3 ms·
Thanks for reading! Yeah, I completely agree with you. One of my biggest influences is Joel Spolsky, and his advice on this is, "If it’s a core business functi
by mtlynch 4y ago
Thanks for reading!
Yeah, I completely agree with you. One of my biggest influences is Joel Spolsky, and his advice on this is, "If it’s a core business function — do it yourself, no matter what." [0]
When I started the business, I thought the money was in the software, not the hardware. I figured people would build the hardware on their own, and they'd pay me to create a nice software experience for them. I offered hardware kits as a convenience for users who wanted to buy all the parts at once. Then, I realized people would pay more for custom parts, and I ended up with electrical engineering at the core of my business before I even realized what I'd done.
In the future, I'll definitely stick to software and domains where I'm competent at the core skills myself. But now that I'm here, I'm hoping to make it work despite having to outsource the EE portion.
[0] https://www.joelonsoftware.com/2001/10/14/in-defense-of-not-invented-here-syndrome/ https://www.joelonsoftware.com/2001/10/14/in-defense-of-not-...
- robomartin 4y ago> When I started the business, I thought the money was in the software, not the hardware. That can happen. I had a business where I set out to develop hardware product A, and then realized I needed product B to be able to make A work. Five years later the business was about product B and A never got off the ground at all. Reminds me of when I bought my first surface mount assembly line from this guy who deals in used manufacturing equipment. We had a nice conversation as the equipment was delivered and installed. He had downsized the business from some 50 employees to just him and a couple of assistants. He could not be happier. At one point he said: You know, a business is like a living, breathing life form that tells you what to do every day, not the other way around. BTW, I still think that businesses that can create a hardware + software lock can be excellent. Generic hardware is trouble, just ask anyone who thought they'd design a cool new 3D printer only to have it cloned by someone in China before their Kickstarter campaign was over (this actually happened at least once, I met the poor fellow). If the hardware is connected to useful non-trivial software and services, the barrier to entry is significantly higher.
- mckravchyk 4y ago> Yeah, I completely agree with you. One of my biggest influences is Joel Spolsky, and his advice on this is, "If it’s a core business function — do it yourself, no matter what." [0] Keeping that in mind in the era of AI-generated code.