Wayne is now a full time KiCad developer
https://lists.launchpad.net/kicad-developers/msg40308.html
Cool stuff.
Just one questionā¦ what does WIT (https://www.wit.com/) get out of this? Really just curious.
He probably told the people and the info will be somewhere (a link would suffice I guess) as I didnāt follow anything (buried in other stuff than electronics/KiCAD at the moment, heh)
Spoiler (answered later in thread):
And while weāre on the subjectā¦ anyone aware of developers (Iām not on the mailing list) having a discussion about how to get KiCAD from hobby open source to professional open source?
There seem to be users/ācustomersā existing that would want to use it, but they crave for a company/party they relate/talk to, if things do not behave the way they expect them to.
Iām not talking about what we do here in this forum, but rather things that free-time voluntary developers will not be incentivized to do as itās not high on their priority list (which is understandable)ā¦ bugfixes for certain features, development of features demanded by such customers, etc.
How would those works tie productively into KiCAD, which is based on voluntary free time development?
How can those works coordinate effectively with the free time developers? Etc. pp.
Reason Iām askingā¦
Hehe, now that is a little amusing, as Dave has clearly been under a rock outback.
He seems to have missed Digikey have already jumped on the KiCad bandwagon, with full time Library developers, and a Digikey library release, as well as funding for kiCad ā¦
are you sure? Their library seems awfully quiet for that. (about 15 commits is all that happened this year. https://github.com/Digi-Key/digikey-kicad-library/graphs/contributors?from=2019-01-10&to=2019-04-24&type=c Not a sign of a project that is alive.)
My guess is they invested just enough time into it to get marketing benefits and then kind of stopped working on it (seems to me more like one guy is working on it every few weeks for a day or two if nothing else needs to be done. Not one guy working on it full time.)
And their general problem is that the lib itself is not organized to be easily scale-able. Their worst problem is that they still have only a single footprint lib which will mean that they will have a major reorganization job ahead if they ever grow their lib. (This is something i told them at the start, they ignored it, i no longer care.)
Itās hard to decode some of githubās reports, but the listings here
shows 17 days ago, many updates tagged āadded 170+ partsā, tho a lot of lines have the same tag, which is suspicious.
and they also mention
āDigi-Key has a separate Partner Library located at the link below. The Partner Library contains parts submitted by manufacturers rather than parts curated by Digi-Keyās Application Engineering team. https://github.com/Digi-Key/digikey-partner-kicad-libraryā
so you may be partly right, that Digikey started, and then hopes to export the work
to their suppliers designers
Iāve not checked the partner library lately to see how active that area is.
Iād expect they need encouragement
from Digikey, and Digikey will need to provide examples and support to their suppliers, and those man-hours may not be directly visible.
The KiCad, essential is a software which source code license is the open-source GPL. The work to produce or use that source-code does not need to be free or volunteer.
People behind Digikey is certainly being paid and doing it at work hours time. Same on CERN that is contributing and maybe other individuals.
There are lots of possibilities to monetize in a source-code project. Most common will be to offer your services: develop special company needs, paid support, paid training, sell online tutorials or lessons, etc
Iām aware of that.
What I donāt know and was asking about is, how the coordination/scheduling is done for developments/bugfixes that special customers need and pay for and what the free time volunteer developers do.
With CERN this seems to have worked / is working. But can it scale?
Question might be moot anyway. The special feature, when its sold has to come with the source and be GPL again, so if someone orders a special feature it probably will be coded against a certain version of KiCAD and thatās it. If he wants to have new features from the main branch the feature has to be adopted by him. If or not the main branch uses that special feature or not (merge) is irrelevant for this.
Overall the problem of customers who need a company to āblame software problems onā and turn to, to solve them, which some open source projects do not have, is kinda an old-way of doing things that will have to adapt to the new circumstances or keep paying through the nose.
I am not so sure it even works that way if you pay for your software. Just try it. Call up autodesk and ask for them to use vector font by default in their lib. (Hint this was a known issue back in 2010. https://www.element14.com/community/thread/2944/l/proportional-or-vector?displayFullThread=true)
If you are not the IT guy of their biggest customer you will most likely get a answer of the lines āwe will look into itā with a āsorry for your inconvinienceā thrown in. With a bit of luck you get a workaround. (Which in reality is what volunteers do here on this forum anyways. Well minus the fake politeness.)
Yeah on that case, that kind of work can or cannot be integrated. But luckily, the license forces that the new source code to be open and available, so if someone wants, it can be integrated on the main branch.
Ideally that would work on a coordination, but most of the projects as they grow will get some chaotic and multiple versions or rebrand flavorsā¦
I am following for years the development mailling list and I didnāt notice any special organization.
There are some big picture guidelines and maybe some developers meet together ( eg at FOSDEM or other ).
but most of the times, for small stuff, people announce āI want to do thisā or āI did thisā. Some times that new features/fixes are accepted to the main branch, other times they are not or delayed to be accepted in future.
Plus one and hearts! Commercial software companies of course listen to their customers but not more than Open Source developers.
BTW, Wayne shares a bit more details in the mailing list thread (here itās again: https://lists.launchpad.net/kicad-developers/msg40308.html).
Hereās the meat
https://lists.launchpad.net/kicad-developers/msg40315.html
by IƱigo:
Any words about who WIT are and why their support to KiCad?
by Wayne:
WIT is a start up that is attempting to open the full stack from top to bottom (software and hardware at all levels). There website is WIT.com but there is not a lot of information there yet but there are lots of things in progress that I am not at liberty to discuss until they are announced publicly. Supporting KiCad is part of supporting the entire open ecosystem so they want to accelerate the development of KiCad. I have been the bottleneck for a while now due the additional developers that have join the project in the last few years. This bottleneck will be removed for the foreseeable future.
Hmm, their website is classic corporate-vague, and I see this detail
For those you who havenāt heard yet, I made the announcement yesterday at KiCon that I am now work for WIT and I will be working full time on KiCad.
The I am now work for WIT
means he who pays the piper, calls the tune, so in the future, the boss may decide other work is more importantā¦
Or wayne wrote that
That was what I assumed, which means Wayne now works for WIT, and is currently assigned to work on KiCad. That assignment can change at any time.
Would Wayne have really been so naive that he would have accepted an employment contract which doesnāt make him a full time KiCad worker, and then written this:
It is indeed a milestone for KiCad. Iāve known for about a month that is was going to happen but I wanted to surprise everyone at KiCon so Ihad to keep in close to the chest. It was killing me not to tell someone.
There has already been quite much assuming in this thread. Maybe we could trust what people say (Iām not talking about big companies) and not assume something bad and dishonest if someone has some commercial interest. Itās funny to see how people always give an impression that they themselves are morally superior but everyone who makes money out of something is corruptedā¦
Well in the past he worked also at some company and did kicad in his spare time. So i donāt see any problem here. (Meaning worst case is kind of what we have right now.)
Worst case would be to fork it if needed. I donāt have cycles to waste worrying about this anytime soon.
I donāt know why anyone should even entertain that possibility. Thereās no sign of anything, even in worst case scenarios based on facts, which would make that a remote possibility. Iām familiar with famous forks (emacs/xemacs, xfree86/x.org, openoffice/libreofficeā¦) and the standard countless small ones, and thereās just no reason to think that it would ever be needed with KiCad.