Ten years ago I was working with malformed data solely in spreadsheets. After ten years of hard work learning engineering and statistics, I finally am blessed to be working with malformed data in queues, matrices, containers, and serialized formats.
— Vicki Boykis (@vboykis) September 14, 2018
Coding on a computer that doesn't have your dotfiles pic.twitter.com/nkoAsG4Nly
— Jake VanderPlas (@jakevdp) September 13, 2018
— Bojan Tunguz (@tunguz) September 11, 2018
Best one I have seen. pic.twitter.com/m4aTkhqQ5l
— Mahdi Yusuf (@myusuf3) September 10, 2018
My function name development process https://t.co/6iGEnD0wpB
— Hadley Wickham (@hadleywickham) September 7, 2018
— Brooke Watson (@brookLYNevery1) September 5, 2018
Who called it “computer security” and not “defense against the dork arts”?
— Daniel Lowd (@dlowd) August 31, 2018
Tired: big data
— Chris Albon (@chrisalbon) August 31, 2018
Wired: massive data
Reality: data_cust_v28-march-34_ 2016_midlo-incomplete_final.csv https://t.co/asesWsAC5B
2008: Hack the planet!
— Chris Albon (@chrisalbon) August 29, 2018
2018: Train the planet! pic.twitter.com/NQIRvsllW9
Next time someone asks for a definition of what #DataScience looks like in practice, just show them the Apollo 13 CO2 scrubber scene, right down to the reproducibility & communication aspect, only the astronauts die bc the parts u have aren’t representative of what’s on the ship
— Nihilist Data Scientist (@nihilist_ds) August 28, 2018
This really is the best. https://t.co/fTpbn3zYdE
— Chris Albon (@chrisalbon) August 28, 2018
Always predict failure.
— Nihilist Data Scientist (@nihilist_ds) August 24, 2018
As an estimator, this has several nice properties: it’s variance is zero and while it is biased, this bias converges rapidly to zero.#statistics #DataScience
