Yes, I also have to do a 'double-take' on these names. I know it's not the point of your comment, but ReactOS started first for those who don't know already.
Because they actually know how to implement fault tolerant scale on AWS. Most companies don't have simian army randomly wrecking infrastructure. Many companies/engineers think they can spin up a few (or few hundred) ec2 instances and think they have solved resiliency.
They are shifting capex to opex. The complexity of solution that can perform OK on aws is significantly higher then complexity of solution that runs on metal. There are use cases for which aws will never be a viable option. AWS in general is a bad option for anyone but Netflix (Netflix consumes over 30% of aws resources if they make a mistake that causes even 30-40% spike all AWS customers are f#$ed)
Does anyone really think Android phones won't be competitive at the FirefoxOS price point? The Moto E is already under $100 unsubsidized, and that's a well-known brand name. Generic imports are even cheaper, and prices will continue to fall.
It's not impossible, but Android phones need to do more - they run not just a web stack but also the Android Java stack - and they have not optimized for the low end, as much so far.
While Android runs on a lot of high-end hardware, it was originally targeted at devices that are under-powered compared to the cheapest smartphone chipsets now available.
Android also uses implementation approaches like a component lifecycle, the Zygote, and limited per-process heap size that reduce runtime memory use at the Java runtime layer.
The people who designed the Android runtime had also designed the Danger system, which used Java and ran in as little as 32MB of RAM.
While it is possible for an app developer to make an inefficient Android app, it is very likely that core phone functionality, like dialers, messaging apps, PIM apps, etc. are a lot more efficient implemented as Android apps than as Web apps.
Well, first of all Android has changed a lot since those early days. The competition has focused on higher-end devices to challenge the iPhone, not the low-end.
Second, it might be true that a single app could be more efficient as a native Android app than a web app. But regardless of whether it is (and I'm not sure it is - we would need to measure), the issue I mentioned is that Android phones have to support both native apps and their entire graphics stack etc., and the web platform and its entire graphics stack etc. Both because users can run both Android apps and a web browser, and because Android apps can embed a web view, so the combination happens even in a single app.
That fundamentally adds overhead. Of course, in theory massive amounts of work could remove it (you could unify both stacks on a single graphics codebase, and to some extent that is true on Android), but that might introduce compromises as well, and no one wants to compromise the high end which competes against Apple.
The bottom line is that Android doesn't ship super-low-end phones. That might not only be due to technical issues like these, of course.
Those are still much more expensive than the lowest-end. I think they cost around $100 last I heard, and a currency conversion on the amazon link there gives $104, which seems to confirm they haven't gotten cheaper.
If you sort this site by price http://www.snapdeal.com/products/mobiles-mobile-phones/filte... you find the cheapest smartphones are about 1500rs. That's about $25 (according to Google). Most of them run Android. Though, once Windows Phone 10 comes out, I'd be interested to see if that's a better UX at the ultra-cheap end of the spectrum. Nevertheless, Android is currently the OS of choice at the low end, wherever you draw the line on acceptable UX. And the latest and greatest Android is available for about 4800rs, or about $77, running on a quad-core SoC.
As for how that's done, it's because Android can aggressively "swap" (really serialize and reconstitute)background components out of memory, and it can do the same to whole processes. You can actually watch your Android app instance switch PIDs, as it comes back from having had all the components, and the underlying runtime instance, killed. Now, there are some fat slovenly apps out there that might not be happy with life in a very limited machine, but I bet those 1500rs phones push the minimum specs pretty damn hard.
It's amusing that the original (native) version ran smoothly on a 286 with EGA but this JS version gets the CPU running hard enough to turn on the fans on my 2014 MBP.
If each domain name can get a non-wildcard cert for free, quickly, why do you need wildcard certs? For multi-subdomain hosting on one server? Just wondering.
For my previous use cases, it's ideal for dynamically created subdomains of a web application. If I know ahead of time, it's easy to grab a cert for any subdomain. However if a user is creating subdomains for a custom site or something similar, it's much nicer/easier to have the wildcard cert.
The lets-encrypt demo makes it look like you could easily script cert acquisition for new subdomains. And the CA domain validation appears to be totally automated (and fast).
Lots of services create dynamic subdomains in the form of "username.domain.com". To offer SSL on those domains without a wildcard certificate, you'd need to obtain a new certificate and a new IPv4 address every time a user signs up. You also need to update configuration and restart the web server process.
It's funny that Microsoft is advertising the Surface as a sort of iPad++, while at the same time making full Office free for the iPad, but not for the Surface. (I know, Surface Pro is a full Windows PC and the iPad isn't, and Windows Office has more features than iPad Office... Just saying the marketing is kinda inconsistent with the sales approach.)
Pretty much every ad aired this year for the Surface Pro 3 compares it directly to a MacBook Air, not to an iPad. (Literally--the MacBook is on camera alongside the Surface Pro in many of these ads.)
To the extent that you perceive the Surface advertised as iPad++, you're probably thinking more of the original Surface ads one to two years ago (which focused on the regular Surface, which did in fact ship with a free copy of Office).
