In this specific case it is also because Absinthe/graphQL subscriptions are built upon Phoenix channels so if you plan to use subscriptions, Phoenix is the easiest route.

Phoenix also is pretty minimal layer over Plug, especially if you don't include the html library.

Famously, something akin to this was done by Voltaire in 1729 (including the crowdsourcing). See https://books.google.com/books?id=BBusxgu8-AAC&lpg=PA259&vq=...

Very interesting, definitely gonna try this out as I have seen similar use-cases.

Any change Github is at anytime going to show the specific merge-conflicts for a PR that cannot be merged?

Got this book as well, it's fun to read through and I can impress friends with simple yet tasty recipes. Bought it mostly because I'm a huge fan of seriouseats and Kenji's posts especially, so I try to show my support.

I second this, also I'd like to see an overview of the (files with) merge conflicts if any instead of a message saying the PR can't be merged. Fixing the merge conflicts might be another thing in the web interface would be cool but I guess I'll stick to the CLI for this.

For completeness I would include the closed form solution in a discussion of linear regression. In matrix notation is fairly simple and it's easy to follow how it falls apart when the assumptions underlying linear regression are not present in the data you're working with (homoskedasticity, independence of errors etc.)

Playing the devil's advocate but isn't Apple hindering web standards a strategy to keep people using the apps from the AppStore. On mobile, where the iOS market share is huge and it's impossible to use any other rendering engine it would be hugely beneficial to stifle progress as native apps and thus the lock-in they provide, will keep the upper hand.

If I look at what features are provided and what are not in HTML5 (especially on iOS), I also get the impression that some features are "conveniently" missing, with the consequence that certain things can only be written as native apps.

Notably: - Network access. You can't write a BitTorrent or IMAP client in HTML. You could have with Java or Flash, but those have been axed (Apple was a major contributor to that). Of course there were good technical and security reasons for that, but it is a "convenient" side effect. - Last time I checked at least, touch input and rendering (WebGL/Canvas) was inferior to native.

Now, I'm not convinced there is a concious decision to keep HTML apps limited, but I do believe it plays a role when it comes to Apple supporting certain features and disallowing others.

Unintended side-effects like these are why I prefer to stick to normal fixtures. They get the job done and save me from adding another dependency to my projects. Fixtures are also pretty straightforward and keep me from learning yet another DSL. I'm sure FactoryGirl has valid usecases but it's not for me.

Passenger is open source and has an enterprise tier. Though written in C++ they very much serve the Ruby world.

Nonetheless, I agree with your point that this model may not be suitable for every business.

The Strava route builder incorporates the popularity of other users routes when you plan your own. So, in an extension to the normal find-shortest-path algorithm between two points, penalties are given to paths not used by other users or bonus given to those that are popular. The result is that you end up with the better roads to ride on. Especially on the bike this is really useful as I can plan new routes in unknown areas and they are pretty good.

For running I found it remarkable how these heatmaps show the inclination to run along water bodies or in forests. It can almost perfect predict where I would or wouldn't run.