EOMA68 at FOSDEM2016
The normal articles and newsreports that I write are entirely third person: this is the first time in twenty years of working with Software Libre that I will be writing an article in the first person. For the past five years I have also kept away from giving any kinds of talks at conferences until now, preferring to stay focussed on the incredibly comprehensive level of detail involved in creating Libre Hardware: component sourcing without NDAs and confirming popularity and availability; designing 3D CAD, designing 2D PCB and schematics and so much more. Breaking the silence was a welcome distraction, where I had decided in advance that an actual working set of PCBs for a first prototype would be the deciding factor. The EOMA68 15.6in Libre Laptop I had just managed to get ready ten days before FOSDEM, so it was time.
FOSDEM was absolutely amazing. Busy, non-stop conversations, talks, demonstrations, discussions, creative ideas and feedback for two solid days. and that was just the Open^WLibre Hardware Laptop Stand. I cannot express enough how grateful I was to all the people that came up to visit me at the stand. John Hearns (from the Intel Design Centre) mentioning that the new up-board has the same SoC that I plan to put into EOMA68 form-factor; Wim from KDE85 for the ideas and support; Hacker Public Radio with a podcast interview - so many people, it was just amazing.
Getting people's feedback also gave me some ideas on how to bring the Laptop and the Micro-Desktop to people, one key idea being to provide kits that can be assembled at workshops under guidance. Also, having the Pi-Top and a Novena Laptop on the same stand allowed for a comparison and a counterpart, thanks to Wim.
Also it was great to meet the people from Ultimaker, and to try printing the laptop parts on their new smaller 3D printers when running experimental in-development firmware. This did not go as expected, so we tried the larger printer, and the parts came out fabulously, with none of the edge stuttering that I regularly have on the personally-assembled mendel90. Also, as they had printed them at 0.1mm the parts came out with a significantly higher level of detail than the default (0.2mm) setting that I had been using: even the inset that I had created for the 1.5mm bamboo plywood panels came out with a clearly-visible level of detail and accuracy. So from this we learn two things: firstly, that I will try printing on my mendel90 at 0.1mm resolution, and secondly, that I will be visiting Ultimaker in Utrecht at some point very soon, to help them to debug their experimental firmware, examine the gcode created by their software, to work out exactly what was going on. The end result that I hope for is that Ultimaker would, by an indirect route which is of immediate benefit to them, become sponsors of the 15.6in Libre Laptop, supplying a full set of parts for a prototype, incidentally having to ensure that their latest firmware is capable of printing such complex interlocking parts. The stable firmware is perfectly capable of doing so, but it is always nice to have test cases that identify problems in code that's in development.
Overall FOSDEM has been absolutely amazing, and just at the right time. I got to meet so many people: Alasdair Kergon, Anthony Stone, Wookey, Dr Stallman, the team from FSFE, and many new faces. I hope that I will be free next year so as to be able to do this again.