I am not sure if teaching this skill would help or is possible.
But I do agree with the direction of your thought. The reading material at least needs to be inherently balanced rather than writing itself getting balanced out by 50% of articles talking about "Using proxies to prevent null pointer errors"(guilty of writing this) and the other half being "Maybe proxies are a bad idea".
>The only oddity I noticed isthe rather large 68MB binary file committed into master in the repository.
Thanks for bringing it to notice. I did this and later realized I can opt for the github release option[0]. I will fix this and the installation url asap.
I understand that this caters to a very specific audience which knows haskell. Hence I plan to support JS as the inline language. I don't want to create a new JS version because I want to learn more of Haskell :p
Not really, in the example of df, I was filtering on the basis of size, which is not int usually but a decimal value, hence I had to provide the type hint.
I agree. The AI part of the idea is relatively new with respect to the network part. To be specific, we have been working on AI since January and to put things in perspective, we started working in December 2014. All this time, we just worked on the decentralization aspect and the plugin framework.
And regarding the implementation of these two separate ideas, well, Grid is both of them at the same time and how we present it depends on what demographic are we dealing with.
"Notify me when the crowd outside [this] stall is below 50"
This situation demands the location of your device. Now, the app would hide your identity by using a hash in place of your name or MAC. This is our initial solution and if something better comes up or if a more complex situation arises then we would implement something better.
It has been just 3 years since I got my internet connection. And thus I am aware of what it means to be disconnected. Our initial idea wasn't even remotely about AI. I just wanted to make something which could act as a backup network, something which people with no internet(we have quite a share of this type in my city) could also use.
You see, its not just about privacy. A decentralized network helps you connect to people around you. And about the AI part being difficult, well, good stuff is always difficult ;)
A bus schedule would have been of great help if the system was pretty organized. But, in my city, it's a circus show. Multiple buses and jams on bus stops. To curb this issue, initially, users would have to mark the buses. This is something like the bitcoin community, where the users of the system help in operating it and in return get its services.
Chicken egg problem. My archenemy. Our solution for now, is to rely on our "Extra feature" ie Plugins. Existing startups and businesses can extend their own functionality by making their plugins. For example, there are tons of anti-rape apps on the market but almost zero which work offline. This would convert their users into Grid users.
We could also get businesses to create exclusive content or offers in return for our Ad platform/Analytics.
Just in case you are thinking that marketers/businesses will get your personal data and profile, then let me say this again, your data stays on your phone. The app in turn will filter out all the ads based on your preferences.
The system is all about crowd sourcing information. Android can tell when a phone is charging or not and phones using our app could tell the grid. So, that gives us electrical outlets in the coffeehouse.
We won't be accessing some public bus time table. Rather, the grid users riding the vehicle could give us real time and reliable data about the same.
And, a person's personal space requirements, well, honestly, that's a tough one, and it needs more thought than what we have given to our current flow.
Since, Grid is all about optimizing daily lives, we would target grocery shops, local general stores and weekly markets, initially.
We are relying heavily on the sensors provided with the phones, along with the most important, location data. We believe we could have at least an alpha if we start with these streams of data.
Sure! You could make a query like "Coffeehouses around me with with at least 2 charging points.", and it would give you a list which satisfies your criteria but would also filter based on what it learnt. In case it learnt that you hate crowded places then it would reflect that in the results.
It can also give you personalized notifications. In my particular case, I have to wait for my bus to go to college and it usually is quite packed when it arrives, so I go to the depot and take the bus directly from there. Now, this assistant of mine could tell me that the next bus is packed and I should go straight to the depot or I could just wait for 2 minutes and take a seat in the coming bus.
But I do agree with the direction of your thought. The reading material at least needs to be inherently balanced rather than writing itself getting balanced out by 50% of articles talking about "Using proxies to prevent null pointer errors"(guilty of writing this) and the other half being "Maybe proxies are a bad idea".