Apple has long had a policy of maintaining versions of their OS's on other platforms in order to insure code portability and good architecture, even when they have no intention of ever releasing on that platform.

Apple has wisely kept the developer tools used on iOS under their control. As such "porting your apps to x86" would likely mean recompile and run through debugging on another device before it is added to the store as an option for those users. Apple is very well positioned to make an architecture transition without losing he advantage of which you speak. Note that I'm not arguing that is their intention, just that your reasoning is flawed in this regard.

I would elaborate on this point a bit. They will have a valid credit card to charge the membership. If this is a stolen card, when the owner notices the theft and/or incorrect charges, Apple can pull the offending apps from the store and revoke their encryption keys as well as begin an investigation into the risk posed to users who had the app. If the app is one that is paid for or makes money from ads, Apple has the actual account info and identifying the criminal is easy for police. Both cases provide real benefit to end users being targeted, although there is more that can be done yet.

[Citation needed.]

Noting that as of a few years ago, research showed the opposite of your claim.

Apple has already started sandboxing binaries they ship with OS X, for example the zeroconf service which is one of the few exposed default services running on OS X. At least some people at Apple seem to "get it" that just because it ships with the OS doesn't mean it should have more access or more trust than it needs. I can't imagine Apple is going to reverse their security trend and start sandboxing fewer binaries going forward.

Clearly it was time to update my Chrome install. It had fallen a bit behind since my last round of performance/compliance tests.

Seem? Seeming is rather relative I suppose. Anandtech's tests on real world laptop battery resulting the following quotes:

Arsechnica, on the other hand concluded from their real world use tests that Macbooks are just, "better than average". So I guess I'd have to ask you how you formed your perception that regular laptops have better battery performance in real world use.

Everyone copies from everyone. Tabs were in Omniweb (albeit with a slightly different implementation) in 1999. Spell checking should be implemented at the OS level and rolled out as a service to apps (as OS X does). Re-implementing it for every app is idiotic architectural design. Numerous browsers on OS X had spellchecking before Opera (since 2000) and they have grammar checking as well. Admittedly, Opera can't do a lot about the fact that OS's have failed to step up and implement spellchecking as a service, but they could at least plug into the native functionality when it is offered. There was Firefox plug-in to synch bookmarks in 2005. That's not to say Opera isn't a fine browser and even the first to bring some features into popularity, but don't go getting blinded by your like of one product. It's too easy to not look hard at the history an just buy into what you want to be true.

Funny. My plug-in became incompatible when I upgraded to 4, but 4 added the feature anyway, so I never noticed. I'm glad they finally got around to rolling it into the core application. It is certainly one of those things that benefits all users and should be a default feature.

This is interesting in an article about unique browser features. Maybe a better article would go through features rarely known about. Like in Safari you can click the "reader" button in the URL bar and it consolidates multi-page articles in to a single page including the images. There is a Firefox extension called "repagination" to do the same thing. Given how much I see people complaining about multi-page articles, it would have been nice for this article to have covered this.

Similarly, Safari and some Firefox plug-ins allow the user to grab the corner of text input boxes and resize them, which is an indispensable feature once you've used it, but was also overlooked in this article.

It is also less than useful to assess threats based upon the number of malware titles of different types, rather than the number of infections caused by different types. For example, self propagating worms have fewer unique names than trojans do, but account for a much larger number of infections per name. When last I looked, self propagating worms still accounted for about half of all infections.

How is that a great business strategy? The number of iOS developers buying Macs contributes basically nothing to Apple's bottom line compared to iPhone sales driven by their large number of iPhone apps. The great business decision is keeping the dev environment under their control and making it Mac only is just easier and cheaper than maintaining it on Windows as well. This leads to apps that conform to UI guidelines and leverage all the built in functionality of iOS and are updated in a timely manner to take advantage of new additions to iOS. Apple doesn't have to wait for third party tool developers to add features to support what Apple puts in iOS in a new release. They build it into the tools and in many cases the next recompile of the app takes advantage of the new function. That is what is smart about Apple's dev tools, not some barely noticeable increase in Mac sales from selling to developers that want to target iOS. It's about promoting iPhone sales because that is where the money is.

The idea that the term "app" was applied by Apple first with regard to the iPhone is misguided. Apps were the standard end user programs on NextStep, which was purchased by Apple and made into OS X. The ".app" ending was the file extension for the bundles. This goes as far back as 1995, at least.

On a related note, I had the opportunity to help a Linux on the desktop noob the other day and was embarrassed by how bad software installation still is for users. on Windows you stick the CD in and double click the "setup.exe" icon. He can manage this. But he has this new Linux (Mint) laptop his brother gave to him and still needs to run some Windows software for work. Here was a chance for me to show him how easy Linux has become.

So While I'd like WINE to be installed by default, I realize that is not always going to be the case. And while I'd like Linux to automatically recognize .exe files and offer to do the right thing (install WINE and run them in response to user command) I recognize it isn't there yet. But at very least I figured I could walk him through opening the package manager GUI, searching for WINE, and installing it. No such luck, as WINE only had a placeholder in the package manager. So on to the Website and click the link for Ubuntu... oops, Mint despite claiming compatibility doesn't know what to do with that. So it's back to apt on the command line which leaves him totally confused and out of his element and afraid to install software on his new machine. So one to the Windows software. Stick the CD in the drive and double click the exe. Nope no luck. it won't run claiming it needs to be executable. So I click properties and try to make it executable. Nope, fails at that too. Back to the command line to find out the non-existent metadata would be stored on the CD so it needs to be copied locally first. Finally it is installed, but no shortcut is in the launch menu and I have to dig through the WINE files to find a shortcut and put it on the desktop. And after all that, WINE can't find the USB ports for some reason. That is when we gave up for the night and started heavily drinking.

My point here isn't that Linux on the desktop sucks. It is that while it might be great for managed installations with set applications, I don't think it is very usable for the home user yet and there are a lot of really obvious areas for improvement. Hopefully Canonical will keep up the good work.

I think you're being overly generous. My understanding from talking to people in the UI design industry is that the ribbon was an attempt to implement the adaptive UI design from the University of Washington, but failed because they couldn't get it working properly so they just kept the look of that interface while abandoning the functionality. That's not to say the UofW did not take Fitt's law into account, just that there is no indication the Office developers did.

A push towards extreme capitalism is not "conservative" and I really wish people would stop applying that term. Moving to tax absurdly less progressive taxes than we had even under Reagan isn't conservative, it's extremist. Literally it is pushing the balance of economics to an extreme not seen since the days of old. Both Libertarians and Republicans (regardless of whether or not one agrees with their economic policies) are advocating for extremist economic policies in relation to historical norms for the last 50 or 100 years. Do not make the mistake of thinking they are pushing the status quo. For the last 20 years the economic balance has been skewing further and further to the extreme end of wealth consolidation.