TLDR; I used Python to create a neural network that implements an F# function to predict C# code. The network was compiled to a CoreML model and runs on iOS to be used in my app Continuous to provide keyboard suggestions.
TL;DR I used Microsoft’s Custom Vision service to train a CoreML model and wrote an iOS app in Xamarin to execute it in less than two hours. It has a loose tie-in with a popular television show. Code on GitHub. You can hear James and I discuss this on Merge Conflict.
I was writing a new language the other day and I thought, “this puppy needs a REPL”!
Drone Builder is a site I created to play with different DIY drone (multicopter) designs. I wrote it using F# and React.
I’m terrible at coding interviews - some busy bee dusts off a tricky algorithm that they studied in college and asks you to (1) originate it from a poorly stated problem and (2) live code it in front of them.
Submitting apps to the App Store is filled with many wonderful opportunities to be rejected. Let’s count them!
What you heard is true, Microsoft has open sourced the CLR and it runs on more than just Windows. They didn’t just dump some ZIP file on an FTP; no, we have another fully functioning, easy to compile, and easy to contribute to CLR hosted on everyone’s favorite file share. Microsoft has even gone so far as to setup a two way mirror with GitHub so that their internal systems stay in sync with what we see. Color me impressed.
I am writing a large iOS and OS X app in F# and am totally digging the honeymoon phase. Not only is F# a crazy powerful language, but it has a great interactive code executer built right into the IDE. In all, it’s a wonderful development experience.
There comes a day in every developer’s life when they have to admit that they just aren’t good testers. We don’t think to hit buttons in strange combinations, we test features in isolation, we don’t re-test for regressions, and we simply don’t do it often enough.
I realized that my favorite apps on the iPhone are the ones that minimize the number of buttons in their UI.
Layout has changed in iOS 6. We no longer are supposed to calculate
RectangleFs and set springs and struts (
AutoresizingMask), we are to use this very advanced constraint solving system. I wrote a library to make writing constraint-based UIs easier.
TLDR; I show how to use the
awaitkeyword in C# to build an interactive help system.
I have toiled away with the new Windows 8 OS, the new Visual Studio 2012, and the new Office 13/365 to present you, dear reader, with this fine set of charts:
I recently journeyed around Seattle to get a sense of the impact motion has on the network performance of mobile applications.
Let’s say you have a zoomable and scrollable UIScrollView all setup in your app. Great! Good job. But how do you programmatically zoom in on something particular? (For example, you may want to pan and zoom into an object that was double tapped.)
The speaker list for Lang.NEXT 2012 has been released, and I found myself trying to decide which speakers to go see. There are a lot of fine choices so I thought I had better get organized!
I submitted the Windows Phone 7 version of iCircuit to Microsoft a few days ago and was shocked (shocked!) that it failed certification by their testers.
I have recently finished porting iCircuit to Microsoft Windows Phone 7 (WP7) and wanted to see how I performed on the code reuse front.
There is a bug in iCircuit that is tarting to feel like my white whale, my wumpus. I know all about the bug: I know why it’s happening, I know all its symptoms, and I know at least one way to squish it.
My last post briefly described the work I did to port iCircuit to Windows Phone 7. In it, I described how the UI was broken into two large chunks: the graphical circuit editor and the “chrome”.
iCircuit is a mobile app that lets you experiment with circuits on your iOS devices. Well, soon, it will also let you do it on your WinPhone7.
Dr. Donald Knuth considered the problem of generating all permutations of an n-tuple whose integral elements were less than some number m in Section 22.214.171.124 of The Art of Computer Programming. Let’s look at his Algorithm M and consider how it can be used to solve permutation problems. To demonstrate the power of the solution, we’ll use C# and investigate the most idiomatic solution in that language.