DISQUS

Bora · 1 month ago

This concept started taking off in the early 2000's with the push of Web Services and today it's at a point where any application you're building needs to take into consideration, what's already out there. What can I reuse that is already built rather than rewriting it myself? Inside or outside of your organization...

There has been a wonderful move towards API based software development and it is really fostering innovation and making it easier to build web applications, enabling people to focus on delivering value rather than spending time on solving technology problems.

Now an interesting thought would be loose coupling in building hardware. It is now fairly easy to get software companies started with low upfront cost but if you have an idea that needs product manufacturing, there is little reuse of existing parts design hence resulting in high upfront costs before you even get a prototype out there. Imagine an "Open Source Hardware Design" concept as detailed in Hartmut Esslinger's book "A Fine Line: How Design Strategies Are Shaping the Future of Business". If we want to foster innovation in hardware space as well, this is the way to go about it.

albert · 1 month ago

There has always been some of that in hardware. Your video driver can break and yet your computer will still be accessible via say a remote terminal. As far as open source hardware goes, you should check out http://www.buglabs.net

Bora · 1 month ago

In computer hardware development, sure, chips, drivers etc but when you think about other types of products, it's more proprietary and tightly coupled. So let's say I invent some technology that would allow cellphones to not have batteries and charge with sun or wind, whatever. But I have no clue how to make a cellphone, I just know how to make that energy component. But I want to have my own cellphone product that looks cool. I want a touch interface just like iPhone's but thinner maybe. If there were some open source designs for these components, I can just use those designs and get them manufactured easily. So if an analogy is to be used, imagine a hardware equivalent of a "ruby gem" - a touch interface - which I can just plug into my own hardware product. From a VC standpoint this makes sense too because it would reduce upfront cost and time to market drastically, lowering risk. The Buglabs stuff looks pretty cool by the way.

Julien · 1 month ago

That is exactly how we built Superfeedr. Our main "motto" for any decision we take is : would that work if this specific piece goes down? And I think this is the secret to "scalability", as we don't rely on anything particular, but on the whole.

In theory, that is now working exactly like expected. (of course, if many components go down, things will start to break apart).

kidmercury · 1 month ago

one of the reasons i think open source software development communities are so promising and are the future of software development -- they naturally evolve and coordinate in a way that embodies the notion of loose coupling. accordingly i think loose coupling is one of those concepts that becomes more important as strategies of "openness" are increasingly pursued.