How GitHub Changed FOSS: What’s Missing in OSHW?

We are blitzing our go public release and are super cramped for time right – so today’s is a short post. Honza Pokorny wrote a great article on how GitHub changed FOSS, see here. And I liked it so much I wanted to draw some parallels to OSHW. Here are my 4 favourites and the ones that I think are very applicable to our community.

Unified place for all your projects

There isn’t one right now. We need this badly, but it also needs to support, or integrate with tools for actually doing hardware. It can’t be a dead space, where projects go to die.

Design discussions

I want to collaborate on designs, and I want that collaboration tracked. I want to look back and understand a bug fix, or why a resistor is 1K. Nothing makes this possible right now, and its a barrier to real collaborative electronic design.

Publish your hacks with ease

This is huge! I need a zero friction way to share my hackery, or else I just wont! The project sites that exist either need to focus on lowering this barrier, or someone else needs to fill the void. Again tool integration is huge here – but is possible. So many great hacks are wasting away in basements!

Discover new projects

Why is this so hard? The harkopen guys are helping, and instructables is good too, but I want a hackernews style occasional, cool project of the day, or week. Hack-a-day does a good job of this on the epic side, but I want stuff that I can obtainably build and make better.

Here at Upverter we are fighting to make these headaches go away. But its a hard fight, and the more people in the space working together the better we can all be. So I guess this is a bit of a cry out for us all to get better, and a bit of a push for stepping back and looking at the landscape. Lets build a github, or at least lets do the same things they did right for OSHW.

The Pitch: Upverter in 30 Seconds

The Pitch: Upverter in 30 Seconds

  • Hi I’m Zak, Mike, Steve and we’re Upverter.
  • We make web-based software for designing electronics.
  • The existing tools were all designed before the web and isolate their users.
  • But because its on the web our software allows collaboration and sharing; designers can finally work together.
  • We are all electrical engineers and we’ve built the software we always wished we had.
  • There are 10’s of billions of dollars in our market.

March 2011 Update

March 2011 Update

So my ma calls me the other day and tells me its -27C at home (Toronto, Canada). And right after cringing and being thankful we aren’t up there right now. We got to thinking, isn’t it supposed to be spring!?!? Like snow melting, flowers growing, spring break = spring? We don’t really know if its spring or not because the weather never changes much here from day to day (Mountain View, California). But we are pretty sure that one one of the things we learned in grade 3 was that girls punch you when they like you, but also that March = spring, and spring = new and fresh (not -27C!). So in the interest of spring, of new, fresh, and exciting, we figured it was about time we sent around another email!

The first big piece of news is that we have finally got users! About a week ago we launched our closed Alpha, which is super exciting for us. To be finally working with our users building the tools they want and need, and to have this incredible base to build on top of is just awesome. We are working tirelessly to add every feature requested, respond to every email, and get the tool to the point where its useful and not just super cool. And I must say we are so grateful for all the incredible feedback! Thanks again to our Alpha users – we really do appreciate it!

So in the interest of setting ourselves up for a lot of sleepless nights ;), I thought I would layout the next big milestone for us. We want to go public as soon as possible. We want to scale, and grow and provide value to as many users as we can. But right now the tool isn’t ready. The following are the tasks and improvements that we are working on right now, and the features that we think once we have, we will be able to open up to a much wider audience. We will probably call this our Beta.

  • Attribute labels

    : This one is a pretty trivial issue. It should be done in an hour or so ;), but without it, its pretty hard to do anything useful!

  • Part editor / Symbol editor / Attribute editor

    : This feature will allow people to start contributing content to the system. You will actually be able to get in there and create the parts the library is missing.

  • Export Netlist / PDF

    : We think this is pretty important to people, and we think its going to be pretty fast to churn out. It might not be super complete early on, but this feature will allow you to add Upverter to your workflow.

  • Public Embed-able Viewer

    : We think this is one of the secrets to getting a ton of people interested in what we are doing. This feature would allow users to add a schematic to their blog, or instructables posting.

  • Import Eagle / Altium Files

    : We are scared of this one. We have no idea how much work its going to be. But as our most requested feature we are working to add it in as soon as possible.

And thats really it for the update! We are continuing to invite people everyday to join the tools and give us feedback and so far the response continues to  be incredible, people really are dying to have better tools! We are here in Mountain View hacking away for another month, and we are enjoying our time, learning lots, and meeting tons of incredible people. It really is a very powerful startup scene down here, and we are so grateful to get to participate in that. Oh, and if you haven’t yet gotten an Alpha key, and you want one, send me an email and we will get you in there right away! Cheers, The Upverters

Interns: How to Hire Porn Stars

Do we actually want to hire porn stars? No. Its a joke stolen from a recent hacker news posting. And despite being flamed a bit in the forum the author made some pretty good points about how funny the job titles we give hackers are. Avery has some thoughts on the interesting job titles and reason for them, and honestly I like his titles way better than the generic ones. And at the end of the day we don’t really want Rock Stars or Ninjas but something a lot more like Porn Stars (willing to work anywhere and probably keeps working at home, doesn’t show up drunk or invisible, etc).

The important part – that we are looking at doing some hiring soon, and looking into our longer term growth. Interns are very important to us, we all used to be co-op students at UWaterloo, we know co-ops rock, and we know the best way to get talent before it gets into the market is to get it as students. Adam wrote a great article about this and I’m stealing some of his ideas here, but the first big problem is no one knows who we are (see avery’s article). We are generic start-up X. And as Adam correctly points out, students, for two reasons, want to work for the Googles and the Facebooks of the world. They get all of the good talent for free, and its not in the students best interests at all. Adam proposed a YCombinator for interns, and I think its a great idea, but way outside the range of what we can do at Upverter right now. So while he is working to make it happen 😉 we have cooked up our own little hacks to try and solve our part of the problem. But first, the problems as we see them:

  • Students don’t get startups. We have been reading Joel and PG since we were teenagers (Steve might have started in the womb). We lived and breathed startup during our co-op terms. We know we would someday be in these shoes. Hell I even dropped out of high-school to start my first business. But other students don’t get it. there are probably tons of students that would thrive in startup culture but have no idea what it means to work for one.
  • Students are irrationally risky. Adam was bang on with this point. You are at the most flexible point in your life, why would you ever chose stability?!?! You could do anything. You could go anywhere, work anywhere. You have no obligations and no ties. Go work for Google when you get bad, or old, or slow, or married.
  • Students want resume juice & prestige. But they don’t know how to get it. They wrongly assume that for their next co-op or real job someone is gonna care that they worked at Facebook. What they don’t realize is that owning a project at a startup and keeping them alive is worth a million times as much to a hirer.
  • Startups are scary & job postings suck. Students don’t know how to pick a startup. They don’t know if a startup is going to disappear and leave them stranded. There are a lot of crappy jobs masquerading as startups, and they are hard to see. And the big boys have realized that fun titles and cool descriptions can hide the cubicle farm reality.

We are going to do 2 things to try and solve these problems. The first is really simple. We are going to convince CECS to allow us to start embedding youtube videos in our job postings. We think this would be magic. The vibrancy, energy and passion that would be conveyed by the three of us talking at a webcam from our ‘office’ cant be faked. You cant cover up a cube farm in a video. You can’t fake interesting. And students immediately get a sense of fit. We think this will do a pretty incredible job of tackling the scary startups & crappy postings problems. The second thing we are going to do is talk to students. And honestly this probably wont scale very well, but we think talking to them about what a startup is, what co-ops do in startups, and what we did in startups is a hell of a first step.

We want to talk about what we look for when hiring and why the name Google will never matter as much as the work you do. We could talk about how none of us ever worked for a buzz name and what that means. About where innovation actually happens and why Google and Facebook buy so many startups. We also want to talk about risk, and give the students a vocabulary for talking to a startup; The questions to ask, and how to know if one is going to disappear or not. The goals of this talk would be pretty simple:

  • Explain what a startup is and why they are the innovators
  • Look realistically at the risk, and the reward
  • Give students a skill-set for evaluating startups
  • Dispel the resume juice myth

And realistically if we could pull off two of these we would be doing pretty well! So whats this look like? We think its us renting a lecture hall and inviting 7 of our most interesting startup friend and colleagues. Fill the hall with a hundred or so of the best and brightest engineering and CS students. Tell stories for about 10 minutes each, no more, and try and get these kids excited about startups. Tell stories that talk about how real and influential the work we do is. Tell stories about the day we killed the internet to Chicago or had a million page views, or changed the world. Tell stories about how the CEO gets the co-ops coffee, not the other way around. Tell a story about hiring and whats really important. A story about what Mom said when PG was on the cover of Forbes. Little, big, passion, excitement, real work, real change, 3 guys in a dorm room. Afterwards we all get pizza and beer together and we can mingle. But the takeaway, and this is where we win or lose, is what the students apply for the next day. Do they still click Google? or do they say “what the hell, why not try a startup”.

At the end of the day its a hack, and we dont really know what will happen, but with any luck we get some great co-op applications. That being said the motivation and the goal is to fix the fundamental problems with the culture around going to work for a startup. I’ll probably be writing more about this as the event gets closer. We think it will be happening May 18th at UWaterloo – but its not set in stone yet. And we are still looking for speakers, sponsors and thoughts, so email me (

Bonus: I found a very cool resume while I was doing some research for the blog post. I’m not sure if CECS would ever allow co-ops to post a resume like this, but it would be monumental in fixing the opposite problem of getting a billion resumes for a co-op posting.

Update: Avery posted a response to some of my points, and I think its perfect. I took a pretty hard line in one direction with this post, and he has done a great job of pointing out where I’m over the edge. I only hope the dialog continues!

February 2011 Update

February 2011 Update

February Sucks!!! All of this sitting inside with no new electronics to hack on, or cool web apps to play with… But wait! Whats this I hear Upverter is now (almost) in Alpha?!?! Haha. Thats right! February is going to become officially awesome (or at least less sucky)!

So I just thought I would post a quick update on whats new around the Upverter ‘offices’ in the month of February. And first off we are loving it down here in the Bay Area, albeit missing home sweet home (except for the cold and snow). Last fall we apparently impressed someone enough that they flew us down for a couple months of intense startup training – and its going great! We’ve met so many cool new friends and colleagues. Its awesome.

A very big part of the program is release early and often, and its keeping us true to our original intentions of releasing something (anything!) and working with the community to build the tools they actually would want to use. Expect us to break things occasionally (sorry) but know that its just us trying to add functionality. But please still tell us when stuff stops working, or what features you need, or that you love/hate what were building. Any and all feedback is welcome and appreciated!

In other news we are starting to look for a permanent home. We are Canadian boys tried and true so its going to be somewhere in Canada for sure, leaving us with just a couple picks. It will probably end up being downtown Toronto but Montreal was also suggested as an option, and who doesn’t like Montreal?!?! haha. Any suggestions are welcome! We also accept couches, cheap rent, and empty basements… just so you know…

Exciting Feedback!

Exciting Feedback!

Today I want to share some terrific feedback we have received recently about the tools we are building. First off I need to say that its incredibly exciting getting emails that we know people sent because they want to help us make this whole eco-system friendlier and easier to enter. So thanks for that! And second, that its just super cool when these emails come from people that we think are incredibly smart, and passionate.And when they write about things that we have thought about or feel strongly about, or even considered solving – well thats just super, super cool.

And thirdly, I want to talk about talking about more. I’ve been thinking and I’ve come to the conclusion that there really isn’t any harm in talking more candidly about some of our long term goals, but more importantly in talking about some of the bigger problems in the OSHW ecosystem – even problems we can’t fix yet. And the plug here is that they are the same thing, we want to position ourselves such that the big problems in the eco-system and our long terms goals are the same things.

So here goes! The following is a (slightly revised) to-do list from one of our recent feedback emails and I wholeheartedly agree with its content. So over the next few weeks I am going to be writing about our thoughts on Manufacturing, Modularity, Over-Optimization and Packaging. You’ll notice these are all very real world physical problems. Problems that we haven’t talked much about yet, and problems we don’t yet have solutions for. But we have ideas, and piles of ambition!

A Hardware Workflow Wishlist

  1. Prototype: I want to buy a collection of pre-made hardware component boards each satisfying a function. Examples are power supplies, processors, FPGAs, and video boards. I don’t really care how many separate little boards it takes, as long as it’s easy to clip them together so I can start building software for it. And I really don’t want to solder anything!
  2. Design: In your tools, let me draw a picture of the exact modules I plugged together in #1  and tweak them into my “final” configuration. I want to export a netlist, bill of materials, and board layout, all automatically.  This should be enough to take the design to a contract manufacturer.
  3. Package: Let me order a standardized case that will be able to contain my new board (either made up of the prototype boards, or as a custom pcb). Automatically figure out external connectors and do the cut-outs for me.
  4. Manufacturing: Give me a “print” button so to speak that will talk to the manufacturers for me, and you take a cut of the revenues.  Trust me, I don’t mind giving up a bit of money in order to not have to deal with the manufacturing headaches.

    Also doing assembly and packaging. You could event put it into a fancy printed box with paper materials that I specify (and maybe a CD).

    QA testing it, and shipping it out to my customers.
  5. Bundling: Handle aggregating pre-orders from random people on the Internet so we can all get a decent price, or find a way to do cheap one-of manufacturing.
  6. Optimize: At this point I’m either done, or I have a marketable product. You could help me transition to hiring people and optimizing my design, or you could scale your company and fill the voids here (like simulation, verification, and design-for-manufacturability).

What a great rundown eh? We have talked publicly a lot about #2 and have some pretty great solutions to it. In fact, I encourage you to head to our landing page and sign-up for the beta of our soon to be released circuit design tool. But over the coming weeks I’m going to give you some thoughts on the others in the list – we have some cool ideas to help fix some of these problems, and we have some great feedback on what other people think is broken. But there is also a discussion that should happen here – and I’d like to start it. As always I’d love any thoughts or feedback, email me: