ptorrone
Active member
yes, i see the brand-new accounts that paul is now allowing whose sole purpose is to “dunk on me.” that pattern is obvious, and it’s familiar. historically, a lot of this behavior targeted limor directly. now it’s sliding over to me because i hold a mirror up to it and i push back.
limor is moving to kicad. it will not matter to dudes
yes, this board was initially designed in eagle. we have been publicly moving to kicad, actively, for a while now. that transition is non-trivial when you have 800+ open source hardware certifications and a very large back catalog of designs. migrations take time. pretending otherwise is either naive or disingenuous.every design limor makes can be opened, viewed, and edited, and manufactured in kicad, you all know this.
open source hardware has never meant “must be created in a specific tool.” it means the design files are published, modifiable, and reproducible. that’s how the definition was arrived at, and it took years of community debate to get there.
arduino published designs in orcad, which is far less accessible than eagle. and eagle has a fun side-story: limor and i personally spent a lot of time with the old eagle owners pushing for xml import/export so designs could move between tools. so yes, the “closed tool” complaint is especially rich. no good deed goes unpunished, and apparently it gets reposted by a long line of dudes who all think they invented the same talking point.
eagle became a de facto standard precisely because it enabled widespread sharing. many foundational oshw contributors used and still use non-kicad tools. bunnie huang does not use kicad. that does not invalidate his work as open source. purity tests like this are for people who can’t be bothered to argue about the actual engineering.
we publish full design files.
we publish firmware.
we publish documentation.
we support migration and interoperability.
and we donate to and sponsor kicad every year, promote their fundraisers, and show up for their events. anyone is welcome to check that record. it would be great to see more organizations do the same instead of posting drive-by takes.
on nordic’s “closed source”
out of thousands of designs, the nordic bluetooth stack gets recycled by the same three people (and, yes, their multiple accounts), so let’s be explicit.there was a vendor-supplied firmware sdk that nordic would not release under any terms. we documented that limitation, published what we could, pressured the vendor for years, and eventually moved away when better options existed. that is what responsible open hardware stewardship looks like in the real world, not the imaginary one where vendors comply on demand because someone yells “open source” loud enough. turning that into “gotcha, adafruit is secretly closed source” is dishonest. it ignores the paper trail and the outcome.
on teensy, compatibility, and why this exists at all
we did not wake up one day and decide to “attack pjrc” or “advertise on their forum.” we were informed we would no longer be able to purchase a core product we had sold and supported for years. that forced a decision. full stop.we are not using the teensy name on a product. we are not claiming drop-in equivalence. we are exploring an open hardware alternative that prioritizes different tradeoffs. some people will want it. some won’t. that is normal.
if your definition of “teensy-compatible” is “runs my existing firmware with zero changes,” then no, this is not that. if your definition is “open, hackable, documented, affordable, and in a similar performance class for many use cases,” then the discussion is reasonable. both positions can exist without dunk on phil accounts and the personal attacks. stick to the technical, we are.
open source certifications are how most people agree something is open source
during covid, when things were brutal for everyone, we chose to keep paying our team for months. we also leaned hard into open source hardware certifications and did the work.adafruit is not only heavily certified, but limor’s libraries are among the most widely used in the space. she even received a github award recognizing that work. certifications matter because they create a shared baseline for what “open” means, instead of forum snark.
and since people love selective memory: sparkfun had an open source certification revoked for not being open source. that happened in their own community. the people who were unhappy with sparkfun’s response asked me privately to help. i emailed sparkfun’s cto. nobody was happy, except the people who wanted the product advertised as open source to actually be open source. i’m sure that made sparkfun like me even less, if that was even possible.
https://oshwa.org/announcements/revoking-certification-for-us002346/
on tone, purity tests, and “dunking”
accusations about my home life, my family, or my motivations are not arguments. neither is pretending that mentioning time, health, or caregiving is “manipulation.” those are just facts of life.what is consistent is that the standard for “acceptable behavior” and “open enough” gets applied very differently to limor than to most people in this industry. that’s not new.
we’re going to keep making hardware. we’re going to keep publishing designs. we’re going to keep moving to kicad. we’re going to keep supporting open tools and open standards. people are free to like the product or ignore it entirely. if there’s a path to resolving issues with sparkfun or pjrc constructively, we remain open to it. if not, we’ll still ship boards, answer questions, and show our work.
that’s it.
