Q: How to balance low cost and high quality (India)
Q: QA in open source solutions? (UK)
I'm going to address both of these questions together. In regards to the the first one, I think we need to clarify the definitions a bit since "low-cost" and "high quality" are very ambiguous terms. They'll mean different things to different people. I've addressed the term "low-cost" in a previous post, where I think there should be less focus on cost and more on value. That means providing a clear benefit/s to the target customer. Cost is a dangerous game to play in where you don't really want a price war unless you have a clear strategy going into it.
In regards to "high quality", this can also mean many different things. To some, high quality means rugged, high reliability, or a long usable life. To others, it means good customer support, documentation, and training materials. In terms of reliability, no matter the cost, you need to have a test setup, more likely some automated test setup. This would be a minimum in terms of quality and QA. You can get customized test jigs made using bed-of-nails and toggle clamps or you can make them yourself. It's then up to you to create the test software to automate the testing as much as possible. I would consider this a standard QA procedure, regardless of whether it's open source or not.
In terms of quality, perhaps it could mean ruggedized which is highly dependent on the application. If it's going to be on an animal or in an environment where it will experience a lot of vibrational shock, then you'll need a vibration table or device to test it against to see where the statistical points of failure are. I've even used recycled massage vibration motors from motel beds to test parameter like this.
If it's going to potentially be deployed in the Antarctic or Sahara Desert, then you need to define a temperature range which covers these extremes and test for that. If you're selling your product at a cost that won't allow you to do this testing, or your customers aren't willing to pay for this kind of testing, what you can do is sell graded versions of your device. The standard version might have an operating range of 0 to 50 deg C which covers most applications. Then there can be ruggedized, graded, or premium versions which are tested from something like -30 to +85 deg C for harsher climates. These would usually require special environmental testing chambers which need to be rented or purchased.
It's also possible that you might need to use special parts for these such ICs which are graded for industrial use. Many ICs have a commercial version and industrial version. These particular devices should of course cost more since they require much more thought, testing time and effort, and also more expensive components.
No matter what, it's important to have documentation, support, and training materials for your products. Without these, it's unlikely you'll get much adoption. If you're selling things at a cost that doesn't value these things, then you probably have a pricing issue or you're focusing on the wrong types of customer. You don't want customers that only think about price above everything else.
So perhaps I would say that rather than looking at high quality vs low cost, I think it's more important to focus on high value, which means features that bring benefits to your customers. This only comes from working closely with the people using your device, listening to them, and incorporating their feedback.