he, him. interested in 🛐mystics, λ👩‍💻functions, ⚛️physics, 🌬️📖process philosophy. occasional tweets about ✝ present and ☸️ past. trying to be a good 🐻.

Kansas City, MO
Joined February 2015
Hey @HomeDepot you f#$king THIEVES give me back my money
1
77
Actually, it's only considered Christianity if it's from the Jewish region of mysticism: otherwise it's just sparkling white messianism.
1
What's the line about outlaws and guns? “If you only allow people to merge perfect code, then only perfect people will merge code”?
I will no longer engage in philosophical discussions about conscious AI/superintelligent machines, and here's why. (long🧵1/11)
137
1,306
357
7,671
To bridge disagreement, create a caring connection and promote understanding.
1
3
5
Programming languages researcher who never programs
Testing researcher who never writes tests in their own code
6
2
8
57
this is how I Feel about my code.
Artists are burning alive to deliver a message, yet some are wondering if the message looks good in their living room.
2
8
Rule one of Kubernetes: if you think you need Kubernetes, you don’t. When you know that you have scaling problems and you don't know the solution and it all seems hopeless, then *maybe* you need Kubernetes. Depending on the sorts of scaling problem you're dealing with.
Don’t get bogged down in theory. Learn to use advanced programming concepts to solve tough, real-world problems with Haskell today. Effective Haskell by @cercerilla is now in beta. Pick it up at pragprog.com/titles/rshaskel…. Or learn more at media.pragprog.com/newslette… #pragprog
5
46
11
237
Can we please standardize the words we use for the length of a function? Just so “we prefer small functions here” vs “we have a lot of large functions, sometimes that is the best way” has objective rather than subjective status... Let me suggest a decimal progression, √10≈3...
1
So a function 3 lines of code or fewer is “fun-sized.” 4-10: “extra small.” 11-30 LoC is “small”, fits on a printed page or most screens all at once. 32-100, “medium,” 3 pages, 15 mins to read ish? 101-300, “large”, you might need an hour to read this 301-1000, “extra-large”
1
After XL I recommend we just use printed physical media for comparison. 1k - 3k LoC ≈ 30-100 pages = “novella” 3k - 10k = book 30k, textbook 100k, a bible 300k, an encyclopedia.
Cleaning out my wife's closet like BRAS FOR THE BRA GOD, SOCKS FOR THE SOCK THRONE
#Mashup ideas Then there you go, And Michael you would fall, And turn the white snow red as strawberries in the Summertime love with my summertime lover Give me a chance I'll treat you like a princess...
In an alternate universe, dialing 822 gets you rapid response legal services. Instead of paramedics, paralegals come, instructed in stabilizing your legal situation while taking you in a car to the lawspital where an actual lawyer can return on your case
it hath been Expanded
14
851
15
3,283
Chris Drost retweeted
I’m tired of winter. Enough already. It’s time to stop plowing the roads. Forecast calls for 2 feet, but that’s not a meaningful metric since it will be light and fluffy. People who are anxious about it can take personal responsibility and shovel the road in front of their house.
169
963
164
5,562
This is a short story about the lengths the political establishment will go to to evade accountability and oversight. I ran a write-in campaign (as a Libertarian) for Auditor of Silver Spring Township, Cumberland County, Pennsylvania this past November. [Part 1 of 19]
64
410
250
1,309
Target made a big splash as “Walmart with slightly higher prices for slightly higher quality.” Cool And I get why the cutting edge tech company still sells books, that's where they started But “we sell you the things so crappy, even Walmart wouldn't dare!” that's... surprising
Snake graphs. You can hook up a Google Form to a Sheet to collect raw data, sum it together into daily buckets, and do a nice moving average. The “mouth” at the end has to do with the moving average including points that are not known yet.
1
The trick there is that I do a linear regression of the series against its values for the last 3 timesteps—an autorrgression—and then calculate the standard deviation of the series points away from what this model predicts they should be, to give an error term.
1
This lets me predict a range of future values plus or minus one standard deviation, and I just plot both the +σ and -σ moving average lines with the same style.