Python on the micro:bit — TouchDevelop or the Mu editor?
You may have noticed that the BBC micro:bit has been launched (for the third time, I think). This time, it’s actually being to delivered to schools and teachers have the opportunity to use it with their Year 7s (11-year-olds).
Some of them will be using the Block Editor, a sort of Scratch-alike provided on the official micro:bit TouchDevelop website.
But we hope that many will choose to use MicroPython, the complete re-implementation of Python specially designed to fit onto small boards and embedded systems. A lot of work has gone into making sure it runs well on the micro:bit and has a special “microbit” module from which you can import all the tools you need to access the LEDs, the accelerometer etc. You can read about it on the micro:bit community page on the Python website.
Earlier this week I helped out at a workshop organised at Maidstone Grammar School by CAS South East to introduce some Kent-based teachers to Python on the micro:bit. One issue which arose was whether to use the TouchDevelop Python editor (created by Nicholas Tollervey) or the standalone Mu Editor (created by Nicholas Tollervey). As I spoke to the teachers at that workshop, there was an amount of confusion & misunderstanding so I lay out here what you need to know about each editor.
General
The most important point is that it doesn’t matter. Both editors have the same purpose: to provide a means to run a Python program on the micro:bit. The same code, the same version of Python, the same result on the micro:bit.
The choice will be down to how the points below interact with your preference, your environment, and the restrictions in place in your school. Also the extent to which you want to achieve commonalities with (a) other TouchDevelop languages; or (b) a local solution (eg because you’ve invested in Classroom for Github).
The TouchDevelop Editor
- The TouchDevelop editor obviously integrates with the TouchDevelop world, saving scripts which can be shared with other TouchDevelop users. It follows the same pattern as the other TouchDevelop languages and therefore benefits from familiarity with those.
- The TouchDevelop scripts are inaccessible except via TouchDevelop. (Although you can obviously cut-and-paste out of the editor but still…). You can’t, eg, hold them on Github or on your PiNet setup.
-
TouchDevelop requires you to be online.UPDATE: Peli de Halleux points out that it works offline. - TouchDevelop runs in the browser so no new software or installation is needed. (I don’t know whether it will run inside the standard browsers on the Raspberry Pi).
- The process to get your TouchDevelop script onto the Microbit is very slightly more involved than the equivalent Mu process.
The Mu Editor
- Mu does not require you to be online
- Mu works like any local editor (I found myself saying “think of notepad”).
- Mu saves files locally so does not share by default but can make use of whatever your standard filesharing solution is (Google Drive, Github, school fileserver etc).
- Mu can copy your code directly to the micro:bit to run immediately.
- Mu requires you to be able to run arbitrary software (albeit without installing) on the classroom machines.
- Mu allows you to program directly on the Microbit via the REPL. However…
- … on Windows, use of the REPL requires installation of a driver which may not be possible (at least without some bureaucracy).