Everything in software engineering has changed within a decade — I was a hands-on technologist 20 years ago, turned to management ten years ago, and am now hands-on again as a consultant. I’m amazed by some changes and surprised by others. Here is a basic rundown of what almost caused my early demise. Unix Is Back for the Developer (Silver Lining) In the ‘80s and early ‘90s, a…

Step thru the process of building & training a custom NLP sentiment analyzer. — Do Vulcans express sentiment without emotion? With the power of Machine Learning, we can find out. We shall evaluate famous Spock Quotes using this (yet another) basic guide to Natural Language Processing (NLP) and Sentiment Analysis (SA) methods. “Once you have eliminated the impossible, whatever remains, however improbable, must be…

How using a blockchain as a general purpose database is the slowest, least scalable, least private, and most costly solution — Ultra Brief History (Web 1–2–3 and 5) Just in case it wasn’t clear: Web 1.0 most people think of as HTML and data retrieval in the Netscape and HTML era (up until the late 90s) Web 2.0 is evolving with JS/DHTML and modern dynamic frameworks like React.js. Web 3.0 is building web apps using blockchain as the…

When I was young I always wanted to rewrite, then I realized it is not possible, now I wonder again… — A familiar cry for help: “the codebase is a mess. It takes more time to understand the impact of making a change than the change itself. Performance is getting slow due to the number of indirections and interfaces. We have a new major change to make, and after staring at the code for hours.. or…

Elastic Python-compute for Machine Learning is coming to Snowflake — it may be a deadly snowstorm for Spark… — I often help design Data Platforms for large IT shops (my other secret job that pays me better). What I encounter a lot is Redshift and Snowflake in the warehouse (more of the latter these days) supplemented by Databricks for Machine Learning (ML) and Spark support. Databricks has done 2 seemingly simple things well

We all know it is slower than C, Java, Rust, but just how slow? — Just as I started thinking about this article, my Medium feed produces a joker post “Python isn’t actually slow” — I read it even though I knew it was not true and was just clickbait. …

My search failed, just like my search for the ultimate pancake — “The modern data stack encompasses a data pipeline, a destination, a transformation layer and a business intelligence/data visualization platform” — Fivetran I read about every article on the Modern Data Stack (MDS) and listened to a few podcasts, youtube videos, and came to the conclusion Everyone has their own definition…