Carlos

Digital Twins is a marketing term that I absolutely disdain 🤢🤢🤢

Just read a pre-print that uses the term approximate independence

Bruhhhhhhhh

Reading some Smart City papers and no idea so far beats getting Docusign + Smartsheets in Govt.

James @cashbonez If welders were like software engineers they'd be tweeting stuff like "the justice system can be replaced with something that can weld two sheets of metal together" 1:13 AM • 28 May 18 1,932 Retweets 8,411 Likes

help we’re trapped in this post from 2018

There’s is definitely change point detection algorithms in R. Not sure about full segmentation

Lumping generic quantitative methods and tools within AI policies has to be the most cringe thing going on rn

Keep fighting the good fight! AI policies are being written that should not include ML.

There is a Del Taco Parcel Map recorded in Santa Barbara County. Lmao

I swear this myopic De Gustibus Non Est Disputandum attitude has held back economics in so many ways.

Koopsman and Beckmann (1957) showed the limit of prices in their wonderful publication. No curiosity from economists about non-price mechanism just price theory ideology

From the gis community on Reddit Explore this post and more from the gis community

Pretty cool thing to do

www.reddit.com/r/gis/s/BwRU...

His fielding though

Like seriously, nothing will make you an unbeliever faster than learning about land subdivision.

Children believe in Santa.
Some Geographers believe in scale-free networks.
To be fair, children don’t know any better

My Roman Empire is knowing that Macys had the best chance of overturning Prop 13 (CA) but they withdrew their case before the SCOTUS because of pissed off customers

Annual reminder: Spotify Wrapped is proof that sometimes count(x, sort=TRUE) is all the data science you need

Sick roster those Seahawks

There’s another one but this one is pretty memorable

theonion.com/marxists-apa...

Clearest image ever taken of Heart Nebula, located 7500 light years away from us

Took a look at a AI combinatorial optimization paper for graphs. Supposedly took them dozens of minutes to solve a MIP for a min dominant set problem on a 1000-1200 vertex graph using Gurobi. Looked kinda slow…

This is the material I crave

WAGMI

This amount is purely anecdotal as it is also how much compute I’ve needed to solve a vehicle fleet size and routing problem covering all of New York City for a whole day in 15 min intervals

I hope the cheapness of RAM opens up the door for revisiting a lot of urban optimization problems with LP/MIP. Yes, we’ve had access to powerful clusters and servers but I argue that 32GB on a desktop is where you can start exploring really massive, meaningful problems

Would

I hate this trend.

Reading sql definitely giving me cte

Double edge sword. Lowers barrier to entry but also gives you a chance to grill them on a very memorable term.

Not cutesy but asking what’s GIS in an interview remains undefeated