Espionage is not science. blah blah blah...
Espionage can be science!
"The most exciting phrase to hear in science, the one that heralds new discoveries, is not 'Eureka!' but 'That's funny...'" - Isaac Asimov
Past Performance Is Not A Guarantee Of Future Returns...
In reality computers solder USB all the time for lower-bandwidth devices(e.g. Bluetooth, WebCam, SD reader, InfraRed, ethernet, WiFi, etc). It's dramatically cheaper and slightly more reliable than including a full USB port because you don't have to worry about charging phones(1000ma+) on a port that will only have a 250ma device on it or buy all the interconnects. The actual chips for most of this stuff are pennies. To build the same thing with interchangeable interconnects raises the power budget, the component budget, and the QA budget...
The main quote comes from a teacher who works for a think tank(that needs funding) talking about conversations he had with other teachers... not stuff he's done himself.
"I've spoken to a number of nursery teachers who have concerns over the increasing numbers of young pupils who can swipe a screen but have little or no manipulative skills to play with building blocks – or pupils who can't socialise with other pupils, but whose parents talk proudly of their ability to use a tablet or smartphone."
- Separate Office - I get bored and many questions are much more difficult to ask via Chat... why the fuck bother coming to an office building, but work from home
- Open plan with everyone scattered - interesting that you talk to more people on a day to day basis, but nearly as hard to actually communicate as when you have an office.
- Open plan with similar people grouped - You can see when people aren't stupid busy and actually talk about issues that are happening. You still run into other departments, but not as often
I'm an iOS developer.
"The Cloud" when done right is hosted servers that can (and will) move around from place to place as fast as they need to; from local servers to in-country data centers to data centers around the world in order to optimize response time and minimize down time. Just because a lot of people do it wrong, doesn't mean the concept is wrong... Just really hard to understand.
Not only is the ideal cloud hard to understand, it's very expensive and hard to implement. Just looking at the one piece of software he mentioned, Jira, it's rather difficult. Jira at least has a cloud based product, but it has different features(e.g. no project imports) which will disrupt their business and force them into different workflows... Setting up good data replication and backups can be difficult(often blind faith when dealing with these fully portable clouds) and testing portable-cloud backup systems usually requires some kind of voodoo.
The actual multiplier is ~2.7X http://web.mit.edu/e-club/hadzima/how-much-does-an-employee-cost.html