So a bunch of companies decided to buy ChromeBooks, presumably without asking their employees.
http://www.amazon.com/Best-Sellers-Electronics-Laptop-Comput...
It's a shame because the numbers do actually show a very large increase in ChromeBook sales, but the tech press has focused on numbers that exaggerate and distort that.
I think that actually is just fine for most companies -- less tech support, and does everything they need (collaborative documents, email, web searches, etc).
My suggestions to consider using OS X rather than Windows have never worked because there are no low-cost OS X options.
There's no longer any excuse or need for OEM Windows machines. Same is true of Mac systems but their walls will take longer to tear down. Nonetheless, fall they will.
Some day historians will only scratch their heads about this long detour away from thin clients that we've suffered for around 40 years and the phenomenon will become a rich research area for behavioral psychologists.
The Android + Cloud model gives you a highly uniform user client which access all the fiddly bits in the Cloud. Until there's an absolutely bulletproof way of providing those services at the individually-provisioned level, that's going to win out for the vast majority of the public.
I'm not saying that the services have to remain as centralized as they are presently -- with Google owning everything (though this provides certain efficiencies). A more federated model in which there are multiple app and/or service providers to choose from _could_ come into being, and the present surveillance environment might help such an environment emerge, but the efficiencies of size and scale (as well as the very thin margins of such services) make this a stretch.
I've been watching a number of projects, most notably FreedomBox, for some time. They're pretty much precisely what you've described: cheap, self-contained, self-provisioning systems based around Linux (usually Debian and its excellent provisioning system), but there's been little noise out of the projects and progress seems slow at best.
If I could run my own servers (on an existing high-speed and highly reliable connection) without much hassle, it really would be quite attractive.
[0] https://www.google.com/intl/en/chrome/devices/chromebook-pix...
I was hoping for something between the Rolls and the Yugo
If you compare the base 1299(wifi) Pixel to a 1299MBA(13") they look about equal as far as I can see.
The Pixel gives you a higher resolution display (2560x1700 vs 1440x900) and a slight bump in processor speed (i5/1.8Ghz vs i5/1.3GHz). The MBA also has an intel HD Graphics 5000 vs the Pixel's HD 4000.
The drawback to the Pixel imo is that there is no 8GB ram option and it's stuck with a 32 GB drive (even if it is an SSD). Also, the MBA has a longer battery life (8h vs 5h) and USB3 ports instead of USB2.
It's actually the battery life that pushes me towards the MBA. I picked up an XPS13 "Sputnik" in an effort to move away from Apple hardware and was immediately disappointed with the battery life. The Pixel doesn't appear to be any stronger on this front.
If your mobile OS relies on Web apps, you might want to think about adding Android compatibility, as Jolla has done even though Sailfish also runs Qt apps.
I would assume it is different from laptop, netbook, ultabook and tablet.
To get a reliable sample of web traffic as a whole you'll need to recruit most of the larger websites (Yahoo!, Google, Wikipedia, etc.), and a sampling of second and third tier sites and then process the data taking into account that browsers lie and that all of your sites that you are sampling are going to be biased in one direction or another. Google for instance probably sees a higher share of chrome than similar sized sites. Add to which there is no standard way of distinguishing browsers other than by looking at the User-Agent string which is error prone and not guaranteed to be an accurate representation.
