<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Posts on D.S. Lachut</title><link>https://dslachut.github.io/posts/</link><description>Recent content in Posts on D.S. Lachut</description><generator>Hugo -- gohugo.io</generator><language>en-us</language><lastBuildDate>Wed, 17 Jan 2024 10:36:52 -0600</lastBuildDate><atom:link href="https://dslachut.github.io/posts/index.xml" rel="self" type="application/rss+xml"/><item><title>PNG Tiles from a GeoTIFF</title><link>https://dslachut.github.io/posts/png-tiles-from-a-geotiff/</link><pubDate>Wed, 17 Jan 2024 10:36:52 -0600</pubDate><guid>https://dslachut.github.io/posts/png-tiles-from-a-geotiff/</guid><description>This is a quick look at how to build a PNG tileset from a raster data set.
Get some data First, we need a data set. You can find the USDA Cropland Data Layer here. Or go here to directly pull the 2022 data:
https://www.nass.usda.gov/Research_and_Science/Cropland/Release/datasets/2022_30m_cdls.zip Download and unzip that file to get 2022_30m_cdls.tif.
This is a GeoTIFF file covering the continental United States in an equal area projection (EPSG 5070). Each pixel represents an square, 30m on a side.</description></item><item><title>PostgreSQL EXPLAIN Explained</title><link>https://dslachut.github.io/posts/postgresql-explain-explained/</link><pubDate>Tue, 12 Dec 2023 17:20:00 -0600</pubDate><guid>https://dslachut.github.io/posts/postgresql-explain-explained/</guid><description>A basic overview of using EXPLAIN in PostgreSQL.
Note: This post is based on this post. I&amp;rsquo;m producing this summary for myself, because The Internet is not Forever.
Up Front: EXPLAIN and friends EXPLAIN shows you what the database thinks about your query. ANALYZE executes the query and shows the actual run time alongside the plan. Be careful doing this with anything that modifies data! BUFFERS shows cache usage Have to use with ANALYZE VERBOSE helps with slow, expensive functions SETTINGS shows performance related values WAL show the Write Ahead Log usage for statements that modify data FORMAT default TEXT Also: XML, YAML, JSON Example:</description></item><item><title>The Internet is not Forever</title><link>https://dslachut.github.io/posts/the-internet-is-not-forever/</link><pubDate>Mon, 04 Dec 2023 22:02:52 -0600</pubDate><guid>https://dslachut.github.io/posts/the-internet-is-not-forever/</guid><description>It used to be (and may still be?) the received wisdom that &amp;ldquo;the Internet is forever&amp;rdquo;, meaning that once something is on the Internet, it will exist forever. If this was ever true, it no longer is. Off the top of my head, there are five ways that things disappear.
Link rot. Maybe you saw something you wanted to remember and you even bookmarked it. But, the site owner restructured the site, or moved to a new CMS.</description></item><item><title>The Heilmeier Catechism</title><link>https://dslachut.github.io/posts/the-heilmeier-catechism/</link><pubDate>Mon, 27 Nov 2023 23:00:00 -0600</pubDate><guid>https://dslachut.github.io/posts/the-heilmeier-catechism/</guid><description>The Heilmeier Catechism is a tool you should use any time you start a data project.
The Problem Scoping a data project can be hard. There are often so many &amp;ldquo;unknown unknowns&amp;rdquo; that you may not know if your project goal is even possible. This uncertainty easily leads to yak shaving and other forms of scope creep.
Every project imposes opportunity costs. There is a limited number of data scientists. GPU compute is expensive.</description></item><item><title>The Re-Launch</title><link>https://dslachut.github.io/posts/the-re-launch/</link><pubDate>Sun, 26 Nov 2023 12:00:00 -0600</pubDate><guid>https://dslachut.github.io/posts/the-re-launch/</guid><description>Like many (most?) people, I used to have a blog. It was a repository for works-in-progress, copies of presentations, technical musings, and how-to guides. Some time in the course of writing a dissertation, starting a professional career, having a few children, moving halfway across the USA, etc&amp;hellip; It now comes up as a HTTP 502 error behind an expired certificate.
The reason In October 2023, my former employer had to do a restructuring and reduction-in-force.</description></item></channel></rss>