welcome to December 1st! Not only the year 2019 ends soon, also the 2010’s will be over and the 2020’s will start! Some time ago, maybe it was 2017 or 2018, I said to myself “this month I’m going to learn Python!”. Which was great because after this month I said “this year” and now I am at “this decade”. How time flies! But now I’m actually in it for quite a while now and become more and more familiar with this language.
But how? And, why?
I started to google “learning python as c++ dev” and found the best quote I read so far. I can’t find the link anymore but it was sth like:
“If you learn a new language, try to learn it as you’ve never coded before. That’s how you get the spirit and soul of the language. If you want to code Python like C++, you don’t need to learn Python.”
I tried different tutorials on youtube and finally came up with buying this book. Online-Tutorials are great, but I still like to make notes in books, putting post-its on the pages and organize my brain in this kind of way. Otherwise I’m only on Google crawling myself through tons of answers – with the book, I kind-a know where the answer is. Still, searching the internet, especially stackoverflow, is the way to code:
So, why did I want to learn Python? For programming microcontrollers it’s almost useless. Games? I’d always recommend Unity or Unreal. Dataviz? vvvv or Processing, maybe also Unity. Interactive installations? Depends, but I mostly did them with vvvv.
But for everything big data and AI-related, Python is the way to go. There are the libraries.
It’s fun to test Yolo v3 in Python and I’m looking forward to find time to try CornerNet-Lite, which seems to be the state-of-the-art right now for realtime object detection(?). Of course these libraries get adapted and wrapped for other languages as well, but I sometimes just don’t want to wait.
I recently did some tests with NLP (Natural Language Processing) and tried to make a chatbot to “talk with Sarah Kane” or at least, with what she wrote. My first attempt was also to do it in Python, and as Python has a really big community, it has the library Natural Language Toolkit (NLTK). I read this tutorial on towardsdatascience.com and the hardest part was to convert Sarah Kane’s scripts from pdf to txt (and you could also read pdfs directly to python, of course). Again, my body was filled with dopamine because this worked out so well!
Coding on the Raspberry Pi, using the GPIOs, Sensors, building web-applications? From now on I’m doing it in Python! Ha!
Another great multipurpose toolkit for coding I unfortunately neglected in the past two years is vvvv. vvvvhat you can use it for is nicely shown in their brand new 8-minute showreel:
It’s a node-based language, meaning you don’t type any code but use blocks aka nodes, that you connect. The basic concept is easy to understand and yet you can make great visuals and almost everything with it. The lead developer of one of my favorite companies in the current media-art-scene, marshmallow laser feast, told me recently while eating a Schnitzel that he can’t code. He’s doing it all with nodes in vvvv. Jaw-drop.
Great thing about vvvv is also it’s community. Not as big as the Python community, but some really talented people doing a great job. Some of the contributions are free to use, for some of them you have to pay (fair!). Same with vvvv, by the way: If you use it for non-commercial projects it’s free, if you make money with it it’s only fair to pay the developers for this great tool! By the way, vvvv is developed in Berlin and they give free introduction workshops!
Next chapter, what happened during this week?
The Spiel && Objekt students were happily in Essen at the Next Level Festivals for Games 2019 presenting their works! Even the television (even the television!!!) was there to make a short clip about it, check it out:
So far so good. Take care for yourself and the people around you!