of users, and they are indeed pretty safe to run. And I often do as others said: I ssh from the chromebook to my other PCs in other rooms, or remote servers, etc...
It's also a great device when travelling if you have a 4G phone providing connectivity or you stop in hotels or conferences with WiFi (I wonder if there's been attempts to bring one to a DefCon or pwn conference to see how it holds... haven't checked).
Sure I find that it hads plenty of shortcomings. But most of them are stuff that are shortcomings for me, professional CS/IT dev/tester/consultant/trainer/teacher/etc... Lots of other user groups will also probably fine shortcomings (like the "elderly" person mentioned requiring Quicken), but it doesn't mean you can't work around the issue (substitute it, use quicken from another machine, etc...).
It surely isn't all-purpose or perfect, but it's still great.
I wholeheartedly agree that I'd be deeply sad to see it sank, way more than I'd have been to see it merged the other way around.
That being said, as recently Android Apps were starting to be available on ChromeOS via the Chrome Web Store and worked decently for the most part, I'm curious to see what the actual integration/folding will be. Because it may be largely marketing-oriented to make Android the prominent part (or just the prominent brand) as it's way more popular in terms of user and device share (billions vs probably a few millions, if even that).
EDIT:
And Pichai's quote does not worry me too much until I see something happening: "mobile as a computing paradigm is eventually going to blend with what we think of as desktop today."
In fact, the article seems mostly unsubstantiated, so as long as I don't see an announcement from Google I think I'm not letting go of my chromebook and ChromeOS. :)
Well there is a huge oversight in the fact that enabling developer mode options (namely for crouton- running ubuntu in parallel) completely nixes any promise of security. I love ChromeOS, and I agree with all the above sentiments, but I wish they had better support for securely running virtual Linux distros!
Not that I agree with their use in education, but "a lot more versatile"? I can't think of anything that a chromebook can do that the iPad can't but I can vice versa. The only advantage I see to a chromebook is a built-in keyboard (and price)
A built-in keyboard is huge for the classroom. Chromebooks in classrooms end up being used largely for written work (typing long essays, creating presentations, etc.) that an iPad is less amenable to than a laptop.
Another thing to break or to lose, and to charge, and it adds to the cost. A $200 chromebook blows away a $600 ipad+keyboard combo for majority of education purposes. Google also has much better deployment & operations for devices & apps.
Source: I know lots of people who deploy and manage these in schools.
I have a 150$ chrome book and it fucking rocks. Does everything I need to do on the web, and I can do javascript/webgl dev on the road.. battery lasts FORever... really amazing machines.
I made a computer illiterate friend get one because I was sick of doing tech support for his windows Lappie and he loves it, and has had zero probs with it.. the only time he's called me with a problem with it was when he was trying to some fucking exe to win a free ipod.
CHROME os rocks.
For me? SSH and a VPS. I don't need whatever random crap I decide to `wget && tar x && ./configure` to have any risk of infecting my browser session. And I can keep my VPS up-to-date just by getting a new one from The Cloud every so often, and it comes out way cheaper than buying a beefy developer laptop periodically. I've had the same, low-end Chromebook for almost two years now and barely feel any need to upgrade it.
How can you seriously answer "SSH and VPS" as apps for development ON the Chrombook? They run on there, but you are not doing development on the machine.
It's like driving a little remote control car to the shops and claiming that you went. You didn't.
I could walk into a library and use a terminal there to type code in, but I'm not doing development on that machine - it's just a dumb terminal.
Now I understand the fact that you can remotely connect somewhere else to do work, but the advent of the microcomputer has meant that you shouldn't need to do that. Power usage for the earth will not be great if everyone runs a device that needs power locally, over a network system that needs powered routers, to a computer remotely that needs power just so you can type code in.
I don't think the math works out the way you think it does. If 10 people share a 1000 watt server, that's better than each of them running a 300 watt pc.
I agree that in the context of this (sub)thread, it's a bit odd to claim that chromebooks are decent developer workstations, when what's meant is that they're decent dumb terminals. I'd still like to run my code on my dumb terminal - but I think it makes perfect sense to work like that - have something dumb and wireless for human interaction, and data/processing on a dedicated piece of kit. Be that a low power intel cpu and ssd in a small box behind the tv, or something in a rack somewhere.
> Now I understand the fact that you can remotely connect somewhere else to do work, but the advent of the microcomputer has meant that you shouldn't need to do that.
And the advent of dynamic libraries has meant we shouldn't need Docker, but, well, that's how it is. It's true that in an ideal world we shouldn't need such a mechanism, but in practice this system works very well for me, and the other ones have been frustrating.
I need a working internet connection anyway to do work: I need everything from software updates to GitHub to IM with coworkers to Google and StackOverflow. And the remote machine shouldn't need any more power than my local one. So my added power consumption is just that of my Chromebook, which is about 10 W.
I admit freely that I am not fulfilling the definition of "development on the Chromebook" in a strict sense, but it is certainly a serious answer for how I develop on my Chromebook.
You could install crouton without a DE and use the built-in shell to do whatever you wanted if you had a need to do development work and wanted to use ChromeOS without a full graphical Ubuntu install. You would have access to apt and vim.
I did this for a while, except my thin client was an X1 Carbon (!)
Have you calculated how much it's costing you?
I ended up scripting up a way to find the cheapest Spot Instance price from Amazon for whatever workload I needed - if I was doing builds I wanted the 8 core machines (c4.2xlarge?). They're ~11c an hour as a Spot Instance, although it's 3x the amount for a Windows instance, and I'd be pushed out of the market if everyone started doing this.
I think I was averaging a couple of dollars a day, so I suppose it comes out in favour of using AWS all the time.
So my current VPS is a non-virtual Kimsufi box for which I'm paying $28 (USD)/month, i.e., $336/year. I'm also paying about $10/month more for a cell phone plan with enough mobile data, which brings that to $456/year. An equivalently-specced laptop that is comfortable enough to use (lightweight, cool, not going to break on me, etc.) would cost me about 2-3 times that amount, and I'd want to replace it at least that often.
(Also if I had a powerful local laptop, I couldn't tell it to do a build, close the lid, and get on the subway without my backpack catching fire. Nor could I have it provide IPv6 to my apartment, host websites behind CloudFlare, provide shells to my friends, etc.)
If I wanted to put some more effort into it, I could certainly do a good fraction of my work (git clone, vi, IRC and email, etc.) on a micro instance. But even a couple of dollars a day would require me to be pretty careful to come out cheaper.
Sweet! Thanks for that information. Until my net connection becomes more reliable it's not really an option for me 24/7.
I think my costs would be a bit higher because Kimsufi/OVH/Hetzner are 300-400ms away from me, so the latency is a bit uncomfortable.
When I was in Canada, a Hetzner box was 200ms away and usable, but for ~$30 a month I only had a dual core box - a bit restrictive when doing large builds.
I might have a go at automating setting up a dev environment on AWS which can be torn down when not in use.
You are not going to sell in third world if your product completely depend on market.and thired world is where you can beat windows with being CHEAPER.and second fact nobody has internet like this in third world.Chromebooks are excellent product with Google fiber.but outside of that,not really(for me as someone lives in thired world country).because even with spending 90$/mounth for internet.my internet is nowhere near to run everything on the web.
Third world? Hell, we're having enough connectivity problems in the first world. Costs aside, a VPS is useless if your ISP connects your entire state with only one fibre link which can be felled by a once-every-few-years storm (UPC Austria and Carinthia, btw.).
I'm not much of an Amazon cloud user, but for compiling stuff like new Linux kernels, or big c++ projects (firefox, chrome) - having instant speed for almost nothing sounds good.
Could you share a couple of lines on how you would set that up? Spin up an upstream-provided basic AMI (eg: RedHat/Ubunut/Debian) - and run a few commands under screen via ssh - and then shut it down?
In the last couple years at least, Chromebooks have often topped Amazon best-sellers. Technically, a lot of people buy them. It's just that compared to mobile devices, laptops as a whole are a small category. Admittedly, most Chromebooks are bought because they are inexpensive (what virtually no one buys is Chromebook Pixel, which I have never seen outside Google except the ones folks got for free at I/O).
I suspect that's a factor of there being very few models. A Windows laptop that sells millions is probably available in 20 versions, each of which is counted separately. Try searching for the current Amazon.com best seller in laptops, the Toshiba Satellite C55.
I think most of what that means is that Amazon isn't a first-party seller of MacBooks. (You can buy MacBooks on Amazon but I believe those are all third-party sellers.)
There is not a huge gap in difference, in terms of HW quality between a Chromebook and a normal cheap laptop (in my experience). Can you provide an example ?
I planned to buy a Chromebook a while ago and do exactly as the parent suggested, but got to the conclusion that it's just not worth the hassle (also, Chromebooks tend to have minimal storage, which is definitely not ideal in a normal laptop).
In general, Chrome OS or not, the HW (touchpad, screen and keyboard even more than raw power) in a 200$ laptop is BAD.
The Toshiba Chromebook 2 ( http://www.amazon.com/Toshiba-CB35-B3340-Chromebook-Celeron-... ) has a great 1080p IPS screen, 9 hours of battery life, and, from what I can remember when I played with it, a good keyboard and trackpad along with decent build quality, for $290. It's got a decent Bay Trail 8W Celeron processor, and 4gb RAM (there are also models with faster processors). There are Windows laptops around that price range also, but AFAIK none with that combination of hardware quality and specs.
As a developer, an unjailbroken Chromebook is my preferred local terminal, because all my work involves SSHing to servers anyway. The Secure Shell and Mosh apps work great (and I even use the Mosh app in Chrome on "real" OSes, because it's simpler).
Before Chrome OS, I ran homegrown minimal Linux installations with a separate user account to run Firefox. It was way too much overhead for me to be excited about maintaining, and paying $200 for a secure architecture and security updates by the most competent people in the industry was super attractive. I can do servers very well; I don't also need to do desktops, let alone my personal desktop.
Not really; my previous job was maintaining a commercial Linux-based desktop. For that one I had a powerful local machine so I could run some desktops in virtualization; I hardly wanted to use my own machine as a guinea pig, because I needed that machine for actually getting work done. I did in fact have a desktop Linux machine on my desk, but in practice I just ran shells and Firefox on it, and half the time I was connected to my screen remotely.
A CrOS-like OS with the ability to run VMs would be pretty great. (And yes, I should check out Qubes.)
There's also pnacl, not to mention the fact that a LOT of what people actually do online is all done/doable in the browser alone. A modern browser is the most full-featured application platform that has ever existed in terms of what comes "in the box".
I'm all for allowing native applications a place to run, however that doesn't mean that ChromeOS was not able to do anything... both my parents and grandparents are all on ChromeOS laptops now... why? It's all they need, secure and I haven't had to remotely clear out any malware in over a year.
> both my parents and grandparents are all on ChromeOS laptops now
This is pretty much one of the two markets I've ever heard of for ChromeOS machines, the other being education space. What they have in common is that:
1) People want them for SOMEBODY ELSE
2) Because they don't trust that somebody else with the power of a less locked-down machine.
Sure, schools are buying ChromeOS machines... for the students. Not the teachers and administrators. It's too limited of a market segment for ChromeOS to grow a large enough userbase.
Actually, my parents got them themselves.. I got one for my grandmother when her computer crashed (I was so happy, I ordered her a chromebook and had it overnighted). She was so happy with it... My mom got hers after she saw it, and knowing that it didn't have to deal with viruses, all they ever did was browse, email, facebook. My dad got his about six months later, and my other grandmother got hers about six months ago.
They've all been very happy with them... well, one of my grandmothers has trouble with using it sometimes, her hand shakes, but that's a problem with desktops for her too.
I used one for about a year and if I were able to use the VPN to remote into work (stupid proprietary VPN software), I'd probably still be using it... I used remote desktop and SSH to other systems when I needed more for work, but was pretty happy with it. Yes, there were times I needed more, I had a desktop for that... one at work, another at home... most of the time I used my chromebook though.
These days aside from at work, I use my htpc for more web stuff than anything else... I've tried a few android and arm based mini systems for that role, but the experience was less than stellar.. currently an i3-5010u which does pretty well for my general use, and htpc chores. I'm on it right now.
Most people only use/need the web for home use... Chromebooks are great for that... There's even a few halfway decent general purpose email clients being worked on (one client to work over imap, etc for multiple accounts) as opposed to typical webmail. That was about the only shortcoming I really felt. Also, getting my private rsa token for SSH setup on the thing was a little bit of a pain.
I think Chromebooks will go down in history as an example of a very viable product for some particular niches, such as those that you mentioned, which isn't a big enough product to be interesting to Google.
If you can be flexible enough with the definition of 'full-featured' to read as 'able to run/produce more', I would argue that 'the browser' – if it encompasses http, html, javascript, and css – is more full-featured than Java. And we wouldn't even need to argue that the former's install base towers over the latter's.
That may be true, depending on how you look at it? JavaScript was the language that, as Eric Lippert put it, was designed to "make the monkey dance.[1]" HTML and CSS and JavaScript are all designed around this ethos of try to figure out what was meant and throw SOMETHING on the screen. It's very forgiving, and it's really easy to get up and running with it quick. I'm not sure that's the same as full-featured; it makes a lot of things more accessible than Java or .NET or Win32 or Cocoa. But it doesn't have the width of functionality they do (it is catching up, admittedly, but they're not required to stay static either) and it makes more complicated tasks more difficult, especially if you need them to be performant.
Most OSes don't have render/styling systems that come close to what a browser does at their core... you can do a lot, but those features you have to often work for, or use libraries on top of the OS. And if you consider UI toolkits on top of the OS separate from the OS that's another issue.
In the end it all runs on an operating system... so it technically falls short of that of course, but building a web-based app is generally easier than raw-coding an application from the OS level.
Don't let nostalgia blind you: Smalltalk was great, but 2015 browsers are obviously a lot more capable. Not because of any great design insight, but because they need to run things like YouTube, WebGL and whatever else.
That's beside the point made in the OP, which attributed Chrome's great security to its lack of functionality, which is just false. You argument is a valid one with respect to its lack of popularity.
That said, while compatibility helps adoption, it is not absolutely necessary. If it were, we would all be stuck with Windows computers in our pockets running on Atom processors having 1:30 hours battery life.
why do you consider running in a sandbox "useless" ?
that's pretty extreme. you really can't think of any use at all for code running in a context where its ability to take power away from the user, and do things without the users permission is greatly reduced? Is your primary goal and idea of utility about doing those sorts of things?
..Or maybe you simply have a deficient imagination?
With Marshmallow Android is moving towards a runtime-permission-prompt model similar to Chrome. They've also done a lot of work on sandboxing using SELinux. Also, after stagefright they started doing monthly security patches and have gotten buy-in from the OEMs to push those patches. Android's biggest security challenge has been getting the manufacturers and carriers to update their software but there has been a lot of work towards changing that, from moving big parts of the OS to userland so they can be updated independently of the core system, to applying business pressure. Presumably updates will be easier on Chromebooks than on phones.
"With Marshmallow Android is moving towards a runtime-permission-prompt model similar to Chrome"
This is horrible from a usability perspective. Note that large bulk laptop purchasers (schools, large enterprises with mostly web application requirements etc.) have been the biggest markets for chromebook. One of the biggest reasons for that popularity is the lower cost of management of chromebooks compared to windows or mac laptops. In that the base OS not only just works, it also just updates. A huge saving in admin time and headache. The other factor is that all user data is automatically in the cloud, which means backups/restores/replacements/upgrades are trivial operations now.
One way chromebooks can retain this ease of use/administration is by basically shipping a laptop with chrome running on Android OS but no other apps installable by the user (outside of chrome sandbox). This will get Google out of the two OS kernels management business but could still retain the essential goodness of the chromebook concept.
> Android OS but no other apps installable by the user
There are quite a few device management solutions out there for Android that can white list, black list, deny all installs of apps. It's very common to setup in business and already completely supported.
It's even far more powerful than that with capabilities to do things like deny certain apps/networks unless you have certain password complexity requirements, etc..
So each app will continue to ask for access to everything and you choose between being info-excluded or giving away your data.
> buy-in from the OEMs to push those patches
How does this work?
Is there any legal obligation for manufacturers/vendors to provide these patches in X weeks?
If there isn't then I won't keep my hopes up.
> from moving big parts of the OS to userland so they can be updated separately from the core system
How exactly? By moving from Android to google proprietary apps? If so then this is a two-edged sword as it inevitably binds you to google spyware. Either you own an half-phone or you give away your data.
Sorry if I sound bitter but I see what Android could be and get defeated by what it is.
I never owned an iPhone but my next phone will be one. Maybe I'll still be disappointed, but at least won't have to deal with CM, Xprivacy, manual updates and all the other shenanigans.
Disclaimer: I'm an increasingly disappointed Android user.
> So each app will continue to ask for access to everything and you choose between being info-excluded or giving away your data.
I think you might misunderstand what the Marshmallow upgrade will allow. Users will have the ability to turn off any given permission for any given app. Don't want google maps to have GPS access? Cool, just turn it off and the app will have to deal with the lack of information. It's not a Windows model carte blanche per-run thing.
Sure, apps can be abusive and ask for more permissions than they need and refuse to run if they don't have them, but that's a problem with any conceivable permission system.
> Sure, apps can be abusive and ask for more permissions than they need and refuse to run if they don't have them, but that's a problem with any conceivable permission system.
The OS could give fake data to the apps. Such as a arbitrary GPS location, or an empty contact list.
Hopefully, that would be unnecessary. If my calculator app won't run without full access to my address book, I'll seek out an alternative calculator which will; at this extreme, with an open marketplace, there will surely be alternatives.
That's not something that developers would stand for. As a developer, I want to know that a permission was denied, so I can take action. If a permission I asked for was denied, it's quite possible my app won't run correctly. And I don't want to deal with idiots leaving crappy reviews because they decided to deny permission.
> So each app will continue to ask for access to everything and you choose between being info-excluded or giving away your data.
Apps under the new model will only ask for permissions when they need it. So, when you decide to post a picture to Facebook, that's when Android asks if you want to give the Facebook app access to your pictures on storage. If you deny it, it lets the app know it doesn't have access. If it's an older app that doesn't understand, it gets fake data as if you had no photos (or no contacts, etc). At least that was my understanding from reading about it.
Note that this security model is similar to web apps wanting to "connect" with you Google account (or Twitter or Facebook for that matter) to access your contacts or similar.
Sadly I am also an increasingly disappointed Android user. I have been with Android since the TMobile G1 and with the lack of security updates, the changing of what's acceptable in the UI every year at I/O (sometimes contradicting what they said the previous year), and the apparent lack of direction I too am ditching all my Android devices and going to the iPhone. Looked a Blackberry Leap but it doesn't properly support video conferencing apps for my entire family (FaceTime + Hangouts + Skype).
It's disappointing. I saw a Chromebook used by a "normal" person for the first time this week. He loved it.
The only benefit for me was the fact you could run Linux in some sense on it, and it was cheap and stealable without too much sadness!
But this merging of browser-OS and insecure Android has signed the death warrant on my devices. Goodbye Samsung and Sony.
> How does this work? Is there any legal obligation for manufacturers/vendors to provide these patches in X weeks?
If there isn't then I won't keep my hopes up.
When a user has control over updates, the company selling the product carries a relatively low responsibility. When a company does not allow a user to update, then that company carries a much higher responsibility.
If Google provides a security update and a company refuses to update their devices (or allow users to update), then that company has a lot to lose.
The Android ecosystem would be a lot better off if there were a couple large class-action lawsuits with big payouts. Companies really don't care until it hits their bottom line.
The liability scenario you claim will fix everything has already been the status quo in Android-land from the beginning. It clearly hasn't fixed the attitude OEMs have towards updates and security in thee last 5 years and other than outside pressure (Apple) continuing to make them look inept at update and support, I don't see any drive from within to fix the situation, and the Android-enthusiest attitude of "just put a custom ROM on it to fix your broken, abandoned product" only reduces the number of people who understand the problem and would be angry enough to file lawsuits or give bad PR a year after their purchase.
Yeah, the problem is what to get instead if staying on Android.
I surely won't pay what Google is asking for their devices, given lack of SD card and changeable battery.
Plus, they also proved they aren't trustworthy by not providing a path for those that gave them money for the Nexus with TI processor. As if Google didn't had the money to keep supporting them.
It's like they don't care, which is sad. They have got into a great position to be a contender in the fickle horse-race that is the mobile phone industry, but the same horse is trampling the people that funded them to get there.
"What about security updates for my phone?" The hurdle is jumped, and the patron has the Google horse land on them, causing death
Unfortunately the latest Samsung devices don't have SD cards or changeable batteries either. They have apparently voluntarily eliminated two of their key differentiating features, not sure why.
I hoped Google would merge Android into ChromeOS, a much more lightweight OS. Now they are stuck with the broken Java UI architecture (too much sits in the main thread, too many object oriented classes). Low latency UI isn't possible by the initial pre-Google (Android company) bad design choices. Even my high end Android phones is just as fast and responsive as the iPhone 6s but has 4 times as much CPU and a lot higher CPU raw speed and 4 times as much RAM. Beside that Android is the good guy, it's the Win95 20 years later - an universal OS that can be used for many things without restrictions. The Android 5+ UI theme is really outstanding great, I personally favour it over iOS 7+ and Win8+. Maybe Android will get HTML5 web apps as first party citizens. And don't forget about the lovely Go language.
> I hoped Google would merge Android into ChromeOS, a much more lightweight OS. Now they are stuck with the broken Java UI architecture
If, as stated in the article, they merge ChromeOS into Android, rather than abandoning ChromeOS in favor of Android (which seems to be how you and many others have interpreted it), they aren't limited to what is currently in Android, since they bring all the capabilities of ChromeOS are now in Android. A merger is functionally equivalent regardless of direction, which just becomes about branding (and its hardly surprising that Android is the more marketable brand.)
There's nothing insecure about Android. It just gives a few more ways to install apps (side loading for example) which clever social engineers/scammers use to their advantage. It's like saying your credit card is insecure because you gave it to the homeless guy on the street when he promised to only buy a sandwich...
Android security is typically horrible when you consider that most phones ship with software that never gets updated throughout their life and has been put together in a quite haphazard way. It's not a fault of Android per-se but it's the case in pretty much anything that isn't a Nexus.
Not really. Android got big because OEMs can do as they like with it. And what OEMs like doing, apparently, is not updating the OS on their device. An Android OS that, from the start, controlled OEMs in some way (by e.g. forcing them to ship updates on a certain schedule) would likely not have become a viable contender in the market.
It's not the fault of Android that OEMs are stupid, but Android was created specifically to cater to OEMs and their stupidity. Android's existence is itself the problem.
(It'd be great if their licensing model up until now was just a bootstrapping strategy, though, and they began requiring things of their OEMs.)
It would be nice if they forced OEMs to use stock Android and update regularly, but they're already under government investigation for requiring Android branded phones to use certain Google services like Play...
To a certain degree, they're forced to either be open as they are now, or sell devices. Given absurd government rulings, forcing stock Android on OEMs would probably bring an antitrust suit against them.
Apple's market share is fairly small compared to Android, on a world-wide basis. And because Apple is closed and proprietary, it can do what it likes. No other companies are installing iOS or Mac OS X on devices, so Apple has nobody to mess around.
> An Android OS that, from the start, controlled OEMs in some way would likely not have become a viable contender in the market.
I'm not sure... From the start, Android was the only viable contender for OEMs. Their choices were create their own OS (good luck catching up to iOS) or take Android with whatever restrictions it had.
Don't forget that were alternatives and I'm sure the large manufacturers were playing Google off against Microsoft, Nokia (Maemo), Samsung (Bada), Palm (WebOS), etc. Those were bigger names in mobile than Google and it wasn't obvious that Apple was going to steamroller the entire industry so quickly.
As we saw, the established companies wasted years and billions trying to play their same games with customer neglect, pointless restrictions and marketing deals, etc. Android became big precisely because Google was desperate to get into the market and agreed to all of it.
Maybe it's a carrier problem? The 5.1.1 update does exist for the S4... And I guess good is relative, and has degrees from 'shitty budget Chinese/Indian no-name OEM' to 'HTC/LG/Moto/Samsung' to 'Nexus brand'.
Android's attack surface is much higher than Chrome OS's. On CrOS I can usually get away with running questionable NaCl code because the NaCl runtime sandboxes what it can do, and on top of that, the browser itself is sandboxed by the kernel. On Android, questionable apps have direct access to the entire kernel system call interface, as well as to other OS features.
Personally, I'm not so much upset because Android is uncommonly bad (it's like any other system that gives untrusted users non-root access to Linux: you can probably get away with it, but eh), but because CrOS is uncommonly good.
> (it's like any other system that gives untrusted users non-root access to Linux: you can probably get away with it, but eh)
This is far from true. Android has always had a ton of sandboxing. Every app is its own user, unable to interact arbitrarily with other apps, things like that.
You won't find any desktop OS that comes anywhere close to the security design of Android. But the thorn in Android's security side remains updates, or lack thereof.
> Android's attack surface is much higher than Chrome OS's.
Not really. All you have to do in CrOS is compromise Chrome (which has happened plenty of times), and you have owned the user. You've won. Game over. And Chrome's attack surface is huge, just like any other browser's. The kernel syscall interface is relatively tiny, and SELinux clamps that down even further.
Apps have direct access to the entire kernel system call interface. They don't run as root, and in particular they run as different UIDs (which is, to be clear, fantastic in its own right) and with SELinux policies. But their "Application Sandbox" is nothing more than that. Apps have as much access as, say, a well-run public shell server gives to their users. That's a lot more attack surface than JS in my browser has.
The update problem is Androids greatest security flaw. Around 80% of Android devices out there run with known security related bugs:
http://androidvulnerabilities.org
That many phone manufactures don't update their software is not an Android problem. Would Lenovo update their devices more often when their operating system was not based on Android? Would Apple update less often when iOS was based on Android?
Lenovo, Samsung, Apple and Google are all in complete control of the devices they sell: they all control hardware, software and services themselves.
If the software is based on Android or not has nothing to do with how often these devices get updates.
It's more like every single business, from ice cream store to accountant, habitually requests the same information from you. It's not exactly an Android problem, but somehow I feel it's less on iOS.
I don't know what's so horrible about an OS that by default encrypts all data on the device, allows applications only to run in a sandbox and uses SELinux.
You can also use MAC and avoid container overhead. SELinux is the worst of the bunch because its a PITA to work with, while AppArmor profiles are pretty easy to write but its not particularly foolproof, Tomoyo has nobody developing good profiles for it, and we should all be running Grsec Kernels + PaX since its the best of all worlds but again nobody wants to put in the effort to develop all the profiles for all the programs we use, and learning modes are just unacceptable.
In all fairness, you don't run any modern UNIX server without giving each sensitive server its own user, which effectively sandboxes it (not quite the same as chroot/containers/VMs, but still). Of course, desktop UNIX/Linux tends to have many important processes running as the same user, but that falls under "Desktop OSes are generally pretty bad".
Everyone sane runs sensitive systems "system high" now, where there's a machine dedicated per task. User/process/thread security in UNIX/POSIX/etc. is bullshit weak sauce, sadly.
You can use VMs for some of that, but that's the limit on sharing (and that's if you trust your hypervisor to be a separation kernel thing; reasonable for many people. Not for others.) Docker/containers isn't enough. Users aren't enough. Processes aren't enough.
Yeah, but they've been getting better over time. It's monotonically non-decreasing security, at least, and the desktop->mobile transition has been amazing from a security perspective. This is a step backward. The only question is how big is that step.
Nah, you just don't see the big picture. You're coming at it from the tech side. It's better to look at the business side. Microsoft has 89% desktop market share, so they basically own the desktop. Apple will never be a real threat because of price. Chrome/Android has an excellent chance to be a real alternative on the desktop. if it gets a large enough market share, perhaps we'll see Adobe's apps, for instance, along with all other major apps.
Is Apple really that much more expensive? Or is that just something that people believe because it used to be true? I keep hearing this repeated, but I bought my dad a $499 Mac Mini.
With Microsoft producing a laptop of their own that starts at $1499, I think all we can really say is "Apple doesn't produce bargain laptops".
Typically not in the segment where they compete (high-end laptops). However, the cheapest MacBook (in Germany) is 999 Euro. A large chunk of the market cannot afford or does not want to spend 999 Euro on a laptop. If you only want to e-mail, upload/view some photos, make a spreadsheet, a 999 Euro MacBook is expensive compared to a 200 Euro Chromebook or 300 Euro Windows laptop.
It was true to varying extents before Apple switched to Intel processors. Since then it's never been true by more than ± few percent for equivalent hardware.
The complicating factor, and why it's such an enduring belief, is that Apple doesn't ship adware and has minimal hardware standards which are higher than the lowest-end PC market. You see this over and over again where someone is either unable to back up claims or, when pressed, has to admit that the $400 notebook they said was “just like” the MacBook Air had a previous generation processor, magnetic hard drive instead of SSD, low-contrast/dim display, mushy keyboard or unusable trackpad, etc. The same manufacturer probably even makes something fairly competitive, too, but it costs about as much.
> Doesn't really matter. Apple makes a very narrow range of products for high-end consumers, and that's a small portion of the market.
It's about half the market as far as profit goes – the low-end stuff is absurdly low margin and customers have very little brand loyalty.
More importantly, the netbook and other low-end market has been fading as people buy tablets and large smartphones. It's not at crossover yet but there's a growing percentage of people who don't have a traditional computer at all and that's going to eat into the low-end market most heavily because the difference in utility between a $200 netbook and a $200 tablet is the smallest and hardware quality is often better on the tablet side.
The demographic shift is somewhat fascinating, with PC ownership declining by 10% for people under 30 in the last 5 years:
“Today, 78% of adults under 30 own a laptop or desktop computer, compared with 88% who did so in 2010. Smartphone ownership, on the other hand, has surpassed both of these devices, with 86% of 18- to 29-year-olds owning one in 2015.”
It's about half the market as far as profit goes – the low-end stuff is absurdly low margin and customers have very little brand loyalty.
Yes, that's true. But Apple needs to have a high profit margin because it only makes $1 billion a week (after tax) and has only $200 billion in the bank. It would obviously be a waste of money to serve more people, or even pay its Chinese workers a living wage.
More importantly, the netbook and other low-end market has been fading as people buy tablets and large smartphones.
The netbook market hasn't existed for a very long time, but sales of $199 laptops like the HP Stream 11 seem to be pretty good. Obviously they don't include the ability to overcharge users by $100 a shot for small increases in RAM or storage.
However, the tablet market -- including iPads -- is declining faster than the PC market. Nowadays even Apple has followed Samsung etc in making phablets.
" It would obviously be a waste of money to serve more people, or even pay its Chinese workers a living wage."
They do. Compare the workers at Foxconn Apple factories to other factory workers. And remember that just about every other tech company also makes their stuff in China.
Every company makes stuff in China, but not under the massive pressure that Apple's big-bang marketing puts them under, where workers are forced to deliver many millions of devices in the shortest possible time.
It was Apple production lines that were driving suicides, and hence the nets round Foxconn dorms.
Pegatron has had Chinese workers doing 90-100 hours a week, sometimes more, just so Apple fanboys can get their fancy new toys without the massive loss of kudos that waiting a month would cost them.
http://www.chinalaborwatch.org/report/107
And Apple workers do not get a living wage, according to China Labor Watch, even though -- as I stated correctly -- Apple is making a billion dollars a week in profits and has more than $200 billion in cash.
Of course, this isn't brutal capitalist exploitation.
I think for directly comparable hardware its not much more expensive, but:
1) the entry price is higher -- to get a minimally functional Apple laptop or desktop you're paying more than any of the alternatives.
2) the smaller number of Apple models compared to any major PC maker (much less the union of all PC makers) means that, unless your hardware preferences happen to match perfectly to an Apple model, you're often able to find a PC that matches your preferences less expensively simply because you are unlikely to have to overshoot your target as far as with Apple.
Apple has a significant, growing share of the computers which people buy for themselves. If you go to a conference, coffee shop, etc. notice that more than half of the laptops are Macs.
Windows has a huge lock-in for the business world due to compatibility, and a few other niches like high-end gaming for similar reasons. Without API compatibility that's a really tough space to get into.
It's quite believable that Android/ChromeOS will take over the netbook market by being cheaper than Windows, and the premium personal market – which is where most of the profit is – if they can get quality hardware shipping at prices competitive enough to make up for the lack of applications (almost all Android apps will need some sort of UI modifications) but it really doesn't seem like they'll get anywhere in the business world. Cracking that means either convincing everyone to drop legacy apps for web apps or remote clients (how many more decades?) or offering API compatibility (invest heavily in WINE?) — otherwise, the cost of a Windows license isn't that much (especially with 10) and certainly not enough to push someone towards the most limited option.
> It's quite believable that Android/ChromeOS will take over the netbook market by being cheaper than Windows
Every store I visit the cheapest laptop-like devices are windows 10 devices, 20-30$ cheaper than chrome os devices. Android ones certainly exist, but only start at roughly double the price of cheap windows 10 machines.
Those windows 10 machines aren't even bad. If you need office + browser they work pretty well.
Just checking today on Amazon:
Cheapest chromebook: 161.99 (and that's the Asus C201, as in STAY AWAY). Cheapest vaguely reasonable chromebook ~210
Cheapest windows 10 laptop: 164.89. This one is actually usable. Screen resolution being the biggest minus point.
Cheapest android "laptop": 165, and this is not something you want. Cheapest one you do want: ~$300. There are cheaper android laptops, but they're chinese duplicates you do not want.
Outside of the US, windows 10 laptops win by a mug bigger margin.
Agreed. Right now, I think much of that would be explained by existing contracts and volume – the Windows devices are shipping in much larger quantities whereas the Android faux-laptops have tended to be curiosities rather than serious ventures.
I think if Google continues to be willing to subsidize Android development at a loss to maintain marketshare, that could change if Microsoft isn't willing to lower the price for a consumer copy of Windows to zero. Anyone who remembers the 90s knows that Microsoft is unlikely to make it that easy.
How do you qualify Android as the worst security-shipping product? Is it mainly due to the lack of sound update mechanisms? Because conceptually, Android's model (each application is its own Unix user) seems pretty good
It depends. Maybe we just see the Play store added to chromebooks, and chromebooks are rebranded as android devices. I mean, there's already so many different android runtimes/versions, what's one more?
That's a pretty bold statement to make. The announcement contains zero detail, to the point that I'm left questioning what the purpose of leaking the information was other than page views.
That's an intriguing possibility. Given their relative marketshares, even if the plan were to kill Android in favor of ChromeOS, they'd probably say exactly what they're saying now.
98 was so bug-riddled that only 98 SE was actually usable. And I'd argue that Windows 2000 was much better-liked than XP on that period. XP was horrible at launch and needed two service packs to become actually liked.
Android will have good security just once they've added more stapled-on legs. Security doesn't come from engineering and good design, it comes from adding more things!
Thank you for saying this! I am an Android developer and I wanted someone to come here and say good things about Android so I could feel safer for my job!
Within the space of general purpose desktop/mobile operating system products: Windows 10 (pretty good). OS X (not great -- which is a shame). Most Linux distributions (OK to good as shipped, can be great). Most BSD distributions (good to great). iOS (fucking AMAZING). ChromeOS (great, RIP). Windows Phone (honestly don't know much about it). BB OS 10/QNX (great, but irrelevant sadly). And then there's Android.
Care to explain the reasons? Doesn't windows 10 phone home? The window xp/7 i used previously would inevitability get affected by viruses. I am using Android for last 4 years have not have single virus or even malware problem.
This is pretty funny considering Windows is the cause for nearly all of the security exploits that occur. The horror stories of people being infected with viruses, malware and ransomware all have one thing in common - Windows. You only need to look at the thousands of security patches this OS has had applied to it over the years and the new ones applied every month to truly grasp what a sieve it really is. BTW, where are all of the horror stories concerning the Stagefright exploit that so many "pundits/idiots" predicted would be a disaster for Android? I'm still waiting for that.
Also, was there a reason you started referring to specific versions of OS's like Windows 10 and BB OS 10 and then disrespectfully mentioned Android without a version number? Don't you think that was a bit disingenuous, and lame, of you to start mentioning specific OS versions and leave out Android? If you're going to use version numbers then shouldn't you have applied them to all of the OS's you mentioned instead of cherry picking OS versions and then shitting out the line "And then there's Android" at the end?
OK -- Android (until absolute latest) = lovecraftian horror.
Android marshmallow = exceptionally bad.
BB10 is a fundamentally different OS (QNX based) than pre-BB10. but Pre-10 was also a good OS from a security perspective, just utter shit to develop meaningful third party apps for.
Windows...used to be horrible. Big changes around Vista (although, ~unusable). Since Windows 7, it's been "reasonable" in a highly managed corporate environment, and not bad even out of the box. 8, 8.1, and 10 have been improvements on that. It's actually easier to do a 10k user highly locked down Windows deployment (although expensive, and involving a lot of experts and third-party tools) than to do a 10k user locked down Mac OS X deployment. Apple had a lot of advantages by starting from UNIX and from basically starting after security was a "thing", but hasn't done as good a job on OS X security as I'd like. iOS, on the other hand, is amazing -- the only serious deficiencies I find with iOS are a lack of "enterprise as sole root of trust" (which no one does, with the possible exception of (Blackberry Pre-10 and post-10), or roll-your-own open source linux/bsd with a lot of trusted computing grafted on in ways which are not at all trivial to do), and a lack of emphasis on anti-forensics on the device itself -- if you unlock it, it contains an sqllite db with ~every message, which shouldn't be how they do it.
OTOH, on cloud services, Google is far and away superior to Apple. The biggest problem with Apple is you're largely pushed toward iCloud which is not amazing from a security perspective. The true win of ChromeOS was you were equally pushed to the Google ecosystem, which is amazing for cloud service security -- the sole problem being you're 100% exposed to Google, Inc. which is both a US company and a single third-party entity, but if you had to pick a single company to be responsible for your cloud services security, you'd probably pick Google on the merits.
Why is Android 6.0 exceptionally horrible? I found one bug in a Play Services API that can crash Play Services remotely, but I can't get code execution so Google aren't interested.
Could you explain why you think Android Marshmallow is "exceptionally bad"? Because I'd really like to know why you think it is. Android gets a bad rap for malware, but it's really a lie perpetuated by people who dislike the platform because they're jealous of its popularity. And even when malware laden apps are sideloaded by people from questionable sites the effects are often minor due to the mitigation strategies employed by Android. Like I said before - where is this pandemic of Android infected phones wreaking havoc? Where are all of the stagefright exploit stories? There's virtually none.
Additionally, Android patches are distributed pretty fast by Google to their phones. It's unfortunate that the OEM's and carriers delay the process for their phones, but Google has no current control over this.
Your explanation that Ryan is jealous of Android's popularity doesn't make sense to me. This is a guy that has been working in security for some time, on both ends of the public / private infosec space [1]. Why would he be jealous of Android, but not ChromeOS?
Someone else expounded earlier: "On Android, questionable apps have direct access to the entire kernel system call interface, as well as to other OS features. Personally, I'm not so much upset because Android is uncommonly bad (it's like any other system that gives untrusted users non-root access to Linux: you can probably get away with it, but eh), but because CrOS is uncommonly good."
I don't think it's reasonable to ask for pointers on how to root an Android device in a public forum, and the size of the attack surface seems to me like a rather reasonable measure of a system's security.
Please, I did not specifically say this person was jealous of Android's popularity. I simply stated that the impetus for people who denigrate Android are generally people that are jealous of the platform's success and try to knock it down a few notches. Also, Android apps are sandboxed and each app is isolated in its own directory. Additionally, could you please cite how these "questionable apps have direct access to the entire kernel system call interface"? If Android apps want to access system functionality then I believe they have to go through a layer of Android framework services which in turn have access to the real kernel system calls.
As for the size of the attack surface being a factor - I agree, but with the exception of QNX and its microkernel (which has issues of its own) pretty much every other monolithic kernel based OS also has a large attack surface so I'm not exactly sure what point you're trying to make other than the obvious.
>Personally, I'm not so much upset because Android is uncommonly bad (it's like any other system that gives untrusted users non-root access to Linux: you can probably get away with it, but eh), but because CrOS is uncommonly good."
Chrome does have great sandboxing, but don't disparage Android just because it works like 99% of the other OS's out there. And I wouldn't be surprised if Android inherits some of Chrome's sandboxing tech when the two merge because I think it's pretty much inevitable that it will.
OK, this is one of those times where infosec people get annoyingly close to "appealing to authority." So far 'rdl has said Android is "the worst currently-shipping security product", "[Blah blah blah], and then there's Android", "Android (until absolute latest) = lovecraftian horror" but the latest is upgraded to "exceptionally bad", and a deeply-nested subcomment about running every task on a separate machine (which I'm sure ChromeOS does... /s) But who needs to hear logical arguments? When the guy from CloudFlare speaks, the thinking is done!
Alright, as something of an Android fanboy it pains me to admit, he probably has a point in this case. But it'd be nice to hear about what specifically is wrong, especially since it would seem there's a fair bit right, like SELinux, automatic encryption, ASLR. The worst thing in my view, is the permissions system, which is probably salvageable (and as I understand it, was improved in Marshmallow.) 'rdl has given more details about pretty much every imaginable system but Android -- are we supposed to just assume Android does every bad thing he mentioned, and the opposite of every good thing?
Edit: I could almost let this slide since the particular counter-comment was equally vacuous. It's still annoying to see "security guy from $BIG_CO => GTFO" (who is apparently smarter than all the security people at Google who've worked on Android, but we just have to take your word for it since you didn't elaborate.)
OK, this is one of those times where infosec people get annoyingly close to "appealing to authority."
This. It's not just "close to" appealing to authority, it's straight up appealing to authority. And yes, it's a common practice in the infosec community. The whole security industry is built on a reputation system. It pays to remember that the number of exploits and vulnerabilities etc. in commonly used software that are reported on practically a daily basis came out of the same reputation based system.
In another comment, he says "user/process/thread security in UNIX/POSIX/etc. is bullshit weak sauce, sadly."
Geofft expounds some more: On Android, questionable apps have direct access to the entire kernel system call interface, as well as to other OS features. Personally, I'm not so much upset because Android is uncommonly bad (it's like any other system that gives untrusted users non-root access to Linux: you can probably get away with it, but eh), but because CrOS is uncommonly good.
> he says "user/process/thread security in UNIX/POSIX/etc. is bullshit weak sauce, sadly."
Android's security indeed relies on UNIX's (filesystem permissions, separate UIDs for each app), but he praised OSX and iOS (I think) for starting from UNIX[0], so it would be inconsistent to also damn Android for it (which I don't think he necessarily did, since dissing UNIX security was in a different context.) From skimming the iOS security guide (and having no iOS dev experience) it looks like it's similar: "The majority of iOS runs as the non-privileged user “mobile,” as do all third-party apps" -- although being BSDish vs Android's Linux could mean that's quite different.
> On Android, questionable apps have direct access to the entire kernel system call interface, as well as to other OS features... because CrOS is uncommonly good.
So CrOS has the advantage because I guess it does not allow running untrusted native code at all, containing things entirely in interpreted/JIT'd VMs. That would be a good point... except every other OS 'rdl esteems higher than Android (Windows, OSX, iOS, Blackberry) probably also runs native code with full access to syscalls (making some assumptions here, but I think it's reasonable that to whatever extent 'geofft's statement about Android is true, it applies fairly uniformly to the rest.) So, why are they good or even just OK, but Android is horribad?
I'm not sure I'm actually addressing arguments 'rdl would have made about Android -- but that's the point, he didn't really say anything that can be argued!
> FWIW, I've heard Googlers familiar with it refer to Android as a pile of crap, as well.
Yeah, it's probably true -- I wasn't saying "Google made it, so it must be good", but underlining the fallacy in relying on a prestigious company association as proof of merit. Android probably didn't have the best talent when it started in 2003 and is now painted into a corner of backwards-compatibility and a huge codebase that would be risky to revamp all at once. I guess my favorite aspect of Android is really just that it's open-source-ish and has enough critical mass to be useful (as opposed to Firefox OS and Ubuntu at the moment) -- so if ChromiumOS were to become the new AOSP as some are speculating, sounds good to me.
To speculate a bit, he seem to favor iOS above all else because:
A) its BSD (Torvalds have offended a sizable part of the -sec community by not highlighting security updates in release notes)
B) it is locked down by default, to the point that there is not even a common storage area for files (this "blew up" in Android a while back because some big name messaging app was dumb enough to put unencrypted user data in exactly such a common area).
Good points. I think bullshit weak sauce is busllshit weak sauce regardless of context. If you don't trust it on a server, why should you on your phone?
The comparison between Android and iOS is especially curious, in part because it's such a competitive space, and also because rdl's opinions varied so extremely.
Both OS X and iOS are based on Darwin, which also suggests vulnerabilities in kernel calls are not what rdl had in mind. When his latest startup was acquired, the blog post [1] described a part of it as:
> Beyond the company’s work in VPNs, CryptoSeal applied Trusted Computing technology to commodity servers, protecting them from compromise by outsider attackers or insider subversion, and guaranteeing the integrity of server-side applications to remote end users.
Perhaps something about the hardware iOS runs on is what impressed him so?
"There are security bugs" does not mean "pile of crap". Everything has security bugs. The question is how bad they are and how much it costs to buy a zero-day / the use of a zero-day.
I agree that everything has security bugs. It's pretty much inevitable. But, one platform has nearly 5.5 times the security issues of the other. So, if you're using security issues to correlate what a "pile of crap" is than you need not look further than iOS.
The biggest difference between Android and Chrome OS for me as a user is how it handles user multi-tasking.
I've tried using an Android tablet for work. It's nigh impossible, even with a keyboard because it doesn't support multi-tasking well. Sure, it supports OS multi-tasking, but it doesn't support user multi-tasking.
Android is based around the single-task model. You have a foreground app that takes up the entire screen and other apps are backgrounded. You can switch between apps fairly quickly, but you can only have one open at a time. There are some attempts to fix this, such as Samsung's split-screen, but none are officially supported. We've even moved to having Chrome tabs use this model.
I love my Chromebook for work though. It does everything I need to do, including having multiple document windows open for multi-tasking.
Chrome OS is built on a multi-task model. You can have many windows open at once and quickly switch between them without losing state or visibility.
Windows tried to merge the two with Windows 8 and it was a terrible user experience. The last thing I want for my dual 24" monitors is to be able to only use one window at a time. Having a messaging app take up the entire monitor is ridiculous. Hopefully they'll find a middle ground. Until user multi-tasking exists in Android I won't be using it for anything but my phone.
I advise against the use of the term "multitasking" to mean anything other than OS/unit of schedulable work multitasking, which is the proper meaning. The same happened to "real-time", and generally having technical terms with concrete meanings be watered down isn't preferable. Multiple workspaces is really what Android lacks.
Multitasking is used in multiple contexts now. Commercials and consumers use it to talk about Android and iOS features. It's too late to try and claim such a generally useful word like "multitasking" as a keyword.
Would be really interesting to see some serious productivity studies done on this. Not to pick on you -- I appreciate you insight -- but humans don't do multi-tasking well.
I wonder (and have no idea, just throwing the idea out there), if people would be more productive in that single tasking environment.
For me its not just about multitasking. Its about distraction. With one window per app on a big screen (eg Windows 8), its just too distracting to switch. I have to go to application switcher, choose appropriate window, and everything looks different now. However, if I've typical two windows side by side of even overlapping, the switch very less distracting...and of course quicker.
That said, I do use a tiling window manager on my 13" laptop for work. I mostly use full-screen mode, but it works well for me, because switching is very fast.
Human ‘tasks’ and computer ‘tasks’ are not the same thing. A productive ‘task’ for me, at work or at home, typically involves concurrently using one or more editors (text, graphic, schematic, …), one or more documentation readers (web browser, PDF viewer, …), and one or more communications tools (email, IRC, hangouts, …).
Seems everyone here is panicking because they think that this means Chrome OS will be killed and Android will be made to run on Chromebooks - but I actually think the signs point to the opposite - Chrome OS becoming the Phone/Tablet OS under a new guise. I wouldn't read too much into a leak filtered through journalists, and instead look at what Google's actually working on:
- Google has Chromium developers working on a DART-based Mobile UI framework and execution engine, Flutter (http://flutter.io/). It's looking to be far better than the existing Android UI system - built for touch and 120fps from the start. This uses the Dartium VM and a bridge to allow the DART apps to use all the native features of the platform, it's much more than just another web framework. Development on this is very active right now, clearly a sizable team working fulltime - and they're building new developer tools also. There was a talk on this a while ago: https://www.youtube.com/watch?v=PnIWl33YMwA .
- Google has built a Runtime to allow existing Android Java-based apps to run on Chrome OS, and is currently testing this and working with developers to get their apps to run on it. It doesn't make much sense to invest in building that out just for chromebooks, since the experience on a Chromebook with Android apps is pretty awful (can't resize etc), but it makes total sense if it's going to be how legacy Java/Android apps run on the new Chrome based phone OS.
The sad truth is that Android simply isn't a very well engineered system - it's been improved over time, but problems persist - like the complex update process leading to unsatisfied users and security problems, poor UI performance (even now, Android can barely do simple animations at a steady 60fps on the latest Nexus devices, and has little hope of allowing for the beautiful animations the Material Design team has come up with), and poor battery life. Google's also at a dead-end with Java given the ongoing legal battles, and with Apache Harmony dead they have to maintain the standard library implementation themselves.
On the other hand, Chrome OS performs great, has awesome battery life on Chromebooks, is quite possibly the most secure end-user OS ever, and Chromebooks get speedy updates for at least 5 years. I know which one I'd choose as the basis for a merged OS.
Yes it's interesting to note that this is all coming from non-tech media interpreting the CEO's comments. The CEO is never gonna say "Android is dead and replaced by ChromeOS", just because of relative market shares. No need to piss off customers and partners. In two years when we realize what's going on, he'll just say that we misunderstood him. In the meantime, if no one closer to the coalface fleshes out the details of this supposed transition, the story will seem less and less likely.
AFAIK Sky is a project of the Dart team without any link with the Android team.
It is an interesting idea but right now it is little more than a proof of concept. Even the demonstration app is extremely underwhelming.
Sky (which is renamed Flutter now) is developing fast though and is clearly a well-resourced project. Certainly fits with a preview next year and 2017 release timeframe. It could be one of several projects competing though.
Update 7:40PM: We've updated the article's headline to be more accurate. A Google spokesperson has confirmed to The Verge that both Chrome OS and Android will continue to exist; Chrome OS is not being "killed."
So... they aren't going to focus more on Android-based laptops and rebrand Chromebooks to something without Chrome in the name?
Because otherwise I'd just completely utterly ignore any and all Chromebooks since it'd be effectively obsolete and probably unsupported as soon as 2017 Rolls around.
Mixed feelings on this. As someone with two Chromebooks (one Pixel, one Samsung ARM Chromebook) , an Android phone, and two Android tablets, here's what I see as the potential benefits and tradeoffs:
+ Being able to run Android apps on a Chromebook would be awesome. There's already some limited support for this, but it would be nice for it to be official.
+ Taking over the Chromebook line would hopefully force Google to make document editing not suck on Android devices.
+ Android running on Chromebooks will hopefully make the display scale better on hi-DPI devices.
- ChromeOS is fairly lightweight and actually runs surprisingly well on cheap devices. I have no doubt that my Chromebook Pixel would run Android just fine, but the Samsung ARM Chromebook would probably chug, even though it's running ChromeOS well enough.
- Making Android run on laptops will hopefully make Google step up their Android document editing game, but as of right now it still sucks. Even the cheap Samsung chromebook was leagues ahead of using my tablet and a bluetooth keyboard in terms of trying to compose documents.
- The ability to install crouton on Chromebooks and have another Linux chroot on the side is an awesome feature of the Chromebook. I imagine that might be more difficult under Android, even though it still uses the linux kernel.
> ChromeOS is fairly lightweight and actually runs surprisingly well on cheap devices. I have no doubt that my Chromebook Pixel would run Android just fine, but the Samsung ARM Chromebook would probably chug, even though it's running ChromeOS well enough.
Plenty of inexpensive ARM devices running Android. I don't see that this is likely to be a major problem.
> Making Android run on laptops will hopefully make Google step up their Android document editing game, but as of right now it still sucks. Even the cheap Samsung chromebook was leagues ahead of using my tablet and a bluetooth keyboard in terms of trying to compose documents.
If they're folding ChromeOS into Android, rather than just killing ChromeOS, I expect that (using similar Chrome apps) post-convergence Android devices will be at least as good as pre-convergence ChromeOS devices at document editing.
Actually if Google going to push coreboot into Android/ARM market more aggressively it's going to help quite a lot with running GNU/Linux distributions there.
It would be easier to convert to linux since Android is already a VM on top of it, right?
I'm hoping that since they're now targeting the desktop OS market that they'll make it easier to run native linux programs on Android. My biggest complaint of using ChromeOS on a pixel 2 is that there are no good text editors on the level of Sublime Text (which works on linux).
> My biggest complaint of using ChromeOS on a pixel 2 is that there are no good text editors on the level of Sublime Text (which works on linux).
Atom works great in Crouton on my pixel 2. I've switched to that as my full-time dev machine. I run everything but Atom in terminals. I run Atom in an X window (xiwi) and have no linux desktop at all. Lean and mean.
There's quite a bit of info online on how to install a chroot under Android. Like with a Chromebook, you need root access though (aka. ChromeOS "developer mode" or a "rooted" Android device).
If this is really where things are going, it wouldn't surprise me if crouton contributors continued the project for Android post convergence.
Took'em a while but better late than never I suppose. I have long been saying (on various forums :)) that this is the right approach.
Flash is dying off and if Android gets better desktop window manager and shell (the bar is already set too low - ChromeOS sucked with its stupid everything including WiFi settings in Chrome, tiny fonts and web-only apps) people will finally have a credible alternative to Windows desktops/laptops. Plus this gives Google a chance to do something like Continuum - without having to be beholden to x86 for apps like MS.
Chrome OS having just a browser UI with web-only apps is what made it appealing. I actually admired Google for releasing two competing operating systems and to me this is sad news, even though I've been an Android fan. You make it sound like Google is the one that needs a "credible alternative" to Windows, when in fact it is Microsoft that needs a credible alternative to Chrome OS and Android.
Speaking of which I don't believe in Continuum. It's an idea that's natural coming from a company that has a desktop OS and wants to leverage that in order to maybe win some market share on mobile. I first heard of the idea from Canonical with their Ubuntu phone, another company trying to leverage a desktop OS.
It's a cool idea for geeks, but made obsolete by cloud-enabled apps like Google Docs. Why would one need Continuum when the changes you make in Google Docs on your phone are reflected immediately on every device or PCs you have?
> You make it sound like Google is the one that needs a "credible alternative" to Windows
With the new Windows licensing Windows is the better alternative to ChromeOS - you can do more at the same price and you can lock it down if you need to a la ChromeOS. In some ways Google is stepping up the game - with Android you can do more.
Good point about Google Docs use case however that's very limited - If I carry the shared storage of my phone with me, if I can blow up the phone to use two monitors and still have my missed calls, SMSs etc in front of me - it enables a greater use case. There is definitely merit in that if continuum is done right.
What Windows licensing are you talking about? Windows 10 is licensed the same as ever.
On missed calls and SMSs, with my iOS device I get both SMS messages and phone calls through iMessage and Facetime on my MacBook. With my Android device I do the same for SMS messages, plus all the notifications I'm interested in with PushBullet.
Windows 8.1 with Bing has been free on devices with smaller than 10-inch screens, and that gets a free upgrade to Windows 10. Windows 10 also seems to be available at low prices....
The US Justice Department forced Microsoft to stop offering special deals on Windows so all the top suppliers paid the same price (1). I assume that's lifted now that Microsoft is no longer closely monitored by the DoJ. Anyone know?
(1) Which seemed to have the (presumably) unintended effect of pushing up the average price of Windows and increasing Microsoft's profits.
Windows 10 is more expensive than Windiws 8. The price listed is $120 for Home and Pro at $200. But that's without taxes because in my country that's $148 for Home and $298 for Pro, after taxes. And I've seen this notion flying that Windows is cheap, but I want whatever you guys are smoking.
Windows 10 retail includes Microsoft support and the right to transfer it from one PC to another. Almost nobody buys it, or if they do (like me) it's usually on a cheap launch offer.
Most copies of Windows are configured, tested, pre-installed, locked to a single PC (not transferrable), and supported by PC manufacturers. They get it at discounted rates, which typically range from $0 to about $45. That's cheap if you think of it as $1 per month over 4 years.
PC manufacturers typically add "crapware" to consumer PCs. These are programs that they get paid to distribute (much like websites that collect $1 if they foist Chrome or the Google Toolbar on you). With luck, they can get the cost of Windows down to $0. They can also get extra discounts or even cash support for advertising from Microsoft and/or Intel for making certain statements, eg recommending Windows or playing Intel's bongs in a TV advert.
In some cases -- for example, if enough users subscribe to the bundled AV software -- the charge for Windows turns negative, ie they make a profit. With a really efficient OEM, the consumer may pay less than $0 for the pre-installed copy of Windows (other costs being hidden).
Since you can buy Windows laptops for £150/$199, the suppliers are clearly not paying $120 or more for Windows.
OK, on the differences of the retail license, you've got a solid point.
In my country (Romania) the laptops with Windows are easily $70 more expensive than those without Windows. I know this because every time I'm looking to buy a laptop I'm doing my research and end up amazed by the difference, given all the crapware they come with, which as you are saying, should subsidize the price.
Big companies like Dell and Lenovo (which owns IBM's old PC business) get the best deals. Smaller companies get worse deals: I've had that complaint from UK manufacturers. Small Romanian companies may get even worse deals.
But I would have thought Acer and Asus would be competitive, because they're global suppliers....
I think the parent poster meant, Windows (including 10, presumably, per Microsoft's 2014 announcement) is free to OEMs for devices with a screen size < 9 inches.
They have a very low cost license for OEMs that sell cheap devices - those get upgraded to Windows 10 for free. I can't find a source for Win 10 licensing but I'd be surprised if they aren't continuing the zero dollar licensing for a subset of devices.
I would be a cool idea for normal folks. Instead of using a laptop they can connect the phone to bigger screen for watching movies/browsing/email etc. That is mostly the case of my friends/relatives who are not in IT/Business sectors.
> Why would one need Continuum when the changes you make in Google Docs on your phone are reflected immediately on every device or PCs you have?
You're not thinking long-term enough. You should be asking, "why would one need a PC when you can just stick your phone in a dock?".
It's going to take a while for hardware to catch up, but in about a decade I think desktops will almost exclusively be used as high-end workstations for stuff like 3-D rendering, scientific number crunching, and maybe hardcore gaming, while anything with fewer demands will be satisfied by a phone using Continuum or something like it.
I think there are multiple problems with that vision.
For one you're assuming that current computing capacity is enough for most people. However current computing capacity is enough only for current applications of it. In a decade from now however people are going to get interested in virtual reality and artificial intelligence. It's inevitable. Plus the gaming industry, just like the housing industry, has always sucked all surplus.
And there are hard limits to how small transistors can get, or to how fast wireless networks can be. But problematic are also the batteries, for which Moore's law never applied. Only 5 years ago I was using a phone whose battery lasted for at least 7 days. How about your current phone?
Another problem is that no matter how powerful a phone can be, it's still handicapped by its small screen real-estate. The form factor of a laptop is optimized for productivity work while remaining portable. A physical keyboard is more productive than typing things on a touch screen. A 13-inch screen is more comfortable than a 4.7 inch screen for doing any kind of productivity work. I can go to a coffee shop, pull out my 13-inch laptop and instantly be productive. And transporting around a dock, a keyboard and a monitor to plug-in your phone in case you need it would be more impractical and would defeat the purpose of that phone.
I concede that there will be people interested in using their phone as a general computing device. Those will be the same people that are doing that today, as in the people that lack the resources to buy multiple devices. A very important and probably big market, but still.
> Another problem is that no matter how powerful a phone can be, it's still handicapped by its small screen real-estate.
With Windows Continuum, you plug an external monitor and a keyboard into your Windows 10 Phone. You get a desktop style experience on the big screen but you can continue to do phone stuff on the phone.
> For one you're assuming that current computing capacity is enough for most people. However current computing capacity is enough only for current applications of it. In a decade from now however people are going to get interested in virtual reality and artificial intelligence.
I'm fairly sure most AI will be done in the cloud. We're already going in that direction, with stuff like Google Now.
VR is an interesting question... I'm really excited about the HoloLens, and I'm really curious to see where things go from there. Honestly, I think AR might become a big deal once we start getting wearables that aren't super clunky.
Actually, introducing wearables to the mix also introduces the idea of a personal cloud of devices. The concept of a PAN has been talked about for a while, but we haven't really seen it amount to much, unless you count things like Bluetooth headsets. Cheap, stylish wearables with phone-quality processors, however, can change all that (I don't think we're at either cheap enough or stylish enough yet... but that's a big "yet").
Think beyond smartwatches or googles, too. Imagine wearing a smartshirt and smartpants and smartshoes. Imagine if every single thing you wore on your body had a CPU in it. I think we'd avoid the problem of how small transistors can get just by sticking sources of computing power everywhere. Of course, this paragraph is all pie-in-the-sky far-future stuff. Prices will need to come down to the same range as normal articles of clothing, which will take a long time. And we'll have to have a seamless way to charge your clothes that won't add any more mental overhead. Nobody wants to say "sorry, I have to stay home, I forgot to charge my pants last night".
> But problematic are also the batteries, for which Moore's law never applied.
I think at some point there's going to be a backlash against thinness, and we'll start seeing bigger devices with thicker batteries. Wouldn't surprise me if portable power sources take off too (I'm already seeing external battery packs being sold at department stores, of all places), especially if inductive charging becomes a bit less clunky. Women might be a good driver of this technology: for example, I can see a system where you just drop your phone into your purse, and it magnetically attaches to a wireless charging mat plugged into a battery pack.
> Another problem is that no matter how powerful a phone can be, it's still handicapped by its small screen real-estate. The form factor of a laptop is optimized for productivity work while remaining portable. A physical keyboard is more productive than typing things on a touch screen. A 13-inch screen is more comfortable than a 4.7 inch screen for doing any kind of productivity work. I can go to a coffee shop, pull out my 13-inch laptop and instantly be productive. And transporting around a dock, a keyboard and a monitor to plug-in your phone in case you need it would be more impractical and would defeat the purpose of that phone.
Of course, you're assuming nobody will make a laptop-shaped dock that you can insert a phone into like a cartridge. Think something like the Asus PadFone [0], but with a laptop instead of a tablet. The only real barrier to that right now is that every shell has to be matched with specific models of phone, because every phone has a different size and shape, but I wouldn't be surprised if we start seeing standardized form factors at some point in the near future.
Better yet, combine that with a setup like the Surface Book, and you'll have a phone that can dock into a tablet that can dock into a laptop.
You're not thinking long-term enough. You should be asking, "why would one need a PC when you can just stick your phone in a dock?".
Instead, ask why you'll need a phone when UI devices become invisible, ubiquitous, and networked.
Eventually, our children (grandchildren?) will laugh at our phone obsession. I'm imagining a silent reel of a Charlie Chaplin figure repeatedly patting all his pockets searching for his phone, dropping it when he does find it, pecking the wrong portion of the screen, and then when he finally answers the call discovering that the battery is dead. In other words, me, every day.
Very good point. I'm hoping the "personal cloud of wearables" becomes reality at some point in my lifetime. Reading the Shadowrun v4 player's guide seriously influenced the kind of technology I want to experience.
Probably true that phones will become good enough for most things.
But at the same time, how many devices will the average person use? More or fewer? If it's more, then no matter what -- even if we don't know what those devices will be -- you want it to be easy to switch between them, which is the problem Continuum is solving.
I think this is Google hedging its bets again and getting ready for whatever unpredictable thing happens next. They just assume it will include a proliferation of devices.
> connectors, that are big, unsightly, and brittle
You know you can run Thunderbolt over USB-C, right? There's no reason that even five years from now, I shouldn't be able to plug my iPhone into my iMac in Target Display mode using the same cable I use to charge it.
If by current infrastructure you mean WiFi, then I agree.
However, I have been using a WiGig dock with 3 1080p displays attached for the past 6 months, and it has been trivial to set up, and nigh indistinguishable from a display connected over wires.
For the record, I used to play FPS games competitively, and I still find the input lag on many IPS monitors completely unbearable to this day. For my eyes, the WiGig dock adds no discernible input lag or compression artifacts to any of my 3 displays, except in games where some frame dropping can still be observed in exceptionally intense moments.
I honestly can't wait for the day when laptops with WiGig and Wireless Charging become mainstream.
Microsoft doesn't need a credible competitor to Android at all. Anything they do in mobile is pure gravy. They're generating $20 billion a year in profit without one, and have seen zero meaningful harm to their profitability over the last five or six year rise of Android. It can easily be argued Microsoft will make a lot more money building software for Android and iOS, rather than competing with them. Office is more valuable than either Windows or Android. Azure is going to be a more valuable business than Android is for Google.
July 2009: Ballmer told the crowd, "I don't know if they can't make up their mind or what the problem is over there, but the last time I checked, you don't need two client operating systems. It's good to have one."
May 2010: Om Malik "Android Has Won — Time for Chrome OS to Move Along?"
I've been waiting for someone to adapt Aura/Ash from ChromeOS to work with a regular Linux desktop. It would instantly become one of the more polished DEs available. Maybe this could help things along.
This is a horrible day for security.