Readworthy blogposts #4
I have collected a bunch of interesting blog posts I want to share from the last month. I will continue doing this, but I will from now on switch to simply making one post per link - with my short personal comment. I have concluded that collecting intersting links over weeks, only made it harder for me to actually sit down and write these recommendations. Breaking it up and sharing it when it is fresh, seems like the better approach.
I have no experience with the term âvibecodingâ, but it has sure been popping up a lot recently. Usually in a mocking context. Ava touched upon the privacy nightmare, which is another good point in the long list of dangers of coding with LLMs, that the hard part is taking it the step further from MVP. Production code needs security, not just a proof of concept.
David wrote a great post titled "United We Watched, Divided We Scrolled" which goes into what it means that we donât have the same shared media experience as we used to. The frame of reference is American, but I think it applies to every modern society. Fragmentation might bring us more personalised experiences, where we can all get something that fits closer to our personal taste, but the price is lack of cohesion and shared experiences.
Now for something work related. Philipp wrote a convincing argument for bigger pull requests, which goes contrary to the philosophy I practice with my team. We like small PRs, several times a day and often deploy to production several times a day. I understand the argument that it can be harder to view the entirety of a new production ready change, when it is spread across several different branches and PRs. We try to solve that with not working on too many different things at once within the team, and make sure our changes can be deployed to production without impact to existing functionality. Though we are a pretty small team, so that approach might be difficult in larger environments with many developers and different teams working on the same codebase.
Fatih Arslan has written about AI coding which is also challenging by worldview, because I am very much in the skeptical camp on relying too much on AI tools. Even though I work with software development, where there is potential to be great benefits, I havenât been using these tools for much more than the occasional rubber ducking. And even that has been more amusing than really helpful. But I appreciate a post like this about someone who has changed their mind on something.
Staying with work theme, Mathew Dugganâs post about Slack really hits home. We use Slack heavily at work and I probably to some extent guilty of keeping myself busy by answering quickly on Slack. Mathew has many excellent points that I think any development team using Slack or a similar platform, should at least consider what applies to them and if there is room for improvement. I try to take the blunt of the various questions and ad-hoc tasks that arise in the company, so the rest of my development can stay focused on actual development. However, even I as a manager by definition need to accept more context switching, I probably donât need to have Slack as the top priority all the time.
Speaking of focus, I enjoyed this post on the loss of focus as something I can really relate to. Everything moves so fast now that we rarely give ourselves time for proper deep focus on anything. I highly recommend reading through this, and try to keep focus and not skim read.
Justin Merhoff has written a thoughtful piece on leadership inspired by his uncle who worked on the original Star Trek series. I enjoyed this little personal piece of history, both as a Star Trek fan and as someone who has a leading role in my career.
Nik has written about how much is possible in just 20 minutes exemplified with older anime shows. I think that applies to tv shows in general, that has aimed for season spanning arcs rather than self-contained stories told in under an hour. I miss those shows too, because they were still able to form a larger story within those single narratives. Think Star Trek or X-Files for example.
Finishing with Carlâs post about how gatekeeping made computers more fun in the past. It might not be the most sympathetic thing to say, but I admit I tend to agree, that "Computers were more fun when they weren't for everyoneâ. Not that we should roll the technological progress backwards though.