4 ms·

Please allow me go offtopic and ramble for a bit. This article feels like a punch in the gut. So before my tech career, I did a 1.5yr Fellowship at a renowned

Please allow me go offtopic and ramble for a bit. This article feels like a punch in the gut. So before my tech career, I did a 1.5yr Fellowship at a renowned research lab -- The Rockefeller University. Our entire programme of research was around the possible causes of multiple sclerosis. One of the reasons I decided to pursue Computer Science rather than clinical research was the slow discovery cycle in clinical research. I got so much faster (albiet smaller) victories in Computer Science. That was 1996. This is 20 years later and sadly we're still trying to determine the etiology of the disease. My research advisor passed away, the other is deep in retirement. We're still trying to figure out the mess of genetic and environmental factors which cause MS (our theory was Human Herpes Virus 6.) I owe so much respect to individuals who pursue clinical research for a generation despite how slow progress can be. I was pulling my hair out in 1996 when I worked on this, because it was so slow with so little signal to success/failure. ...In the meantime, I get antsy if my build takes longer than two minutes or if a scrum sprint goes over by a day. PS this is also why I admire Craig Venter (https://en.wikipedia.org/wiki/Craig_Venter https://en.wikipedia.org/wiki/Craig_Venter) so much -- he worked so hard to institute the practices of automation into biomedical research. It is also why im so excited by Mousera/Vium

swuecho 9y ago
small world. I just quit my job at UCSF and I was a programmer in the MS group.
Balgair 9y ago
Its only gotten worse in the last 2 decades. There are a myriad of articles that pop up on HN about how bad bio had gotten. No one can replicate anything anyone else is doing and it seems like ~45% of published research these days is p-hacking to graduate students. It used to be 'publish or perish' but now its 'funding or famine'. Why not fudge the results, who is ever going to check, let alone prove p-hacking? The researchers have mortgages, college-funds, and college debt too. Nature came out and said that a P of 0.005 (not .05) is the way to fix it. Like, guys, the culture is rotten from the core on up, you need to fix that, not add a 0. It makes sense, the funding is (long-term) drying up relative to the number of researchers, there is less money to go around, so we get to see who is swimming naked.

swuecho 9y ago

small world. I just quit my job at UCSF and I was a programmer in the MS group.

Balgair 9y ago

Its only gotten worse in the last 2 decades. There are a myriad of articles that pop up on HN about how bad bio had gotten. No one can replicate anything anyone else is doing and it seems like ~45% of published research these days is p-hacking to graduate students. It used to be 'publish or perish' but now its 'funding or famine'. Why not fudge the results, who is ever going to check, let alone prove p-hacking? The researchers have mortgages, college-funds, and college debt too. Nature came out and said that a P of 0.005 (not .05) is the way to fix it. Like, guys, the culture is rotten from the core on up, you need to fix that, not add a 0. It makes sense, the funding is (long-term) drying up relative to the number of researchers, there is less money to go around, so we get to see who is swimming naked.