Time, People and Money: Exactly Why ‘Hardware Is Hard’ | Hacker News Reader

Time, People and Money: Exactly Why ‘Hardware Is Hard’

medium.com

26 pointsmojuba7 comments

I helped fund this project on IndieGoGo, originally set for delivery of June 2015. I was super confused when they went to Kickstarter after already raising one crowdfunding round. Then I realized - they just used IndieGoGo to fund their prototyping phase. I have yet to receive my Flow/Nuimo/whatever it is they call it, or even an expected ship date.

It seems a little duplicitous to me that they presented this concept as largely complete at the start of the IndieGoGo campaign and then turned around and started crowdfunding again. I know, they're saying "we're making the product better! It will ship so much cooler!" So what? You said it was plenty fucking cool up front. I thought it was plenty fucking cool up front! But I'm really frustrated as a customer - I felt a little bait-and-switched.

On top of that, it kind of bugs me that Senic seems to be focusing a bunch of their energy on writing a ton of self-congratulatory blog posts instead of getting product out the door. Sure, they went through this really hard, grueling design and manufacturing process to get to where they are. So what? They haven't moved the dial on their total users at all in doing so.

/rant. Mostly frustrated that I haven't gotten my Flow/Nuimo/whatever it's called yet.

Well the impression i have is that Indiegogo only got traction because Kickstarter stopped accepting projects that didn't have working prototypes to demo.

I was always wondering what the difference between the two was. Now I feel like an idiot for not knowing this up front. Oh well, $100 I may or may not ever see again.

I worked for a hardware startup for 2 years. In my experience, part of the reason "Hardware is hard" is because it's not something that a lot of people in the tech space have spent their lives tinkering with in the way that they have with computers. Version 0 of software just runs on your computer than you already have. Version 0 of hardware (beyond 3d printed prototypes) isn't fundamentally impossible, but most of the startup people I know aren't the sort who know how to cast aluminum in their backyard, for example.

We shot ourselves in the foot by paying for tooling for castings before we'd validated it beyond 3d prints. It ended up having minor flaws that we couldn't fix because it required adding material to the tooling, which isn't something you can do easily. If we'd had somebody on the team who was the sort of guy blowing shit up, setting stuff on fire, and making aluminum ingots in a homemade forge when they were 16, I'm confident we wouldn't have screwed up in such a basic way.

>>>not something that a lot of people in the tech space have spent their lives tinkering with in the way that they have with computers.

Its one of those 'truths' that are so true that people just don't want to deal with it: you can upgrade your software within seconds, but it takes - at least - a few minutes (if you're lucky) to upgrade your hardware.

I've also worked for hardware companies, over the years. Hardware is hard because you're pushing atoms around. Not electrons. The difference between the particles is reflected in the difference between the kinds of engineers: you can express, in words, mere electrons - in seconds. But it takes a lot longer to make that same expression in hardware.

>>>We shot ourselves in the foot by paying for tooling for castings before we'd validated it beyond 3d prints.

We need better tools, and this is pretty clear to the hardware engineers who are trying to bring the design and production of hardware products to the same level of software.

Imagine if, one day, we can build new hardware near-as-rapidly-as new software. In many ways (e.g. OpenSCAD: http://openscad.org) we could compile hardware components as simply as we can software.

When the gap is closed, we will have a new class of society: simply, wizards.

Hardware is hard because it's slow and expensive to iterate. Furthermore, it's hard/expensive/impossible to update products that are in the field.

Going to crowdfunding before contacting factories and thinking about design for manufacturability seems dangerous to me if you haven't done hardware products before. Real products are rather different than arduino / raspberry pi prototypes.

Reply on news.ycombinator.com