2,885 karma · joined March 14, 2012
Co-founder, Kami - kamiapp.com
You can see this with the Material Design specs as well - the rendered demo videos (e.g. https://www.youtube.com/watch?v=Q8TXgCzxEnw ) from the Design team are amazing, but what's actually been implemented on Android is far less sophisticated and even then they frequently skip frames. Interestingly, many of the Google apps are much closer to the Design specs and perform better on iOS and even on the web, at least in Chrome.
Hopefully this will finally be resolved with the Chrome OS merge (Chrome OS has great performance) - It must be extremely frustrating for the Google Design teams when they can't implement their designs properly on their own platform.
I'm not sure how good the v2 docs are right now (haven't dived into them much yet), but I'd say you are probably best off to start learning v2 now and get a head start - Egghead.io has Angular 2 tutorials up already so you could try those. Angular 2 is looking really good, and seems to be a big improvement in performance and simplicity over the first version, so I'd recommend learning it.
So the threat model is someone hijacks the firmware in a voting machine to record votes other than what the voter selected, and it's much easier for someone to flip millions of digital bits without being detected than for someone to change the box which is ticked on millions of bits of paper.
It'll be interesting to see if the zero-commission model works for them though - The 20% Uber takes is actually fairly reasonable (and much less than what a traditional taxi company takes from their drivers) when you consider CC fees & Collections, customer support costs, customer acquisition costs, marketing, etc is all covered by that. Uber is spending huge amounts of money on growing the customer base (which is good for drivers as they have a liquid market of ride requests coming in all the time), and a service like this will most likely not be able to invest as much in that since they don't have the commission and investment to pay for it.
Ironically, one of the biggest problems with New Relic is that their interface is bloated and performs poorly :)
Google's working on a new UI layer and potential backwards compatibility for older Android apps, which points in that direction:
- Chromium developers are working on a DART-based Mobile UI framework and execution engine, Flutter (http://flutter.io/). It's looking to be far better than the existing Android UI system - built for touch and 120fps from the start. This uses the Dartium VM and a bridge to allow the DART apps to use all the native features of the platform, it's much more than just another web framework. Development on this is very active right now, clearly a sizable team working fulltime - and they're building new developer tools also. There was a talk on this a while ago: https://www.youtube.com/watch?v=PnIWl33YMwA .
- Google has built a Runtime to allow existing Android Java-based apps to run on Chrome OS, and is currently testing this and working with developers to get their apps to run on it. It doesn't make much sense to invest in building that out just for chromebooks, since the experience on a Chromebook with Android apps is pretty awful (can't resize etc), but it makes total sense if it's going to be how legacy Java/Android apps run on the new Chrome based phone OS.
The sad truth is that Android simply isn't a very well engineered system - it's been improved over time, but problems persist - like the complex update process leading to unsatisfied users and security problems, poor UI performance (even now, Android can barely do simple animations at a steady 60fps on the latest Nexus devices, and has little hope of allowing for the beautiful animations the Material Design team has come up with), and poor battery life. Google's also at a dead-end with Java given the ongoing legal battles, and with Apache Harmony dead they have to maintain the standard library implementation themselves.
On the other hand, Chrome OS performs great, has awesome battery life on Chromebooks, is quite possibly the most secure end-user OS ever, and Chromebooks get speedy updates for at least 5 years. I know which one I'd choose as the basis for a merged OS.
Of course, if the merge is more Chrome OS than Android, they'd be saying exactly what they're saying now - The last thing they want is an Osbourne effect hitting the current Android phones.
- Google has Chromium developers working on a DART-based Mobile UI framework and execution engine, Flutter (http://flutter.io/). It's looking to be far better than the existing Android UI system - built for touch and 120fps from the start. This uses the Dartium VM and a bridge to allow the DART apps to use all the native features of the platform, it's much more than just another web framework. Development on this is very active right now, clearly a sizable team working fulltime - and they're building new developer tools also. There was a talk on this a while ago: https://www.youtube.com/watch?v=PnIWl33YMwA .
- Google has built a Runtime to allow existing Android Java-based apps to run on Chrome OS, and is currently testing this and working with developers to get their apps to run on it. It doesn't make much sense to invest in building that out just for chromebooks, since the experience on a Chromebook with Android apps is pretty awful (can't resize etc), but it makes total sense if it's going to be how legacy Java/Android apps run on the new Chrome based phone OS.
The sad truth is that Android simply isn't a very well engineered system - it's been improved over time, but problems persist - like the complex update process leading to unsatisfied users and security problems, poor UI performance (even now, Android can barely do simple animations at a steady 60fps on the latest Nexus devices, and has little hope of allowing for the beautiful animations the Material Design team has come up with), and poor battery life. Google's also at a dead-end with Java given the ongoing legal battles, and with Apache Harmony dead they have to maintain the standard library implementation themselves.
On the other hand, Chrome OS performs great, has awesome battery life on Chromebooks, is quite possibly the most secure end-user OS ever, and Chromebooks get speedy updates for at least 5 years. I know which one I'd choose as the basis for a merged OS.
The more useful metrics are over a time period - like the sessions per month one. Looking at percentage growth over time for a cumulative metric seems pretty meaningless actually - among other reasons, it can never be negative even if your business just tanked.
Monthly Growth usually refers to the growth in a monthly metric per month, and would be 0% in your case.
I've long felt that a major disadvantage for the Android ecosystem is that the OEMs simply don't have the resources to compete with Apple when it comes to designing new hardware, and it leads to situations where the iPhone is years ahead for certain features. One solution for that is for Google to invest in these technologies and make them available to all the OEMs, a bit like Android but for hardware too. I wonder if that's what is happening here? Looking forward to a teardown.
RE threads on MRI, I think it comes down to what your latency requirements are and the type of work as you said. I'm happy with sidekiq on MRI because it doesn't matter hugely if the process gets bogged down for 60 seconds while a ruby-heavy job runs and the other jobs run slowly - jobs are only pulled from redis when there is a spare thread, and sidekiq jobs won't timeout like web requests do. IO-heavy work will get a boost with threads on Puma, but if you get a single Ruby-heavy request running for 20 seconds serializing something, you are going to get timeouts happen.
I guess it does depend on the type of app you are running - Ours is actually quite Ruby-crunching heavy, because it has to serialize some big json responses, which is still incredibly slow in Rails. Apps which don't have occasional Ruby-intensive requests wouldn't have as much trouble with all this :)