I agree, this is the most sensible path for someone with IT and programming experience. I spent over 10 years following this path, worked for top astronomy institutions and observatories. It was quite fulfilling and allowed me to contribute to many research papers and projects. But, this career is almost always based on soft-money (grants) and it is almost unheard of to get a truly tenured position without going the PhD/academic route. I eventually left astronomy due to fizzling of grants and the desirability of the skills I gained while becoming a data-analytics software engineer (eg. data scientist in today's parlance). I also wanted to start a family and have more control over my skills, location, and career options. Finally, working as a telescope support or astronomy software engineer, expect your salary to be in the $45k-80k (US). The skills you use as a successful astronomy software engineer will easily net you 1.5 to 2 times this salary, in industry.
True, it may seem inefficient in terms of old school stare-at-one-point astronomy, but this is one example of a new field in astronomical research: data mining of large sky surveys.
Projects like the OGLE surveys sample many millions of objects many times: just to produce this sort of variability data. Its not a question of reliably finding the objects during an observation, but more being able to identify them when they do occur, throughout a long sequence of observations.
For me, purely random shuffle is nice if done within a genre, but what if your collection covers many types of music? Randomly jumping around is fun for a bit, but for slightly more consistant playlists I use generators such as gjay (http://gjay.sourceforge.net/).
It incorperates user classification in a multi-dimensional space, as well as using crude bpm and frequency finger-print matching. Its not perfect, but it sure keeps you from jumping between ambient and death metal (or something equally as bad). And its got more depth than some '5 star' ranking system.
I'm curious if anyone else has a prefered smart playlist generator?
Slashdot Mirror
I agree, this is the most sensible path for someone with IT and programming experience. I spent over 10 years following this path, worked for top astronomy institutions and observatories. It was quite fulfilling and allowed me to contribute to many research papers and projects. But, this career is almost always based on soft-money (grants) and it is almost unheard of to get a truly tenured position without going the PhD/academic route. I eventually left astronomy due to fizzling of grants and the desirability of the skills I gained while becoming a data-analytics software engineer (eg. data scientist in today's parlance). I also wanted to start a family and have more control over my skills, location, and career options. Finally, working as a telescope support or astronomy software engineer, expect your salary to be in the $45k-80k (US). The skills you use as a successful astronomy software engineer will easily net you 1.5 to 2 times this salary, in industry.
True, it may seem inefficient in terms of old school stare-at-one-point astronomy, but this is one example of a new field in astronomical research: data mining of large sky surveys.
Projects like the OGLE surveys sample many millions of objects many times: just to produce this sort of variability data. Its not a question of reliably finding the objects during an observation, but more being able to identify them when they do occur, throughout a long sequence of observations.
For me, purely random shuffle is nice if done within a genre, but what if your collection covers many types of music? Randomly jumping around is fun for a bit, but for slightly more consistant playlists I use generators such as gjay (http://gjay.sourceforge.net/). It incorperates user classification in a multi-dimensional space, as well as using crude bpm and frequency finger-print matching. Its not perfect, but it sure keeps you from jumping between ambient and death metal (or something equally as bad). And its got more depth than some '5 star' ranking system. I'm curious if anyone else has a prefered smart playlist generator?