Today is my 365th consecutive daily blog post. What were the most popular posts this year?
- Reflections on 10,000 Hours of Programming. I distilled 31 different bullet point pieces of non-obvious programming advice.
- Wordle: What's the Best Starting Word?. I was early to the Wordle trend. I had written a program for the optimal strategy for a similar game years before, and I was able to adapt to Wordle quickly. (hint: maximize differential entropy)
- AWS is Not a Dumb Pipe. Will hyperscalers win at building higher-level SaaS or only focus on infrastructure-level building blocks? A comparison to telcos and why this time it's different.
- Why Did Heroku Fail?. Nearly every tech company has built an internal PaaS at scale. Yet, companies that sell PaaS haven't been as successful. Heroku continues to win developer mindshare but isn't as ubiquitous as its reputation.
- An Overview of Docker Desktop Alternatives. Four layers of abstraction deep can get confusing for even the most seasoned engineers. Clearing up some misconceptions on an area I worked on at Google.
- Hire for slope, not Y-Intercept. An algebraic metaphor for life.
- Pair Programming Doesn't Scale. A contentious post (to say the least). Thinking through one facet of scaling engineering teams and managing large projects. A practice that is ubiquitous in the technology consulting field but surprisingly absent in nearly all big technology companies.
As I wrote more posts, I generally categorized them into three overlapping buckets.
- engineering for technical posts (166)
- startups for strategy, trends, and management (146)
- misc for everything in between (110)
So what are my plans for the coming year? I think I'll keep blogging daily. I would also like to write longer Substack-like pieces that synthesize ideas from multiple daily posts.