I’ve been working on a new feature for Visual F# Power Tools extension, which would creating and maintaining folder structure within F# projects. I have to admit, it was not a pleasure to dig into VS SDK and try to get it working. The most annoying part: part of SDK related to Solution Explorer still uses COM interfaces… Second most annoying part: Solution Explorer complaining about completely correct project structure…
I’ve been trying to learn F# for quite a long time now, but there was never a good way to do it. First of all, I need some way to evaluate my work. It’s really easy to learn new language, but you never know if you’re using it right. That’s even more likely to happen when not only language is new but also general idea behind that language is much different. That’s the case with my F# learning. Almost every programming language I’ve used so far can be classified as object-oriented-first language. F# is different. It’s functional-first language and because of that it’s more about learning how functional programming looks like, not how F# syntax looks like. But I think I found a way to learn F# right. And the answer is: Open Source.
I’ve started learning F# couple weeks ago, but unfortunately after few days I got quite busy and after these days I still know almost exactly nothing about functional programming at all. That’s why I decided I have to look for some challenges and samples I could work on while learning. That’s how I found F# and Machine Learning Dojo slideshow and Digit Recognizer competition on kaggle. I decided to give it a try. Of course, the main goal is to learn F#, not to get 100% correctness in the competition. You should have it in mind while reading the post :)
I just watched The Future of C# talk by Mads Torgersen recorded at NDC London conference over a month ago. Among all the stuff Mads talked about one really drawn my attention: methods/properties with multiple results. Unfortunately, instead of providing some new, crazy idea how to make it possible in C# he focused on making current usage easier. I think language designers could go much further and create real multiple results experience!
I’ve spent last two days working on my first open source .NET library named CloneExtensions. It gives you a smart way to clone your object instances without implementing any interface writing any additional
Clone method at all. It uses Expression Tree to compile that
Clone method for you right before you’re trying to use
GetClone for given type
T for the first time.