retro image of a Michelson mechanical Fourier analysis and synthesis machine. it's very complicated and large and mechanical.
Me: Can we get some Fourier analysis?
Michelson: We have Fourier analysis at home.
Fourier analysis at home:
Should have tagged: @ryanhatesthis.bsky.social
What is a 'memecoin' anyway? Ryan Broderick in the Garbage Day newsletter describes it succinctly:
"Crypto firms want to monetize the very concept of a meme. Which is as insidious as it is extremely lame."
(Garbage Day newsletter and content here
www.garbageday.email/subscribe?re...)
It needs to match the ✨vibes✨! This is totally legit.
Loath to admit it, but 95% of my hair product choices are driven by whether it smells good.
Cropped screenshot from a Google search that shows the ubiquitous AI sparkly diamond and displays the text "An AI Overview is not available for this search"
This feels like beating a final boss in an AI game.
Like the screen should pop up with job postings for becoming an AI engineer at Google if you can type a search that stumps their AI process.
Standard Likert vibe scale.
Vibes were:
⬜ Sus
⬜ Off
⬜ Vibing
⬜ Immaculate
✅ Unmatched
Something is rotten in the state of Ohio
#databs
Blueskibidi
Don’t think of it as whether the data has error or not. Most of the time it will.
Think of it as whether the error in the data will cause you to make a different decision than if the data was perfectly clean. Most of the time it won’t.
One of the biggest tips I have for anyone doing data analysis, especially data from people, is to spend some time drilling down to the most granular data and just looking at individual records. You will find the craziest shit you never imagined and your analysis will be better for it #databs
I have the Kreg accu-cut track and use that w my DeWalt circ saw - I don't have to rip full sheets often but it works really well when I do!
I use a diablo blade and it is great at limiting tear out.
I, too, have uttered the words "it's the freakin' weekend baby I'm about to have me some fun" often in reference to sitting down to work on a side project, which eases the guilt a bit.
At work, I always start with the problem to solve.
In hobby projects, I agree that it's fun to just explore and see where the data takes you!
When you are working on a data side project, do you usually start with the problem to solve and find the data to answer it?
Or do you start with the dataset and go down the rabbit hole to see what you find?
What are the current alt methods we are iterating through as a species that will be the telegraph/Morse code combos intrinsically linked in future humanity's experience?
(Highly recommend The Information, btw!)
Telegraphs and Morse code have always been intrinsically linked in my mind. Reading "The Information" by James Gleick @around.com and learning there were at least half a dozen early iterations for conveying language through an electrical wire is fascinating.
Also begs the #dataBS question -
Telegraph tech and Morse code have always been intrinsically linked in my mind. Reading "The Information" by Gleick and learning there were at least half a dozen alt iterations for conveying language through the medium of electricity that were tested and then abandoned is fascinating.
Yep, agree with this - even if prediction of which customers will churn isn't all that's needed, the analytical process to dev a model - EDA, feature engineering, feature importance - can still lead to actionable info.
And then model output provides the list of customers to pull those levers on.
YES. Love this question and totally relate to this answer!
Tangentially related to the original post - but say the obvious thing! Connect the dots!
One of the biggest differentiators I've seen in good senior/leadership folks in tech is that yes, they have next-level skills but also, they SHARE THEM IN A CLEAR AND OBVIOUS WAY.
Data Weaseling and Data Hoarding sound like top tier #databs skills
My inbox filling up with meeting cancellations due to holiday schedules is the BEST early gift.
Welp.
How can we visualize what a book ISN'T talking about? With an anti-tag cloud! See the most common English words that are never mentioned in a text.
www.bewitched.com/demo/anti/
Brb, changing every bullet point on my resume to say "tedious, undifferentiated task"
I have no knowledge of this case, but have seen lots of pathways to the outcome.
Misattributing events as users. Poor/no de-duping. Incentives that drive users to sign up for multiple accounts.
Our biases lead us to accept a number we like and scrutinize those we don't.
Happy working on side projects day for all who celebrate! 🥳🤩
A corollary: if there is coding shown in a movie/TV show, I am def pausing it to see the code.