There is always a balance of security and freedom. To this point, I'd say Google has erred on the side of freedom. And at least with Android, if Google were to go rogue in a doomsday scenario, we could always fork it.

While Google probably technically could use the remote uninstall function on an app not installed via their Market, it seems extraordinarily unlikely that they would. And as you suggest, there already exists CyanogenMod and other third-party firmwares; it's fairly easy to strip Android of any connection to Google if you so desire.

More order means more economic activity.

If you have any knowledge about security and cryptography, and you start planning out the security for a new kind of app, one keeps on running into difficulties, one after another. It just occurred to me the other day that these difficulties are just like the ones 3rd world businesspeople face. As far as security goes, we are living with a poverty of infrastructure. Basically, too much of the information infrastructure is broken, to the extent that we can't do diamond-hard key management.

Secure medical information through the browser? People are in fear of regulation sets like HIPPA, because browsers are one of the richest sources of security holes! Private information? People say ridiculous stuff like "privacy is dead" and young women just hope the pics their ex-boyfriends posted are lost in the throngs of other evilly posted pics on the internet. It's insanity. How is this any different than someone in the 3rd world trying to build a widget delivery business, but finds that the roads are rotten, he can't get parts for his trucks, and he has to bribe people to let his trucks past, bribe people to let his warehouses be unmolested, and to get the business license in the first place?

In terms of security, even the 1st world is like the 3rd world. The upshot: whoever attempts building the internet's first mature security infrastructure will have huge risks. Whoever succeeds will make billions of dollars. My startup is doing precisely this. If you understand what I am saying, you can email my username at gmail dot com.

Google gets it. This is what they're trying to do with Chrome. This is how it will start. People will try to build little pockets, little pieces of infrastructure. These are the walled gardens. Some will fail. Some will succeed and even interconnect. Eventually, workable networks will securely connect the entire world.

I feel like security holes are inherent in the nature of technology and its rapid development and evolution. The security threats that plagued the internet in the 90s are now obsolete. With new technology comes new threats. That's the way it will be and the only option we have is to stay vigilant and attempt to think of every edge case before black hats do.

To say that a walled garden is the solution is a dangerous statement. While this may provide a very safe platform such as iOS, it is also a stifled and languid platform due to the bottleneck of the approval process. We will be safe from security if we never make anything new. That's not a solution to the problem that's sacrificing future potential for economic activity.

No, it's only the first step! Think about how infrastructure gets established. It doesn't appear everywhere at once. That's usually a logistical impossibility.

I feel like security holes are inherent in the nature of technology and its rapid development and evolution. The security threats that plagued the internet in the 90s are now obsolete. With new technology comes new threats. That's the way it will be and the only option we have is to stay vigilant and attempt to think of every edge case before black hats do.

Sounds like a ratrace to me. Sounds like you're giving up on the idea that things can be better. Sounds like a lot of work. Sounds like the general security situation sucks. Sounds to me like there's latent demand here.

I don't see how "more order means more economic activity": if anything, I associate "more order" with more bureaucracy and higher barrier to entry for new economic agents.

Actually, I'm thinking of doing exactly that.

I don't see how "more order means more economic activity": if anything, I associate "more order" with more bureaucracy and higher barrier to entry for new economic agents.

It's contextual, actually. I presume you are commenting from an industrialized nation. For you, more order is going to mean more bureaucracy, more friction, and less economic activity.

For someone living somewhere, where the rule of law doesn't even apply, to the point where it's hard to maintain basic infrastructure, there isn't enough order yet. In that case, more order = more economic activity.

I'm saying, in the context of digital security, the Internet as a whole still suffers from a poverty of infrastructure that inhibits many kinds of transactions. There is tremendous risk and opportunity here for those daring enough to do something about it.

Essentially, "social trust benefits the economy and a low level of trust inhibits economic growth".

You're not going to open a store if you're pretty sure it's going to get robbed. Order doesn't have to mean bureaucracy.

This is why you don't see very much inward investment in places like Russia, Somalia and so on. The barriers to entry are very low for anyone with capital and an idea, but the barriers to exit are very high. Consider the case of Mikhail Khodorkovsky, for example: http://en.wikipedia.org/wiki/Mikhail_Khodorkovsky

Bureaucrats need to get real jobs.

I think you are only partly correct: there was a natural barrier imposed upon free-ranging economic expansion right where the west coast of North America meets the Pacific Ocean.

Although I don't dispute any of the points you make, you left out the question of economic scarcity. In the case of the cowboys, if land were free then moving a herd to fresh pastures would be a simple matter of calculating time * # of animals lost per day of transit. but once you run of free land, then you have to start estimating the opportunity costs of your neighbors as an input to transaction costs, and it's cheaper to put contractual arrangements in place than to settle disputes.

This doesn't affect my analogy so much.

...you have to start estimating the opportunity costs of your neighbors as an input to transaction costs, and it's cheaper to put contractual arrangements in place than to settle disputes.

There was emotional reaction against barbed wire. The barbed wire fences were there, otherwise the contractual arrangements wouldn't work very well. With the fences, it's a lot easier to know whose land you are on and whose cattle are grazing where.

The only difference here is that it is harder to disable Android's remote push/remote kill. As is noted though it would be relatively easy for manufacturers or AOSP based projects to disable those market functionalities or provide an alternative market. Presumably if they are given a reason to want to some will.

All the kill switch did was remove the horse after the soldiers had already jumped out. But Google got super lucky and was able to also fool these particular soldiers to commit suicide.

On Android phones /system is read only. To make changes there you have to get root--then you can remount /system read-write and make modifications there. DroidDream did that and installed a root shell in /system/bin/profile. To remove the infection, Google used DroidDream's own root shell.

Now imagine DroidDream's /system/bin/profile didn't just accept any old command directed at it by just anybody (perhaps simply using public key cryptography). Could Google have removed from a non-rooted phone?

Once you're looking at human-driven innovation and technology-driven innovation, you can start to see some major differences. Human-driven pulls money whenever it brings high technology to people's lives in a more synchronized fashion. Technology-driven pulls money when it invents something so new it literally shifts the power structure into groups of those who have it and those who don't.

I find it pretty liberating that human-focused innovation is having such a renaissance right now. I don't feel so hurt if it's at a (possibly small, globally) cost to technological innovation since Apple clearly optimizes for psychology. I feel like real technological innovation can still happen as well, but a walled garden acts as a filter that separates those who just made the next poorly designed CRUD application from those who'll get bought.

I'll still add that it's a little scary since I have an engineering fear of big monolithic entities like Apple and Google over well-organized, loose-integrated, and smaller guys.

Market isn't so much a walled garden but a preferred vendor, in the same way that a Debian repository is.

  There is no "Kill switch" in Android

I didn't say I agree with it, that is just the way it works

Facebook becomes the internet you go to when you care about knowing who you're dealing with. IMO this is bigger than Google/Chrome, iOS, or Android.

There's a significant difference between code and data.

Similarly, when AOL lost to the open web, it wasn't because their code was closed, it was because their community and their content was closed. In a similar manner, iTunes needs to offer a web client if it wants to compete with Amazon, Google, and the rest in the long run. Or at least make Android/WebOS apps.