An alert from the Microsoft Download Center reminded me of the existence of Kodu Game Lab, a tool that lets users create games and learn some programming along the way.
Kodu, from Microsoft Research FUSE Labs, is a visual programming language aimed primarily at kids. The PC version is free to download and the XBox version available on the Indie Games channel of the XBox Marketplace (only in the US) costs around $5.
Kodu Game Lab can be used in the classroom, in after-school clubs and in the home to teach creativity, problem solving, storytelling, as well as programming. You can see how Kodu motivates kids in each of these areas in this overview video:
The idea is being easy to use, anyone should be able to use Kodu to make a game, a "world" in which characters and objects interact, without needing any design or programming skills.
There are plenty of resources to get users started. As well as an "official" Kodu Classroom Kit, a set of lesson plans and activities, there are plenty of tutorial "worlds" posted on its community site where Kodu creators share their games.
The most important improvement in Kodu 1.2 is its new storytelling features. A new Kodu Community Website makes it easier for users to share their games, exchange ideas and get feedback. There's plenty of activity on the site, with lots of new "worlds" being posted.
The latest release is Kodu 18.104.22.168 is a minor update, which fixes a few bugs adds a feature which according to its announcement:
takes the contents of another programming page and includes it inline on the current page. The advantage to this is that kode which you previously needed to duplicate on several pages can now be just done once and “inlined” as often as needed.
There is also a debug overlay for users engaged in tutorial creation.
So one way and another this visual programming language is becoming quite sophisticated and a good way for teachers to introduce programming into the classroom.
The only real problem with the previous release of Google's TensorFlow was that it would only work on one computer. Training neural networks is computationally intensive so the good news is that the l [ ... ]