This has been experience as well. If there is a hard problem which needs to be addressed, generative code helps me break the inertia by generating the first draft and then I get really curious to poke holes in the generated code. I tend to procrastinate when I come across a gnarly issue or something I am not really familiar with, justifying by saying I need a big block of time to work on it. I use generative code as a pushy "mom/boss/coworker/spouse" to get stuff done.
If you have something interesting to say then people will figure a way to understand it. Early in my career I remember coming across “Essential .Net” by Don Box and I don’t think Don Box had a particular audience in mind. He just unwrapped what’s under the hood in Common Language Runtime (CLR). It took me multiple times to really understand “essential .net”.
I worked on an autocontouring model but we could not get very high accuracy for it to be adopted commercially. The algorithm would work for some organs but would totally freak-out on the others. And if the anatomy was out of norm then it would not work at all. This was 5 years ago, I see Siemens [0] has a similar tool. I remember shadowing a dosimetrist contouring all the Organs-At-Risk (OAR) and it took about 3-4 hours to contour one CT image of thoracic region. Do you know how much better the autocontouring tools have become?
Great point, to me it feels more like running barefoot versus with padded shoes. Barefoot running will build your foot muscle based on your natural form over time, whereas padded shoes may interfere with your natural form and at worst encourage bad form which may lead to injuries as your mileage increases.
Over the years in building software, I tend to spend more time thinking about how all the pieces (i.e. from package manager) fit together rather than building them from scratch and fitting two pieces together require deeper understanding of both the pieces and how the combination of the two behaves in the environment.
This is how I assume one of the ways an expert system would play out:
1. A system which understands not just the language but the meaning behind it and the cultural context around it.
2. Let's assume it's multi-model and have all the senses having the same level of technical maturity.
3. It has read all the text available for all the major sciences and understood the meaning behind it including the possible applications of it.
To that end, having the above conditions satisfied I am assuming there should be major breakthroughs in all sciences. So far what I have seen is that the AI system is able to selectively brute-force it's way in finding a solution. Maybe I am just being impatient.
Firstly, congratulations! This is very inspiring and I love the pragmatic approach to starting a startup. You wrote about converting to a contractor with Help Scout from being a full-time employee. How did it help you free-up your time?
Initially my intent was to go from full time employee to a ~16 hours/week contractor so I could spend the rest of my time on Preceden, but I wound up also picking up contracting work with my former employer (Automattic) and found myself working 30+ hours/week contracting. So much for focusing on Preceden.
Eventually I stopped contracting at Automattic and gradually reduced my hours at Help Scout until it made sense for me to go full time on Preceden. I'm fortunate both Help Scout and Automattic were open to me contracting as it made this transition much easier than it would have been otherwise.
