Apps? No root? Your device serves others, warns Berners-Lee
zdnet.com
zdnet.com
"When we turn a computer into an appliance. We're not making a computer that runs only the "appliance" app; we're making a computer that can run every program, but which uses some combination of rootkits, spyware, and code-signing to prevent the user from knowing which processes are running, from installing her own software, and from terminating processes that she doesn't want. In other words, an appliance is not a stripped-down computer -- it is a fully functional computer with spyware on it out of the box." [The Coming War on General Computation]
Video: http://www.youtube.com/watch?v=HUEvRyemKSg
Transcript: https://github.com/jwise/28c3-doctorow/blob/master/transcrip...
... at the expense (decreased availability, and public tolerance) of the general ones. Possessing or promoting general purpose computing might well wind up on the FBI's list of suspicious behaviors, along side such threatening habits as paying cash for gum; "normal" people are still basically terrified of magic/the unknown. Hackers are the new witches. A plethora of embedded devices, where one general purpose computer could have done the same, would be tragic.
Of course, all of this is just my humble opinion.
As for hackers being 'the new witches,' people aren't exactly scared of computers. If anything, they take the complexity of the digital age with too much complacency.
Admittedly, I don't think general purpose computing will in any sense "disappear," but I do think that we are already in a technological depression ("where's my flying car already?") and the witches metaphor is particularly apt. Until "normal" people catch up on the details of general purpose computers ("what's a filesystem?") and feel empowered by them (confident enough to secure their own systems, and write software to automate their lives), we will continue to stagnate, and embedded devices will thrive. The proliferation of embedded devices is a barometer of the public's aversion to generality, pure and simple... I would even expect a new model for general purpose computing to result from the current Cambrian explosion eventually, and be a net win. But I think it's important to realize we spend less time inventing today than 5 or 10 or 20 years ago, and more time dealing with the political and social response to the full implications of the current breadth of computing if only because it is still entering the mainstream. Anyone interested in pushing the limits today has to focus as much, if not more, on addressing the distaste for change, as on the underlying technology (at least if they want to put their name on it). Designing and coding Bitcoin is the easy part...
The original iPhone is still useful if you are a dev, but not so useful for regular people, as many apps can't run on it any more.
* nearly always closed source, in the rare cases where you have the source you can't modify the code that actually runs when you use the service, usually no control or even ownership of your data, it runs on a machine you don't control which can make your data subject to mining by foreign governments or advertisers
You're confusing "the web" with "networked software".
When you use the web, you have no control of the server but full control of your client (if your browser is open source). When you use a networked app, you have no control of the server AND no control of the client.
"Do I want control of the software I run on my machine?" is a separate question from "do I want that software communicating with other computers that I don't control?"
Currently the number of web apps (by which I mean something with a html/css/js interface that runs in a web browser) that I can run locally with my own data that I store locally (or not, my choice) is tiny. The number of native apps that give me full control of my data is huge, because native platforms make that easy to do. Local file access APIs are relatively new in the browser.
You are correct that there are plenty of tasks that are inherently about communicating with another computer, but that is not what we are talking about here. Writing a document needn't involve communicating with another computer. Editing an image needn't involve communicating with another computer. Maintaining my todo list, calculating how much time to bill a client; these needn't involve communication with another computer that I don't control.
One might even argue the idea of individual control is somewhat at odds with the logical structure of FOSS, particularly in it's GNUist form - i.e. for a Lisper (and Stallman was/is one), there's no semantic distinction between code and data. Your position requires distinguishing bits from bits.
Solutions which distinguish bits from bits are political, not computational.
I struggle to understand how this can be true, and I say that as a Lisper (spiritually if not currently), because 'this data does something meaningful when interpreted by a computer' is exactly a semantic distinction from 'this other data doesn't'
Or to be Wittgensteinian about it, point to the part of a datum which is the meaning.
Being able to open up a word processing document without it being able to execute arbitrary code on my computer is one of the many benefits.
Oh, wait, Microsoft Word macros...
How is that better?
If I had root I could get at that data.
That's not taking away from your point that Web apps may have similar problems but there's at least 3 distinct platforms
* Open Computers with native apps (Windows, Mac, Linux)
* Closed computers with native apps (iOS, Android, XBox, PS3, ...)
* The Web
The article is ranting about closed computers with native apps.
"I don't always send data over the network, but when I do, I use an open-source browser that uses open protocols."
But "freedom" is a cluster concept [e.g. "game" - see Wittgenstein], and Berner's Lee is pointing to another aspect of what we mean by "free". Again JavaScript is a useful example - the user has the ability to turn it off in their browser if they deem that to be in their interest.
Or rather, the user has traditionally had that ability. Smartphones and other appliances increasingly are curtailing or eliminating the user's freedom to do so. And a user might want to do so based on the concerns about data you raise.
The FOSS idea of freedom is important. But the problem it solves is of another era when those accessing computers rather than terminals tended to be programmers. Stallman's concern arose at MIT AI lab, not among owners of Atari 2600's or Commodore Vic 20's.
Berner's Lee is addressing a world where our game consoles and computers are collecting and transmitting detailed data about what we are doing, and we are not given permission to turn the collection off.
While I do not deny the unsavory side of Stallman exists, I am intrigued at how smoothly we've moved from his detractors mocking his "The Right to Read" [1], to his detractors ignoring it, to us standing on the precipice of living in it, even as the mockery continues unabated and even the HN zeitgeist seems to be that he's some sort of whacko who should be ignored... some sort of whacko who, I might add, appears to have been a great deal more correct about the future than the people labeling him such.
I question whether it remains a primary issue in regard to the future impact of computing on the world. In no small part this is because Stallman's principles have achieved so much. The threats to GNU and Linux do not appear to be existential.
I am a bit troubled by the tribalism of the larger FOSS community and the times at which it focuses on ideology over problem solving and its tendency toward a tribalism which casts so many as as unthinking malevolent "others."
Stallman's principles are valuable analytical tools but they are not a hammer which also saws wood and turns screws. Some problems are better solved by trying wrenches first.
> I question whether it remains a primary issue in regard to the future impact of computing on the world.
There are more threats to freedom now than ever. Look at UEFI on new laptops, graphics cards or wifi hardware that only work with closed-source drivers, and the whole world of smartphones that do not run without closed-source binary blobs.
A proprietary BIOS has always been a thorn of contention, but with UEFI you can't boot your own OS at all unless it is blessed by Microsoft.
In the "olden days" things moved a lot slower, and some companies - like Intel - even released drivers under the GPL. Now when a new phone comes out, its SoC is 99% powered by closed source binary blobs. The Nexus 4 is a disaster - look at this list of proprietary blobs https://android.googlesource.com/device/lge/mako/+/77f8e6b51... - how can you say there's no issue here? Almost all the interesting stuff this phone can do is powered by non-free software.
In addition most phones ship with proprietary bootloaders that don't let you change the operating system. I'd say that freedom-wise, today is much worse than the situation with computers from 5-6 years ago, especially if you expand computers to include phones and tablets.
I'm actually not worried about people cackling about how they can finally enslave the masses. Generally they do such a bad job when that is a primary goal that they end up failing. (Remember DivX, the time-limited DVDs, for instance?) I'm worried about people incrementally making a small choice here and a small choice there and before you know it, we've still pretty much ended up in the same place. Yes, FOSS as a community has done a lot of stuff but it worries me to see so many people declaring the job done and that we don't have to worry any more. Using the FOSS community as the freedom backstop doesn't work if the FOSS community no longer thinks it has to worry about freedom because of past successes.
The problem of appliances locked down with spyware is an example. I believe it is a problem whose first order implications are more serious than those which FOSS (and even GNUism) are intended to address. The implications include threats to physical safety and political liberty.
I understand that there are different opinions. I acknowledge that it might be the case that the best possible solution adheres to GNUist principles. I just think that an unwavering GNUist approach is likely to an impediment to achieving a good and practical solution
I use "GNUist" because FOSS doesn't prevent lockdown - and "Stallmanist" sounds too much like "Stalinist". And like "pragmaticism" is too ugly to be co-opted for other purposes.
My Symbian Phone would load open source code written in Java, but only after it was signed. I was granted the means of self-signing, but only by Nokia's grace. And that turtle goes all the way down. The OS could have been FOSS and the device still constructed so as to require code signing for the OS.
The depth of his impact and contributions is really only obvious with the benefit of hindsight, though. Even among relative friends, it would be hard to find a time in that evolution that people weren't questioning similar issues of net future benefit, practicality and the reliance on strict dogma at the expense of practical compromise.
Based on that I'm inclined to believe his vision is as vital as ever and that the traits that at times make him and the FSF hard to work or deal with could very easily be the same ones that lead to long term benefits.
If you send a message to my computer, and my computer sends a response (or no response), how does that relationship justify your access to the source code of my computer even under the most liberal interpretation of GNUist principals?
Assuming of course that software on my computer is not GNUist. And even if it were GNUist, should I be required to notify you of that with each transaction to make you aware that you may modify the code with which you are interacting? I'll skip the logical extension of GNUist thought to the point where you can modify GNUist code on my computer.
GNU was designed around single computer, multiple users. But today's world is single user, multiple computers. The pressing problems of that world are different than those from the days of dumb terminals even if those problems are a result of the success of the dumb terminal era solutions.
(1) Dangerous insecure apps-can-trash-my-machine desktop OSes with complicated or confusing install / uninstall semantics and "OS rot" due to installs/uninstalls actually modifying things outside the app.
(2) Jailed devices, where you control very little outside the app world and someone else has root.
(3) The cloud, where you don't control anything.
I don't disagree with your assessment of the general concerns web apps, but from a practical perspective its a lot easier to provide FOSS web apps that temper some of these concerns (self hosting, etc.) while still operating within a well tested security framework. To end up in the same position with most current native mobile apps usually requires a device configured for developers with all the security relaxations, difficulties with redistribution and often program fees.
Neither one is very awesome, but its not like many ios and android apps dont wall away your data while mining it on remote servers.
"the learning curve is a bit much, and the software is not quite there yet," answers the question about how hard it is to run a server.
While I'll grant that the learning curve is a bit much, and the software is not quite there yet (and even then, I would say that the software exists, it just hasn't been integrated properly)
Setting up your own web server isn't much harder (I'd actually argue it's easier) than rooting or jailbreaking a phone. On top of that, it's legal. And once you've rooted/jailbroken your phone, how many services will you continue to use that are hosted on someone else's hardware, using closed source software? The software to host yourself is available, it's just not well integrated yet. Also, the lockin to things like FB is a not insurmountable, but difficult problem.
I apologize if I have taken you out of context.
I don't really develop for Mac, but there was a Linux tool which also shipped a 32-bit Darwin binary. I thought "I'd like to use that on my 64-bit Mac, surely I can just compile it with XCode". Downloaded XCode from the App Store (after giving my mother's maiden name, my blood type, etc.), and it doesn't appear on the PATH, because it can only write to /Applications/XCode. Jesus [2].
My point is, somehow subsidized phones turned into locked-down tablets, turned into a compiler that can't install itself to /usr/local/bin. That seems daft to me.
1. ASUS let me root my tablet for free, no strings. So it's opt in, but not terribly evil.
2. If anyone has a good Mac homebrew tutorial, I was wondering if I could compile with a different toolchain than Xcode? I'm really not up on this stuff, I need to learn.
I don't think what matters to the average user should matter to any thinking person. You're right, freedom doesn't matter to lots of people in the western world or, for that matter, in China either. Lots of people choose to be powerless or have no desire to be. Lots of people simply fail to see the long-term implications of paying more and more for less and less freedom. If control is one definition of ownership and possession is nine tenths of the law, by giving up even the expectation of control, we are all abdicating ownership for good. "Good luck with all of the of that." Berners-Lee is right (and Stallman, etc...). Deliberately and only buying freedom is the only sensible investment; anyone and anything that tries to remove individual (user's) freedoms should be dismissed as absurd, IMHO.
I agree with you generally, that we as technical people and experts need to be aware of, and create awareness of, these issues. I think you're falling into the same trap as RMS, which is to get absurd about the abstract notion of freedom, without making change attainable or actionable for the average user. If you want to have a massive shift in user attitudes towards the App store, you're not going to achieve that by comparing Apple to China. Nor can you simply 'dismiss them as absurd', because you're alienating your audience: they think Apple is reasonable.
What I liked about Sir Tim here was that he makes the argument pretty reasonably: it sucks that apps aren't cross-platform, that you have to depend on a proprietary interface to save them, share them with other users, etc. He makes it relatable in the present, he doesn't just fall back on the dystopian slippery-slope argument.
:|
ISP/Youtube (etc) copyright school, anti-circumvention laws, three/six strikes, the CFAA, etc... Prison often involves "public service"/forced labour. To people agreeing with those articulations it even seems "just": surely criminals deserve what they get (people even seem OK with rape, as long as it happens to a murderer). To people who don't agree with the current imaginary property laws, there's no practical difference between the threat of prison and the threat of a prison "camp." Suicide might even feel like a viable alternative. I know memory is short, but that was basically the front page for about a week, and only as long ago.
"I think it's sort of offensive to imply that because someone isn't aware of the implications of the app store, they aren't a 'thinking person'."
That wasn't my meaning nor was I trying to be offensive, only blunt. None the less, an uninformed person who bases his opinion on an average of the public's opinion, going no further, is abdicating thought, and can be accurately called "unthinking." (Obligatory Socrates reference... "I appear to be wiser than he, because I do not fancy I know what I do not know.")
"I'm not really sure where you're going with the possession thing[...]"
Simply but strongly: that it's not worth spending money on something you don't ultimately control (be it stocks, socks, tablets, or smartphones; don't buy shares in dictatorships either, even if they are going up: you're only betting that they will win). Everyone has that choice, and it is as important as ever. Save your money for freedom.
Wikipedia: "Possession is nine-tenths of the law is an expression meaning that ownership is easier to maintain if one has possession of something, and much more difficult to enforce if one does not." ( https://en.wikipedia.org/wiki/Possession_is_nine-tenths_of_t... )
Now is the most effective time to hang onto that presumption of control, since it's harder to get back when it's gone.
"he doesn't just fall back on the dystopian slippery-slope argument."
It's an oldie, but a goodie. While I wish it weren't applicable, I do foresee a dystopian future, at least compared to the present, where we have sold the "right-of-way" to the devices and data that augment and define us for brushed metal and an Apple logo (not to single out any one company). At least give me a bit of credit though: I didn't once mention Hitler. :)
Edit: I'm not trying to be combative. As you said, we mostly agree.
Before installing, note that from within Terminal you can use the XCRUN tool to launch compilers and other tools embedded within the Xcode application. Use the XCODE-SELECT tool to define which version of Xcode is active. Type "man xcrun" from within Terminal to find out more.
Downloading this package will install copies of the core command line tools and system headers into system folders, including the LLVM compiler, linker, and build tools.
drwxr-xr-x 3 root wheel 102 Jan 25 14:52 Documentation
drwxr-xr-x 7 root wheel 238 Jan 29 13:33 Library
drwxr-xr-x 7 root wheel 238 Jan 29 13:11 Makefiles
drwxr-xr-x 5 root wheel 170 Jan 25 14:53 Platforms
drwxr-xr-x 3 root wheel 102 Jan 25 14:54 Toolchains
drwxr-xr-x 22 root wheel 748 Jan 29 13:12 Tools
drwxr-xr-x 7 root wheel 238 Jan 25 14:54 usr
Basically the developer tools used to live by default in /Developer at the root of the filesystem, now the Developer directory is inside Xcode.
The average user will find it very convenient when their technical friends can fix, upgrade or unlock features in their device. So even if the average user never uses it, they will benefit greatly by having root access to their device.
Remember back in the day when your friends would have you fix their computer? Has that ever happened with phones/tablets? Why not?
In my experience supporting users of PCs, Macs, and Android and Apple phones/tablets, very few users have any desire for root access or any idea what to do with it when they have it. Their device is going to belong to others no matter what, maybe it's better for it to be Apple or Google or a somewhat vetted developer in a walled garden than the owner of a malicious bot-net.
I'm not sure what you mean about Xcode though. It's the same toolset (minus private headers) that Apple uses to build their own OS X and iOS apps. There's one extra step to put all of the command line tools on the path for you (install the Command Line Tools package). Then just get Homebrew and you're all set.
Open source components of Xcode (parsers, compilers, linkers, command-line tools) here: http://www.opensource.apple.com/release/developer-tools-45/
Homebrew: http://mxcl.github.com/homebrew/
Why the parenthetical? What's different about a computer you can hold in your hand that makes it justifiable for the vendor to lock it down and advocate criminal penalties for attempting to run software of your choice?
They know it's the death knell for desktop OSes though.
"They" have already decided that desktop OSes for the general public are deprecated. Steve Jobs years ago said that PCs are "trucks" that most consumers shouldn't use, and Microsoft has no hesitation about sacrificing Windows 8 usability on desktops in favor of tablets. Even Canonical seems to be headed that way with Unity.
In fact, I've seen that message now for the past few years. I'm still awaiting the day when I can no longer run my own software. I hope it doesn't come to pass.
I agree with you on the locking down issue. I don't like the ways things are going. :-(
For Mac command line compiling, You need to install the xcode command line tools package
In Xcode 4.5: see Menu->'Xcode'->Preferences->Downloads pane->Components tab. Then click the install button on the "Command Line tools" package
This should give you a working compiler and tool chain (llvm or llvm-gcc)
Hopefully, anyway.
I think the package is actually available separately from the Apple developer tools site, although I don't have a link for it. Links on the ADC site tend to change so often, it's hardly worth quoting them.
The fact is that there will always be gatekeepers, whether it be the OS maker, the firmware maker, or the chip makers -- root be damned. You'll never have true ownership of your device unless you created, understand, and own those three parts. Maybe I'm stating the obvious but, I think that phone OEMs and OS makers would do better to focus on making more function available that don't require root access. You shouldn't need root to install a tethering app or be able to uninstall some shitware that comes pre-installed by the phone carrier.
What can't you do in ML that you could in OS X 10.1?
If and only if the app signing and approval by Apple gets mandatory to run apps, a la iOS, then you'll have a point.
As for now, all the new signing etc are additions on top of the regular infrastructure, and the reason they were added was security.
Most security nerds find app signing and sandboxing a good idea.
Those same security nerds have been doing both on FLOSS operating systems (BSD, GNU/Linux) for years, so that point is irrelevant to the original discussion.
So your point is even more irrelevant to the original discussion.
What are you talking about? Nothing about app signing and sandboxing requires open-source repositories.
By using the web, my device is serving others (rather, serving me to others) on a massive scale, every day.
Without root you cannot count on being able to do that.
(1) Operating systems have an outmoded security model. Most focus on multi-user security, which still has its uses, but they fail to focus on application isolation. Applications should be installable by anyone and isolated completely from other apps unless specifically granted permission by a user/administrator. The entire malware problem can be laid at the feet of this.
(2) The (related) poor state of installability. Mac has this problem the least with drag-and-drop .app installation, though sometimes even that can be confusing (and .dmg packages are weird beasts... why?). Linux has .rpm or .deb, which applies a massive and complex band-aid to the otherwise awful state of installability on that platform. Windows is absolutely horrible... it's like Linux where you have "installers" that have to do package management instead of a formal package system.
The open source world -- and even commercial vendors that want to keep the PC alive -- have to either address this problem or accept the dominion of the locked-down vendor-controlled consumer compute device. This will require abandoning the old fashioned Unix design philosophy (and the similar way Windows works) and thinking seriously about the problems of installability and isolation. It would be worth taking cues from iOS and Android here, though there's also a lot of room for new ideas.
Oh, and I forgot to mention. If we don't address these problems, all app vendors will pay a ~30% per-sale tax to Apple, Google, and Microsoft in exchange for the valuable service of a platform that provides installability and application isolation. And you know what? The market will pay it, because for most users those things are that valuable.
Point, click, install, with no fear of damaging my system. If I don't like it I click and uninstall. Anything else is completely broken.
It's interesting to note that the prevalence of virtualization is also a sign of the failure of operating systems. OSes in a box (whether via complex container overlays like OpenVZ/Virtuozzo or hypervisors) are an ugly hack to fix the fact that the OS security model is broken even for multi-user operation. The fact that everything requires root to install is the deepest issue, along with the lack of permission structures for things like network interfaces. It should be possible to run an OS and sell accounts to the general public, not VMs, and people should be able to run whatever they want from their local account and this should be safe. The fact that this isn't viable is because OSes are broken, thus we have the huge overhead of virtualization as an ugly band-aid to fix it.
Incidentally this is exactly what Android does. They provide the isolation by creating a new user account for each app and by default apps get no permissions. The permissions you accept at install time then allow extra access. (Behind the scenes there are also mechanisms to provide security permissions across processes so that a second process doing work on behalf of your process is limited to your processes permissions.) Multiple problems have happened.
The initial default permissions in Android 1.0 were a little too permissive and for example allowed reading from the sdcard. While that seemed harmless at the time, it was in retrospect a bad choice.
There is no hard validity checker. For example if an application ignores the API call to find the location of the sdcard and instead just uses /sdcard then no one would know. This means that it is very easy for apps to not comply with best practises to be available, the developers don't even know it, and the platform has problems if they want to fix the issues due to the large installed base. (If Angry Birds stopped working because they hard coded things instead of using APIs would Android or Rovio get blamed?)
Android also made a decision that you either accept an app and its permissions, or not install it. There are good arguments both ways of only allowing them as a whole or being able to disable some permissions. But the net effect is that many apps have more permissions than they do use, and have to declare them if there is any chance they could be used. Users can choose to not install apps with too many permissions, but you can't leave a review explaining that. I see a lot more pushback over permissions in app store reviews these days, but the many years of prior practise mean far too many apps have too many permissions. (BlackBerry had this scheme where for each permission you could set it to grant, deny or ask on each use. For the tech crowd it is great, but for someone who just wants to play a game it is annoying and probably not helpful.)
The permissions themselves are not sufficiently granular - for example "Internet" is on or off. An email app contacting my email server is great, but contacting a command and control server in another country is a huge problem. Of course some sort of per connection prompting could happen, but isn't really practical for the vast majority of users.
Commercial issues have a been a problem too. For example the Google Play Store is not available in China (and many other countries). Consequently users have to use a different store, and some have different incentives as app stores are not lucrative.
Finally we have the "fragmentation" issue. Google did not implement Android in such a way that the majority of core could be upgraded by Google. Consequently you end up with full stack releases which depend on manufacturers and carriers deploying after they have already sold a device - something they have no incentive to do. This would need to be solved by aligning incentives, and is hard.
TLDR: the hard stuff is hard. Application isolation is already done by all the mobile operating systems. It is the nitty gritty details beyond that that need solving.
I don't like the user-account-per-app paradigm. It seems like an ugly abuse of multi-user access control to compensate for the missing app isolation piece.
I agree that it's a hard problem.
Preventing an app from opening other apps or arbitrary files if the user desires is awful for the users. Siloing data so that a user cannot use an arbitrary application to read a PDF is awful for the users. One-click installs and refusing to run unsigned apps is awful for the users when the only CA is Apple or Microsoft.
All of the current systems are poorly implemented. They could either be a step forward or two steps back. I'm not sure of the direction they are taking yet.
Remember software engineers have been stamping out bugs in the x86/64 arch for decades and they still find new bugs all the time. ARM arch is no different, they'll be stamping out bugs for decades too. By adding a hypervisor and virtual machines you're layering a whole new set of bugs on top like a bug sandwich. Complexity skyrockets, security is the first casualty.
As for not having root on a proprietary device it's a double edged sword. On one hand you're firmly in the land of feudal security leaving everything up to the developers, on the other hand if you root the device you're now opening up priv escalation and NFC exploits galore.
So we end up with shitty phones and tablets that try to make consuming old media easier, instead of users making their own media and applications.
There is also the Sandboxie approach to installing apps.
Once installed the OS monitors the app (and any spawned processes) for any and all changes it makes to the OS and once closed the settings are reverted (except local settings). When uninstalled any files/settings/etc it changed are also reverted by the OS instead of the uninstaller.
One criticism people have of this approach to application development is that web apps aren't as efficient as native apps.
This pro/con discussion of web vs native apps reminds me of the C programming language, why it was created and what for.
The C Programming language was designed to be a portable language, meaning that it can be compiled on virtually any platform, any operating system. Yet C was also designed to constitute the minimum abstraction away from a given platforms native assembly language.
Thus a program written in C can be compiled on almost any machine, and run as efficiently or nearly as efficiently on that machine as a program written in that machine's native assembly language.
So perhaps we could use something like the C programming language for the web. A technology which allows us as developers to write our applications once in a portable open format, without needing to sacrifice in terms of performance.
Any thoughts?
Root access would be very helpful for the tech community in doing this, but on its own, it's probably too esoteric to make much of an impression on the press or general public.
Focusing a bit more on the whole root access thing, I think it would be well to remember that we tried the open root thing already, and it turned out to be a security disaster for normal users. I think it's a good thing that the mainstream computing platforms (Windows, MacOSX, iOS) are trying a new approach. I also think that it's a very good thing that there are more open solutions, such as Android (which, just to be clear, is also a mainstream choice) and Linux, that allow us to see what can be done with more open access. Then the closed platforms can see how to go about providing that functionality in a more secure manner for less tech-savvy users.
The argument seems to be that if you don't have the ability to access whatever is the equivalent to the root account on your device at all then this means that whoever does has more control of your device than you do.
This means that you have to trust the device or OS manufacturer. For example if you buy an iPhone there is no way to "untrust" Apple without throwing the device out or doing a jailbreak.
With a more open system such as Debian you can decide to untrust the OS vendor by simply replacing the entries in /etc/apt.sources with something else.
This kind of model is often used in corporate environments where computers are locked down by the IT dept, therefor the IT dept can make choices on behalf of the users as to what the security settings should be, what software is installed etc.
To pun on Bertrand Russell: "It's roots all the way up!"
Sometimes it's OK to just rent. If I want to own, I'll move elsewhere.
The most valid argument that I can think of is that, in a "device appliance" context, many people aren't aware that they're renting. I don't suspect that they'd change to owning once they understood that they're renting, but it might make them more sensitive to the agreement under which they are renting.
Except that with renting property you have specific rights clearly defined, and there are people with guns and handcuffs and jails to uphold them. You have no rights at all with your phone, and nobody to keep your "landlord" honest. Nevermind the obvious fact that you are not renting your phone, you purchased the device, it is your sole property.
I do own my phone. I have every right to throw it off a cliff if I want to. I do not own the software that runs on it. I license it, and when I bought the phone, I agreed to that license.
Apple cannot stop me (legally) from running whatever software I like on my iPhone. The caveat is that they're under no obligation to make that easy. I bought the device knowing what software it runs. Were I to buy the device with the intention of running different software, then caveat emptor applies. Apple does not represent that their mobile devices can run any other software, so I really don't see where there's a valid argument that these "renters" are anything but consensual. I have the right to do whatever I want with the device, but I may not have the ability.
When you get down to it, the argument being made is that one should want to own, rather than rent. Go ahead and make that argument, but it's not fair to dress it up under the precept of "rights".
Not for lack of trying. Jailbreaking is only legal because of a specific DMCA exception granted by the Copyright Office, which Apple opposed. And IIRC that only applies to phones, so it actually is illegal to jailbreak an iPad.
That's the claim they want you to accept, but it isn't true in many jurisdictions, and in those where it is (US) people should be fighting to fix that mistake, not using it to excuse immoral behaviour. I do not require a license to use software. I did not agree to any license when I purchased my phone. I did not sign any contract.
>Apple cannot stop me (legally) from running whatever software I like on my iPhone.
This contradicts your previous claim that you do not own the software, and that you must abide by some license. Their license says "you can't bypass the restrictions we put on you". You can't argue that you are both bound to it, and not bound to it.
>but it's not fair to dress it up under the precept of "rights".
Why not? We have rights in other areas, specifically to protect people against predatory companies doing shitty things because "haha you agreed to it!".
I agree that we should have better laws protecting consumers from abuse, but I'm not sure what to make of "I do not require a license to use software". If I create something that I do not want to give away, I expect to be able to define some restrictions on its use. You may have a different expectation, but I don't think you have majority support on that.
No, you are assuming the only case of software restriction is limiting my OS choice. Apple works very hard to try to stop people from helping others to fix their crippled phones, so that they can install sofware on them that apple has not approved.
>I'm not sure what to make of "I do not require a license to use software"
Software is protected by copyright. Just like other creative works, like a book for example. Copyright gives the author exclusive rights to certain things, like distribution, copying, and creating derivative works. It does not give them exclusive rights to use the work. Copyright law prevents you from selling a book written by someone else (without their permission). It does not prevent you from reading it. The same is true of software. You can't sell itunes, but you are entirely within your rights to use it.
>You may have a different expectation, but I don't think you have majority support on that.
The law supports me on that.
Actually, you can't sell copies of the book. You can sell your original.
Even if you think that you own your phone, you still have to get it onto a network for it to become useful and you'll be renting that network access and you'll have no choice between contracts because they're all the same.
Not a straightforward question. As a renter, I can pack up and leave on relatively short notice, with limited liabilities. But as a homeowner, I might as well be an indentured servant to my mortgage company. If I own my home and suddenly find that I need to sell it and move elsewhere for work or other reasons, I can expect to take a six-figure bath in today's market.
Personally I would rather be a serf than a prisoner... hence my preference for renting.
2-5 are perfectly reasonable offerings and the corporations that control the data pipes have all accepted our tax breaks which contribute to them being able to maintain their virtual monopolies. My point is that the rent analogy does not work at all. It's too simple.
Most people are simply ignorant about the details because they trust an authority to keep things in order (e.g. the FTC). Perhaps trust is more misplaced with Apple and Google than with your water supplier, or perhaps those working in the food or water industry will be as wary about water quality as we are interested in whatever spyware is running on our phones - and we will find it a bit ridiculous.
If it happens to server others collaterally, well, I'm not the jealous type.
Real world circumstances render this correct ideal view mostly or entirely not true.
The browser should have stayed a document only thing.
I think the browser manufacturers will take that deal when the App Stores decide to not allow anything that doesn't actually have any purpose other than showing blobs of text on the screen.
When did a newspaper become an "app"? What new functionality does a newspaper bring to a phone that it needs its own binary code clogging up the flash drive?
To read NPR, I need to install the NPR app. To read BBC, I need to install the BBC app. To read CNN, I need the CNN app. And I need to keep them updated and if a new version of the OS comes out, they need to be updated to use the new APIs and... oh, what a waste of life. I just want to read the damn news. Perhaps next, I'll need separate televisions to watch different TV channels.
Because of bad designers. Most native apps suffer for the same reason.
>For one thing, they often lose input methods available to the underlying system.
Such as?
>For example, how many web apps use context menus? How many _can_ use them without breaking browser's own UX?
I have no idea how many do, but they don't break anything unless the designer messes up.
For example, right now in Opera I have 17 entries in context menu when clicking on a empty spot on the page, 9 entries when clicking on selected text and 15 entries when clicking inside text edit field and yet another set when clicking on a link. Some of the options control behaviour of the browser, some deal with current page, some deal with specific page element. If HN wanted to use context menus for some app-specific reason, all of those options would be wiped out (at least for affected elements) because the app would be in direct conflict with the browser.
Touch events are often another casualty -- sometimes they are commands to the OS (swipe to change active desktop), sometimes to the browser (pinch to zoom), sometimes to the app (drag to drag around a map, or maybe pinch to zoom?).
Additional layers always introduce problems and cause things at the end of the chain to have worse UX or adapt by using lowest common denominator. A crazy example: gmail running in browser running in windows running in a windows VM running on a mac machine which is remotely connected via VNC from an android tablet. >:D Android -> VNC app -> OSX -> VMWare -> Windows -> IE -> GMail app. Ok, this doesn't illustrate much besides the point that browser _is_ an element of input chain (that makes a disproportionate effect on UX too -- the other links are at least trying to be transparent), I just like the idea. :)
Of course, and you've probably seen much worse than that. Incompetent web designers are plentiful, but the capabilities of a web site are not limited to "broken sites made by incompetent designers".
>If HN wanted to use context menus for some app-specific reason, all of those options would be wiped out (at least for affected elements) because the app would be in direct conflict with the browser.
If HN wanted to do something incredibly stupid, they could do something incredibly stupid. Right. I'm not sure what the issue is there. If they wanted to not do something incredibly stupid, that is also an option. Why judge only on the worst possible scenario and assume every website is designed to be as broken and shitty as possible?
>Touch events are often another casualty -- sometimes they are commands to the OS (swipe to change active desktop), sometimes to the browser (pinch to zoom), sometimes to the app (drag to drag around a map, or maybe pinch to zoom?).
This applies equally to apps vs web, and isn't even specific to phones.
>Additional layers always introduce problems and cause things at the end of the chain to have worse UX or adapt by using lowest common denominator
That statement is false, and not relevant. The browser is no more an "additional layer" than an app is. The same number of layers in both cases.
But context menu is a useful concept (sometimes). It _would_ be nice if it could be used when it is called for. The problem is that it is impossible (I think) to use context menu or _any other right click action_ you could think of without breaking browser functionality. It would be like you said, incredibly stupid.
Thus web apps effectively lose one whole mouse button, thus reducing available input bandwidth.
Another example of lost input is lack of system-global shortcuts. A web app in my knowledge can't have them. I listen to music while I work and I tend to switch tracks or pause them once in a while. So when I listen to youtube playlists, I have to go back to the browser, open the playing tab and do it there. (and you never know if the video currently playing is SFW :))
> That statement is false, and not relevant. The browser is no more an "additional layer" than an app is. The same number of layers in both cases.
The browser is a layer, and the native-app is a layer, except to get to web-app, you need to go through the browser first and a native-app sits at the same level as the browser (so a portion of input capabilities is not used up for the browser itself).
Of course. Which is why you would use it only where it is useful, not just hijacking a particular input event globally.
>Thus web apps effectively lose one whole mouse button, thus reducing available input bandwidth.
No, you can use right click all you want, and it need not interfere with the browser's use of right click. Also, very few people use a mouse on their phone, so this does seem like a bit of a stretch.
>Another example of lost input is lack of system-global shortcuts. A web app in my knowledge can't have them.
Right. There are a few other things like this that web apps can't do. And apps that people would want this sort of interface for should not be web apps. But as I said originally, virtually all the apps people use are not of that nature (and those that are almost always are apps that came with the phone), they are just websites turned into clunky, awkward apps for no reason.
>except to get to web-app, you need to go through the browser first
The browser is the app.
Navigation
Shazam and similar
All modern mobile browsers do.
>but Instagram does more than takes a picture.
Be more specific then, I assumed that's what you thought couldn't be done. Tell me what it is you think can't be done so I don't have to guess.
Actually, you did not ask about'what can't be done', but 'example of anything other than a game that is actually "taking advantage of the underlying hardware and OS"'.
Instagram takes advantage of the underlying OS and hardware when it uses photo filters. Same as native games do. Could it be done in browser? Probably. Same as games.
You left off the rest of the sentence though, the part that said "in some way that the browser can not".
>Instagram takes advantage of the underlying OS and hardware when it uses photo filters. Same as native games do. Could it be done in browser? Probably. Same as games.
The difference is that photo filters can be done, quite trivially, in the browser. Many games can not, because the performance of the javascript engines and the limited hardware accelerated drawing.
At the time of Instagram release that wasn't the case.
One of the best uses of law is to place limits, regulations and standards on what businesses are allowed to get away with.
Same with Blu-rays.
>Trust will disrupt Openness
There are a couple of factors that will disrupt the Web architecture.
The Web has trained us to build dumb clients and centralize anything of value on the server, at a huge cost and never enough trust. We can safely predict today that light-weight protocols, mediated by the mobile OS (and its Platform) will directly challenge the Web architecture, precisely because we can leverage the platform trust model. That evolution is extremely profound.
For instance, apps running on your device can securely and privately share information without requiring a complex temporal integration involving a 3rd party service (such as Google AdSense). The information is produced and consumed on the device or the device of a related end-user. What happens on your device can now stay on your device.
Just to be clear, and to show how disruptive that architecture is, the primary key of your private data becomes your phone, not your identity. Merchants no longer need to identify you. They can’t care less about YOU, they just care to know some information about you. The problem with the Web Architecture was that the only way to do that was to associate PII to a primary key on a server and hence merchants needed to identify you to track your every move (and they shamelessly did).
The second factor is just as profound: the very open nature of the Web is driving scale over scope. The Web has successfully nurtured the largest Catalog, the largest Search engine, the largest Auction site, the largest Social Network, but I see this as a negative side effect of the Web architecture because it limits the scope of what people can do. In other words, the scope of what Amazon, Google or Facebook offer is limited by the scale (and hence the revenue) they can achieve.
I actually argue that a trust-based neutral Platform will support a more vibrant and diverse ecosystem than a truly open model because in essence a Web business couples the leve of trust it can achieve with the functionality it can deliver. The Platform decouples the trust from the functionality and it enables much smaller actors to deliver a lot more scenarios while relying on the trust establish by the Platform.
I would be surprised if the Web can resist being disrupted by the Platform. Actually, I think it already is.
The best you can do is sort-of trust the device. But that's not much better than not trusting the device in terms of the kind of architectures and products you can build.
How many get rooted without the user's intervention?
As another example: In the earlier days of iOS, you could root it just by visiting a website. Who knows how many fell pray and have malicious code running on their devices today? With a targeted attack, you don't even get the benefit of security researchers all scouring over the code like you would with something more widespread, thus there really is no way to even keep track of what might be out there.