> Last December, Keller Scholl and I posted a working paper suggesting that this whole narrative is bullshit, at least so far.
Thing is, when you are calculating the color of the sky, then it is either a bluish looking spectra, or you are in trouble. And consequently physicists will look outside immediately after that calculation.
The entire post needs an explanation of why they don't find any automation in that timeframe? I mean, just from administration we have things like package managers, containers, several cloud stuff, and so on. If they don't see that in their data, then my immediate assumption is, that they need a more reasonable approach.
> I mean, just from administration we have things like package managers, containers, several cloud stuff, and so on.
Automation replaces human work, business can do better now, then they want to produce more than before so they up hiring as well. It's much easier to set your target higher (production, profit) than to automate anything. As soon as you have automated something new work pops up, work that was probably not done before for lack of resources.
Programming has been automating tasks for decades yet here we are, so many of us. Cloud should have automated many jobs, yet with this new found ease of deployment we chose to deploy more rather than hire less. As long as humans have needs and desire for betterment I don't think we can surpass our ever growing ambition with automation.
The article reads like sour grapes about an article that didn't get as much attention as the author believed it should, and yeah, absence of evidence is not evidence of absence.
Thing is, when you are calculating the color of the sky, then it is either a bluish looking spectra, or you are in trouble. And consequently physicists will look outside immediately after that calculation.
The entire post needs an explanation of why they don't find any automation in that timeframe? I mean, just from administration we have things like package managers, containers, several cloud stuff, and so on. If they don't see that in their data, then my immediate assumption is, that they need a more reasonable approach.