If Microsoft ever does applications for Linux it means I've won.
-- Linus Torvalds
I've seen a lot of whining about special snowflakes always needing passing grades, but in this case I think the overrule was the correct call. From the Inside Higher Ed write up on it, this is the section that gets me:
Asked if the decision to fail every one of the 30-plus enrollees was fair to every student, Horwitz said that "a few" students had not engaged in misbehavior, and he said that those students were also the best academic performers. Horwitz said he offered to the university that he would continue to teach just those students, but was told that wasn't possible, so he felt he had no choice but to fail everyone and leave the course.
Instead of failing just the students that deserved it and giving appropriate grades to the rest of the students, he decided to fail everyone because the school wouldn't let him quit the course. So several students are doing the work and paying the tuition only to get a failing grade on their transcript because the professor wants to make a point. That's why it's getting justifiably overruled.
Rather than distribute more proprietary services, how about ownCloud for Drive, K-9 Mail for Gmail, OsmAnd for Maps, and F-Droid for an app store? Mozilla and DuckDuckGo provide Free Software search providers for Android, too. With Google neglecting the Android Open Source Project and Cyanogen partnering with Microsoft, the future for Free Software Android as anything but a shell for proprietary software looks bleak.
How much money are those services going to offer Cyanogen to be included? I'm pretty sure at least 90% of the reason for this deal is that Cyanogen Inc needs revenue and Microsoft was willing to provide it in order to increase their toe hold in Android devices. Open is nice, but the Cyanogen people need to pay the bills.
The best reason I can think of is portability. Visiting your parents and using their computer? No need to install a native app, just open the web page (hopefully in the future at least, this says you still need to install a plug-in). Want to use it on Linux? No need to wait for Microsoft to update the Linux app (if they ever bother to update it), just use a standard-compatible browser to open the web page. Microsoft wants to add features? No need to make sure it works on a bunch of different OSes and versions, just make sure the web page is still compliant with the standards (you still need to make sure the browsers handle the standards correctly, but this should be an easier target).
Not that this is perfect by any means. But nowadays computers perform well enough that most users won't even notice the problems you mention, and the other advantages more than make up for the problems.
That means you can't have any sleep-monitoring functionality in the watch, and health tracking is the only thing that will get many of us to strap something to our wrists, since watches have been made obsolete by our phones. A 24 hour charge, with a very fast charger or easily replaceable battery is the only way to make this work. Charging one battery while you're using the other, and making a quick swap is probably the best solution with current technology.
I suspect the number of people wanting to strap a device to their wrists for sleep monitoring are a very small subset of the potential smartwatch market. The changes you suggest would hurt the already struggling battery life of this thing, and make it bigger to boot. Given the small number of people that would take advantage of those things for sleep monitoring and the number of sales that would be lost due to the battery and size changes, it wouldn't be worth it.
The slight curvature also reduces visual geometric distortion. When you watch a perfectly flat TV screen, Soneira explained, the corners of the screen are farther away than the center so they appear smaller.
I have a 30" computer monitor at work, and while I like it better than my old dual-screen setup, I've noticed this issue with windows placed close to the edges. I wonder if there are curved computer monitors in the works, or if this is just for huge TVs. The main problem mentioned with curved TVs (distorted view for anyone off-center) would rarely be a problem with a screen that usually only has one viewer, and it would fix the edge distortion problem.
Computers are not intelligent. They only think they are.