The carpenter comparison isn’t a good one. The output of the carpenter’s labor and (importantly) material resources yields a hard good and that process does not

The carpenter comparison isn’t a good one. The output of the carpenter’s labor and (importantly) material resources yields a hard good and that process does not scale. Software on the other hand scales extremely well and the marginal cost of giving software away for free or selling it to the next person who wants it has next to no additional cost. The expense model of software enables economic models that would never work on physical goods. Unlike hand crafted furniture, software has next to no durable value beyond the transactions it enables. Don’t get me wrong, software investment can be incredibly durable, e.g. banking and insurance mainframe apps built 50 years ago. But without a specific business application software itself is nearly worthless. Writing open source software is probably better compared to published research. It’s an academic endeavor and also a way to put your name out in the world in a way that recognizes your contribution and may lead to commercial opportunities for you and/or your software.