Check this out from my pal DaShaun: in May of 2025, 60% of Spring Boot downloads were for versions no longer supported by the open-source project. That’s a lot. Too much, really.
I can see how it happens. You get an app into production, it runs fine, and then you moveon to the next thing. Meanwhile, the version drifts into “End of Life” territory. OSS support ends (like way back in mid-2023 for the Spring version DaShaun mentioned).
Posts in "tech"
Here’s a cynical, but good overview of Anthropic’s attempt to use Claude code for generic knowledge work, a tool called Cowork. One part:
Cowork may find fans among knowledge workers who have to create presentations but take no pride in their work.
Come on.
This is an application of that caustic Protestant belief that godliness requires suffering and pain. At the very least a bunch of tedious bullshit. The easy path is the Devil’s work.
Sometimes easy is greatness. This gets more true as you get older. E.g. “this overnight success only took 20 years.” In fact, isn’t making things easier, like, the point of…everything?
See also: “spellcheck and digital publishing, for those who take no pride in their writing and lithography.”
Original: Anthropic floats Claude Cowork for office work automation. See also a coverage round-up of Claude Cowork.
Winning in AI, story mode
Winning in AI is now defined as having a great chat app, not the AI infrastructure running that app. Amazon (AWS) has no chat app (that I know of, which is the point!), so people don’t think they’re winning. If AWS did have an app, and evolved it as a product, they would be at the AI table.
It doesn’t matter that AWS might have the “picks and shovels.” That’s not where the definitional power and attention and leaderboard is.
I wasn’t really sure what hardned images were, let alone “distroless,” so Tony and I were lucky to get William on this week’s Tanzu Catsup to sort it out. We also discuss how it fits into platform engineering.
This week’s Software Defined Interviews episode is with Lian Li:
In this episode, Whitney and Coté talk with Lian, a “cloud-native human” with a 15-year career in tech. Lian discusses her transition from tech to performance art, her experiences in amateur musical theater, stand-up comedy, and improv theater. She talks about platform engineering, the importance of community building in tech, and balancing professional life with personal projects. They also cover her unique improv workshops for engineers at conferences and the popular KubeCon karaoke parties she organizes.
Listen and subscribe, or watch the video (above) if you’re into that kind of thing.
The Alpha and The Omega, Software Defined Talk #554: This week, Brandon and I discuss AI’s impact on Stack Overflow, Docker’s Hardened Images, and Nvidia buying Groq. Plus, thoughts on playing your own game and having fun. If you don’t like the traditional podcast format, check out the unedited recording.
Is Your AI Assistant Creating a Recursive Security Loop? from Camille Crowell-Lee
AI-assisted coding is starting to eat its own tail: the same LLMs that write code are increasingly asked to review it, explain security decisions, and even override their own warnings. That creates recursive trust loops where “explain your reasoning” becomes an attack surface, and models can literally talk themselves out of being secure. The fix isn’t better prompts, it’s old-school architecture - separation of concerns, non-AI enforcement, and treating LLMs as assistants, not authorities.
Check out more in her article.
Virtual machines still run the world
The above is from a recent IDC white paper.
Container use is growing. Even then, VMs still run everything. Most of those containers run in VMs:
IDC forecasts that 85% of containers will run in VMs in 2028. Meanwhile, there is a huge installed base of traditional applications in VMs that will be around for a very long time.
And:
nearly all public clouds continue to run their containers in VMs for reasons of multitenant isolation, scalability, and utilization maximization.
Adding security and governance to Model Context Protocol - How Broadcom uses MCP
Here’s the benefits Broadcom is getting from using MCP (running on Tanzu Platform):
This capability allows for seamless task completion, such as linking internal ticketing systems directly through to code commitment, significantly reducing the time required for ticket resolution. Specifically, developers can utilize natural language processing to direct an AI agent to find their next task, have the AI coding assistant implement the requested changes, and automatically submit a pull request.
A great platform as a product paper, and a fun platform philosophy thereof
I like this platform as a product paper a lot. You should check it out if you’re into DevOps, SRE, platform engineering, whatever. It’s also available in O’Reilly if you have that subscription and don’t want to lead-in yourself.
Here’s some fun parts:
Adopting a product mindset starts with continually evaluating the business context to manage “build versus buy” decisions. Contextual factors such as scale, compliance requirements, or the diversity of the workforce skill base and technology stacks often require organizations to opt out of an off-the-shelf solution and instead invest in a set of integrated capabilities designed for its specific needs.