Wizard on the MagicShifter and Open Hardware
Stefanie Wuschitz speaks to Philipp Tiefenbacher aka Wizard about his work in developing the open source hardware project MagicShifter and the Vienna hackerspace Metalab.
Can you tell us a little bit about yourself?
So I started to program when I was around 14, 15. And I was super fascinated by computers and the possibility to create stuff or simulations inside the computer, just by programming. And I think I was a little bit scared of the outside world when I was a kid. So this was super exciting. I clearly remember the following thought:
‘Oh skate boarding is so hard, programming is so much fun! Because if you do something wrong in programming, well, the computer crashes — back then the computer crashed — this was in DOS times, you could really crash the computer. I think I was really shy and scared when I was a kid.
Well a skate board can crash too, right? It crashes with you.
Yes, but there it’s much worse!
How did you get into open hardware?
Yes, that’s a funny story. Basically it was one super enthusiastic teacher in my school and later at university. That was Dr. Herbert “Happy” Hörtlehner. The Happy Lab is named after his nickname “Happy”. So he was in the HTL Spengergasse (a high school with a focus on technology) where I went. And the funny thing was that during school I successfully avoided electronics, because, to put it bluntly I was afraid of the soldering iron. It was too much in the real world! But for some reason Hörtlehner liked me, because in the theoretical and mathematical parts I was good. And so super funny story: when I was 19 and studied Mathematics and Physics I was walking near the University of Technology and suddenly a motorbike stopped next to me, and he took his helmet off and it was this professor. Then he said: ‘Tiefenbacher, won mochma denn wieda wos?’ (‘When will we again do something?’). Me and a friend of mine Stefan Farthofer, also from HTL Spengergasse went to his his house near Naschmarkt. He basically had a whole house with a flat that was fully stashed with all kinds of electronics stuff. And there were meetings each week, students and pupils meeting there, mostly building robotics. Then I found out ‘Yahoo! I can program the robots, so I can still program and they move around in the real world and that was really super exciting.
And how did you get into hacking?
Well, through the CCC or people who knew the CCC and took me and said, there is a hacker camp and you really should go there. And so I went there.
So this one teacher was really influential for you?
Yes, he was really my mentor for this. But there was also a lot of critique about this teacher, because other pupils said that he either liked you or didn’t like you. Or, it was not like or didn’t like, it was more like he somehow judged you: ‘this person is smart and wants to learn something, this person is uninterested’ and then I guess…I was quite good at school, but I heard from other pupils that they could not follow his teaching and that also they had the feeling that he didn’t like them. So he was maybe not the perfect teacher but, he was the perfect teacher for interested students. But of course you have this situation in school that the pupils are not there on their own will, you know what I mean? They have to be there. So he was frustrated with those people who just sat there, basically just wanted to get some grade and get school to be over. Also these project that you have in school, or even at university, usually we had one big project for each semester, and when the semester is over people just forget it.
Do you think that open hardware could be an educational tool that works differently from that?
Well, certainly it helps that you have open hardware designs available, but I guess you need some tutorial or introduction from some other human. When I grew up there weren’t any online resource available. Maybe you can all learn it from the internet nowadays, but back in my days I had the feeling that I couldn’t have started with this without basically him showing me things that are possible, like controllers and stuff like this.
If there is no mentor who makes you enthusiastic and curious, I guess, no internet resource helps, if you are not curious?
I think if there are good tutorials, and I think there are great tutorials out there, you need just to get the initial impulse, maybe it could be enough to see a video of someone talking about this. But to somehow get the idea. And I saw this many times, that even if you are interested… I think I was lucky that I was able to program already, because other people have to learn two things at the same time, you know what I mean? So this really helped, because I was really confident with programming. So the mysterious things were only the electronic things. Because with hardware you have a lot of bugs. And I think you have to develop a mental model to think about these things. Because you don’t see if there is voltage supply, or you don’t see …you somehow have to detect the signs…
There is a certain level of abstraction.
Yes, because it’s not so super intuitive. And I often try to explain voltage and current, but there is no analogy in the real world that truly fits. Water is a little bit similar, but it’s not the same.
So what is actually the Magic Shifter 3000?
It’s a little hardware gadget, it’s mostly open source, I publish the schematic, but not the PCB itself. The previous version was totally open source, like — everything! But it was frustrating, because basically nobody cared about it. It was not something people wanted. So its really…
You mean they were not interested in the design that you put online or to buy the actual hardware from you?
Nobody re-programmed it basically.
There was no community building around it? Of developers?
So basically in the old “shifter” the whole software was more like a demo mode. That was my idea: let’s make it open source so that people can learn programming. And stuff like this..but almost no one re-prommed it. So I was a little bit unhappy, because the user interface was not so super…So it was more meant as a programming platform, but everybody used it with the pre-programmed firmware, which was not so user friendly — also the two buttons interface was a little bit…so I realized that most people are more interested in a nice piece of hardware with a polished interface, you know what I mean?
Yes.
Also with the old version I was still learning electronics. I did not design it to be easily manufactuable, or manufactuable in a cost-effective way. So the new design can be assembled much cheaper. So I thought that the chances are higher that it will just get copied. If will be successful it might just be copied — so I just didn’t want to put the PCB online, but the schematics are online.
And specially the whole software tool chain is completely free, because now we have a different concept, you could also re-program the firmware if you wanted to. But now you can also do it in java script and so we thought maybe people are more interested in doing this on a higher level. Through the web server — we have a certain API — you can access this via java script easily. So you can re-program the firmware as well, but if you don’t need a low low level function, if you just want to set the LED to some value, then you can do that in the html java script. And this is maybe more accessible — I don’t know. I still have the feeling that 90 percent of the people just buy it for the way it looks. They are either not interested or able to take advantage of the open source aspect. So this was a little bit… a disillusion, because we were super enthusiastic ‘Yay so its open source and you can print your own case’ And then one person after the other said, ‘I don’t have a 3D printer, I don’t wanna print the case by myself’. Because it was really like an inversion of thought. Because at the first exhibition we went, it was always like ‘I don’t have a 3D printer’ and then I adapted to this and said ‘Here, we sell them with the 3D printed case already’. And I would have liked it to be an open source project and people hacking on it and not a posh product and then even when I went to the CCC congress, I think 2013 or 2014, I went there with lots of pre-assembled shifters. And there are the hackers, there are the real hackers, and they were saying: ‘Can’t I buy it without the 3D printed case? I have my own 3D printer’.
So who is your target group actually? Who do you want to use it?
Well, as it turns out our target groups are younger people — half of the people are from 14 to 20 or so..and then I think also some older people who have the money already, and for them its — they just buy it either to use it as a gift or maybe they are interested in electronics themselves. The idea for the ideal target would be — some young person who is interested in programming and hardware.
Did you ever think of collaborating with a teacher in school who has to teach java, who has to teach html? And wants to use it for teaching?
Yeah, yeah, yeah! Actually we were already talking to schools in December. But back then we didn’t have the hardware. Now we have the hardware, but I didn’t get back to the schools, at least before the maker faire. Because it took us some months to create the firmware, I always underestimate this. Specially now with the web interface you have quiet a complex user interface. In order to make everything work smoothly, like the web interface and the animation — it took us months to get rid of all the bugs! In the beginning, when we went to our first exhibition in January, it crashed and then the web interface only worked every other time and stuff like this. So I told these people from the school to please give us some time, because I do not want to frustrate people too much.
Well, yes, it needs to be stable first.
It’s frustrating for people when it just doesn’t work.
And is there now a community around the new gadget, the new shifter?
I think it’s much more accessible now, but that’s also because I’m not doing everything by myself anymore. But this was my own fault. In the old version I did not manage to deligate any tasks to anyone else. I was always thinking: ‘Too bad, nobody except me can do this, because they don’t know to do it and I don’t know how to explain it to them.’ But now we are basically three people. And one is doing the firmware and one is doing the web interface. Now the firmware is much cleaner and more accessible, because before I was doing the firmware by myself and I just didn’t find the time to clean up lots of stuff, I was kind of overwhelmed. And now because it’s more accessible and I think the difference is also that we put it on github, there are more people. But as I said, most people are not interested in the open source aspect at all. This is really just for 10 percent, this is optimistic. Most of the people want only the gadget.
And what was most surprising for you since you went to the market with your design?
Well, as I said, we made it much cheaper, the old version was much more expensive. Now we have half the prize, but I think it’s still quite expensive for many people. And I think an eye opener was this group of people from Switzerland. They said ‘Ahhh… under 100 Euro, it costs nothing, then we buy some’. So five swiss citizens were buying Shifters just like that.
And when I went to the faire in Rome the Italians waved with their hand ‘Mamma mia, this is way too expensive, way too expensive!’.
You need region specific prizes. He, he, he.
We have to sell all to Switzerland.
Income sensitive prizes.
So this was really interesting, because there is some threshold I guess. If it’s below it’s good, if it’s above then they just don’t think about it anymore. Which I totally can understand. And I really don’t like this …If I build something new, I like to show it around, but after I have shown it around 20 times it gets a bit boring to me. So I don’t really like these sale’s duties, but yeah I guess…Some people really like being sale’s persons. They have fun talking to people. And I also like to talk to people, but…
Can’t you get someone on the team who is doing that for you?
Yes, but this is …we also have to pay people for doing stuff… we don’t have the income. Financially it’s not really a success. I have a normal job, so this is more like a hobby. It’s not were I earn my money. At least for now it’s more like a hobby. We don’ sell enough to really finance this.
Metalab recently celebrated its anniversary, was it 10 years?
Yes, 2006.
What is your relation to Metalab?
Well, Metalab also got funded in part, because this Hab-Lehner died. You know, because there was a huge group of people who went there and had this space there basically. I think he died 2005. There was one group, specially Paul Böhm and also me and some other guys who were like..wanted to start Metalab and there was another group of his students and pupils who started Happy Lab. And I would say the one group was more like the nerds and hackers and the others were more like.. a little bit more straight..not so chaotic.
You went with the chaotic crowd?
Yes, exactly.
What could you recommend to other hackerspace communities, how to keep their spaces going for such a long time as metalab is already going?
Mitch Altman and other people collected some patterns and anti-patterns for makerspaces.
I think Mitch Altman had a talk at some CCC camp. These patterns have been around for a longer time and I think they are really good. Like: ‘Don’t solve problems before they arise.’ On the Metalab mailing list now they discuss problems that are purely theoretical and waste a lot of time. Generally it’s good to have do-ocracy and things like that.
But I don’t know why some things work and some don’t work. You know what I mean?
Is there anything you would like to share with the people who come to the open hardware summit?
Well, I think there are a lot of different reasons why you do open hardware or open source software. Our project is more like.. we started super enthusiastic and I think for some applications open source makes more sense or works easier. And for some applications it’s harder and I fear, that what we make, basically a toy or educational toys…just as a small example: now we found a project management software in our business for organizing teams and we found out that all these projects, there were a lot of open source projects there, they all seemed to have developed into some kind of..’ oh yeah it’s open source, but if you really want to use it you better use this web service where the original developers of the software offer the software with additional plug-ins you don’t have in the open source version. I had the feeling, specially with this project management software,usually if you are in the position that you need something like this, usually you have some funding. So people who write this as open source somehow feel they are giving money away for free. Because they don’t get paid, but they see other people using it for commercial projects. So I think there are some things where open source is easier to use. With a browser for example like Firefox..well Firefox is also a little bit commercial now. But the idea is — you have some applications where open source is easier..I don’ know — maybe software defined radio..and some applications where it just doesn’t work. And these applications where it just doesn’t work is project management, but also everything where you have consumers, who really just want to use it and don’t care about how it is made.
So I’m not sure.. I got a little bit disillusioned about this whole thing. Because when we made the shifter we were like ‘yay, it’s super open source!’ and all the other people were like ‘I don’t know how to do this..’ So we rarely found anybody who wanted to program.
It’s really like the Arduino Leonardo, the old version. It was quite easy to do that.
