mechsy
Funnily enough I’ve found Python to be excellent for modelling my problem domain with Pydantic (so far basically unparalleled, open for suggestions in Go/Rust), while the language also gets out of my way when I get creative with list expressions and the like. So overall, still it is extremely productive for the work I’m doing, I just need to spin up more containers in prod.
If your test suite is up to the task you’ll find defects in new updates every now and then, but for me this has even led to some open source contributions, engaging with our dependencies’ maintainers and so on. So I think overall it promotes good practices even though it can be a bit annoying at times.
You’d also have to ‘fix’ DNA: unless we can re-engineer a bunch of key enzymes and then re-encode the entire genome (or maybe key parts) with forward error correction without breaking everything else, it might work. You might also break evolution to some degree by making random point mutations less likely.
But what I learned so far is that as soon as you’d attempt something like this in bacteria, the fitness advantage from an evolutionary standpoint is negligible compared to the efficiency loss introduced by FEC, so your colony would get outcompeted by other bacteria unless there is a niche your resistant bacteria survive in (high radiation environments?). The efficiency loss induced ‘disadvantages’ would probably be less pronounced in mammals though - If (big if) you manage to not also break anything essential in the wonderful yet surprisingly efficient Rube Goldberg machine that is life.
On a related note: I think it's kind of puzzling that all this data is freely available but seems to be rather complicated to access. One would assume building/operating satellites is more complex than making that data available in formats the average startup web developer can use almost instantly. IMHO you can see a similar pattern in weather data: it seems like there are a bunch of platforms which are basically wrapping public weather service's domain specific formats in a 'nicer' REST API. One could argue there's documentation on these formats, just read the docs, duh. However, I think lowering the barrier to entry by providing easy access and documentation could enable way more people to work with this data.
IMHO if you know some basic python and know your way around the requests library, it should be trivial for you to get started working with these datasets. That's just my personal opinion based on my experiences from a few years ago. Maybe it's way easier by now and there's a bunch of great free open APIs for this stuff. It always feels like they fulfill all the basic requirements to open the dataset and write some documentation but the last (small) step of writing the basic 'getting started' docs/blog posts and maybe some REST API for access is missing. Instead, confronted with lower than expected usage, a bunch of industry-transfer projects are spun-up to increase adoption. I guess DX also matters for open data hosted by public entities. /rant
[2] https://dataspace.copernicus.eu/
[3] https://browser.dataspace.copernicus.eu
That sounds interesting! How come? Been doing devops stuff for a few years by now but also have some background in the Unity/HCI space which I enjoyed a lot. I'd be curious how this combination might be applicable in the VFX/media production industry.
I've also been on the fence of learning Unreal for a while now. Maybe you could shoot me an email (address is in my bio)? I'd be very interested in how the industry works outside the 'classic' VFX artist space.