kasbah

I am faced with feeling pressured to add a CoC for one of my own projects in order to apply for funding. I am unsure what to put up. It seems like a massive task to me.

I quite like the CoC of the EMF camp as it seems to the point, un-ambigious and written in language I understand. I may try and adapt it for online interaction. What do others think?

https://www.emfcamp.org/code-of-conduct

I love these kinds of grid step sequencers and have had an obsession with them since I saw a Monome as a teenager. I tried to make my own capacitive touch hardware once that could work as one. Eventually I just settled on writing a MIDI Node.js one which works with a Novation Launchpad [1].

This one is really well done and gives me some ideas. Would be cool if it took advantage of bigger screens.

[1]: https://github.com/kasbah/launch-step

Thanks for the tips. I checked out the O2 amp and it's extensively documented!

For 2. it might be worth integrating with some of the part managements systems. https://github.com/monostable/awesome-electronics#inventory-...

Not yet, one day I want to add links to these new type of assemblers.

I do take requests but mostly for documenting other people's projects and getting them in a state to be put up. I was going to have a go at HackRF [1] next but if you know any good projects you would like to see put up, let me know.

https://github.com/mossmann/hackrf

P.S. I should really have a go at this power supply too: https://github.com/eez-open/psu-hw

Yes! I had this idea too (or maybe I stole it off you, who knows :). Though I changed my mind at some point and thought I would first do it according to preferred retailer (since I normally want to order at Farnell and then RS, if Farnell doesn't have it, because both will deliver next-day in the UK for free). Check 1.6 and 2.0 on the 1-click BOM roadmap [1]. Though it may be better to integrate this with Kitspace.org rather than build it into the extension.

[1]:https://github.com/monostable/1clickBOM#roadmap

Cheers! I don't think it's in competition with Tindie either. I want the site to serve a niche of creators that can't be bothered with the rest (i.e. packing and selling). I think this opens up a world of designs that don't have commercial viability. My special areas of interest are music gear and scientific instrumentation (but of course that doesn't dictate what types of designs people decide to put up).

The cost for a one-off order is always more than something that is made multiple times so I it's not a question of competing on price. Kitspace currently optimizes for one-off orders (batching orders of a single design for multiple people would be interesting too though).

I am curious, how do you think it lowers the support burden? Do you mean they can just say: "you soldered it, I don't know, works for me" or is it something more? :D

I think the stencils are best left to the batching services and Aisler and PCBway, which are linked from Kitspace, offer them too (and they should pick out the relevant solder-paste Gerber files).

EDIT: Oh, and regarding planned features check: https://github.com/monostable/kitspace#planned and the work-in-progess (WIP) pull-requests: https://github.com/monostable/kitspace/pulls

I like it, I might add that to the description. But what sets a Heathkit apart from a regular kit? Sometimes I describe it as "a place for virtual kits". (Also, didn't Heathkit make a little bit of a comeback? http://heathkit.com)

It would indeed be great to have more types of designs. We've had an influx of animal shaped PCBs recently (partly my own fault [1]) so I am on the search for some rectangular green, serious ones to offset that. I feel like it's slowly getting to a point where it's worthwhile adding categories.

[1]: https://blog.monostable.co.uk/posts/etching-oscillators-in-z...

Ah nice, hadn't seen those two. The first one I came across was [1] and I have been fascinated with the idea ever since. I kind of agree with Dave Vandenbout [2] though in that it might not scale. I kind of shifted my fascination towards using code to do PCB designs and giving talks about that [3].

Best practices, it's a good idea, but it is hard enough to get people to do the minimal thing that I came up with for Kitspace and I think only certain kinds of people will read a best practices guide. At the Global Open Science Hardware Gathering [4] we are trying to come up with more general best practices for open source hardware though and doing a more focused one for electronics would be worthwhile anyway.

About attending OHS, not unless I can get some travel funding :(

[1]: https://www.evilmadscientist.com/2011/improving-open-source-... [2]: http://www.xess.com/blog/schematics-really/ [3]: https://fosdem.org/2018/schedule/event/cad_pcb_code/ [4]: http://openhardware.science/gosh-manifesto/

Hey Abe (I follow you on GitHub!). We don't actually sell anything on Kitspace. Since about a month we have affiliate links to Aisler.net and PCBWay. So far I have earned about 7 Euros through links to Aisler.net and waiting on the numbers from PCBWay. I have spent about 4000 Euros on the site and a lot of my time so definitely running at a net loss still :)

Yes, I totally can't wait to take more advantage of the Git back-end! I like the idea of visual diffs though they can have problems (if you move the entire design by 1mm say). Just having tagged versions will be the first step and we can experiment after that.

I have been pushing people to better document the projects and helping with getting the required files into place (and sometime helping them use Git/GitHub). I have been doing what I call guerrilla BOM building (never shout this at an airport!) where I sort out an OSHW project's bill of materials and send a PR. I was doing some of this today: [1][2]. I will be applying for grants to do more of this and maybe get some help with it.

[1]: https://github.com/prometheus-science/Flypi/pull/2 [2]: https://github.com/rwb27/openflexure_nano_motor_controller/p...

All great feature requests and will be implemented one day (if I don't get hit by a bus). Linking up the board preview with the parts list, to help soldering and debugging, has been on the roadmap for Kitspace for quite a while. Here is a badly collaged concept drawing of that idea: https://github.com/monostable/kitspace/blob/bf017e245ed92066...

If you click on the "repo" link for most (not all) projects you will be able to download schematics (the majority are done in KiCad).

Haha, yeah, I just fixed up the tomu [1] BOM because of out of stocks and started dreaming up a self-healing BOM solution (pick the in-stock parts), I need to implement quite a bit of stuff before that though.

I forked the ergodox project and will have a go at helping the creator put it up when I get a chance.

[1]: https://kitspace.org/boards/github.com/im-tomu/tomu-hardware...

Thanks for the tips. I actually have a GitHub org [1] which serves as a todo list for projects to try and get into a state to be put up.

Source code for Kitspace is on GitHub as well: https://github.com/monostable/kitspace

Tindie is a really cool platform for selling kits and assembled projects and I need to figure out how to best reach people there. Kitspace is kind of trying to serve the niche of creators that don't want to put in the effort to put together and sell assembled projects or kits. There is probably still some overlap though and some people do have projects on both sites already I believe.

Kitspace is built with React and Semantic-UI (and pcb-stackup! [2]) and is compiled into a static site hosted on Netlify. I chose to implement my own build system using Ninja because I didn't get on with the frontend build tools at the time and it needs to process a lot of assets (and I do embedded systems programming by day, so I actually like Make and Make-like build tools). This was before Webpack and I had a go at understanding Webpack a few times in the mean time but never got my head around the config language.

I am currently working on moving away from a static site and using an instance of GitLab as a back-end. I am eyeing up Parcel and Next.js as possible routes for building the new front-end.

[1]: https://github.com/kitspace-forks [2]: https://github.com/tracespace/pcb-stackup/

Hey HN, I created Kitspace in an effort to help open source electronics be a bit more like open source software. The aim is to make it as easy as possible to replicate designs without the creator having to bag components to make kits.

I have been working on this for a number of years already (started with the browser extension) and have a few things in the pipeline such as integrating a GitLab back-end for easier file-uploads and adding a bill of materials editor.

The source code is available at https://github.com/monostable/kitspace (give it a star if you like it - it really helps a lot with grant applications). There is an alpha version of a BOM builder tool at: https://bom-builder.kitspace.org. Feel free to ask for more details of course :)

That reminds me that I should make the extension work for mobile Firefox. I keep wondering though: do people order electronic components on their mobile?

You can try with Digikey, Newark or Farnell. Once I find time I'll do a shared-cart trick with Mouser as well so the extension will be needed less, though the extension can also be easier expanded to more retailers that don't offer cart-sharing links.

Thanks! You can order from Digikey and Newark/Farnell without the extension though it's still a little bit better with it (less steps for Digikey and references are added for Newark/Farnell).

The extension source code is available at: https://github.com/monostable/1clickBOM and it's been extensively reviewed by Mozilla for getting it up on addons.mozilla.org.

(Source code for Kitspace is on https://github.com/monostable/kitspace by the way.)

This is the first year with a combined CAD and open hardware dev-room organized by the folks from CERN.

https://fosdem.org/2018/schedule/track/cad_and_open_hardware...

The client seems to be open source but their site doesn't. https://github.com/itchio/itch.io is just an issue tracker.

It's held on a university campus so it couldn't really be any other time.

Thanks for your answer. I'll look into it though I haven't met anyone that has used Antlr and was happy with it.

I use the "Make Medium Readable Again" extension for when I _am_ interested in the content.

https://github.com/thebaer/MMRA

Thanks for you answer. To clarify: Freegeopip is not my project but it's pretty easy (for someone who knows how to run a Linux server) to host your own instance.

I host my own instance of https://freegeoip.net. Any advantages this would have over that?

For electronics check out my list of resources. https://github.com/monostable/awesome-electronics

Signal can now work without GCM, they merged an alternative that uses websockets.

I will reiterate some of what others have said:

- Use version control!

- Hire a professional software developer

- Good architecture and abstraction comes mostly from practice but if you really want to do some reading on software architecture check out "The Architecture of Open Source Applications" (vol 1 & vol 2)

http://www.aosabook.org/en/intro1.html

While, as others have pointed out, that with 100k non-paying users the $1,600 figure makes more sense. Your example still makes no sense to me, what are you _actually_ doing with all that server power?

Is it me or is it nuts to be paying $1,600 a month on servers when you have just over 1000 users? Seems like something that could be run on a $50-$100 VPS.

Another great resource for learning constraint programming is CS294 "Programming Synthesis for Everyone" [1]. It focuses on using Racket and Rosette [2] to synthesize programs.

[1]: https://homes.cs.washington.edu/~bodik/ucb/cs294fa12.html

[2]: https://emina.github.io/rosette/