If I could only give one piece of feedback for the rest of my career on all tech marketing content it would be this: rewrite to describe the activities done, not the outcome achieved.
Posts in "tech"
Claude Skills are top on my list of “important things no one is talking about” for this year. They’re both an AIPaaS and showing a new programming model and mindset. The educational angle (“View Source”) is a good take.
🔗 What MCP and Claude Skills Teach Us About Open Source for AI
The original is long, so I finished reading it with a summary from one of the discussed robots:
🤖 MCP, Skills, and the Architecture of Participation in Open Source AI
Summarized by AI.
Open source AI is not just about releasing model weights. True innovation comes from an architecture of participation, where developers can inspect, modify, and share small, composable components. Historical breakthroughs like Unix, Linux, and the early web succeeded because they allowed modular contributions—viewing source, remixing, and building on others’ work—rather than requiring deep engagement with the most complex layers of the system.
Anthropic’s MCP (Model Context Protocol) and Claude Skills embody this participatory model. MCP servers let developers give AI systems new capabilities via simple, inspectable interfaces to data, APIs, and tools. Skills are atomic, shareable instructions—bundled expertise that can be read, forked, and adapted. This is the opposite of OpenAI’s GPT “apps,” which live in a closed, app-store-like ecosystem where internals can’t be inspected or reused. Skills and MCP servers are components, not products, and their openness allows a collaborative ecosystem to flourish.
The long-term potential lies in creating “fuzzy function calls”—reusable, human-readable instructions that formalize what LLMs already understand. Just as early compilers and UI toolkits let developers move “up the stack,” MCP and skills will let participants focus on architecture and composition rather than raw code generation. This evolution could preserve mass participation even as layers of abstraction and complexity emerge, as the web did with HTML, CSS, and JavaScript frameworks.
The economic stakes are high. Today’s AI market is extractive: training data is used without recognition, value capture is concentrated in a few companies, and improvement loops are largely closed. MCP and skills could enable participatory markets, where contributions are visible, attributable, and shareable. To reach this future, the AI community must embrace open protocols, inspectable artifacts, new licensing models, and mechanism design that fairly rewards contributors and encourages ecosystem growth.
The future of open source AI will be decided at the interface layer, where ordinary developers and even non-programmers can create reusable skills leveraging their own expertise. If AI development mirrors the open web instead of proprietary app stores, it could become a generative ecosystem that expands opportunity rather than consolidating power.
Links
🤖 What MCP and Claude Skills Teach Us About Open Source for AI - Explores how MCP and Claude Skills could enable a participatory, open-source AI ecosystem similar to the early web, contrasting it with closed, app-store-like approaches.
Summarized by ChatGPT on Dec 3, 2025 at 7:04 AM.
Getting ready for an AI app influx. Did we learn from the digital transformation era?
Will IT get ahead of the chaotic introduction of a new technology, AI, into their organization? Probably not, they rarely do, creating Shadow Whatever the New Tech is. But, for those that do, here’s Tony and I’s recommendations. Platform engineering’s etc.
This our second Tanzu Talk livestream. We do it weekly on Tuesdays at 4pm Amsterdam time/10am Eastern time.
🔗 Getting ready for an AI app influx. Did we learn from the digital transformation era?
An XDG library for Java - xdgj
Each time I write a small utility, either a command line one or for an MCP server, I need to store state and config in the file system. I’ve come up with many ways of doing it. You know, like, you want to store a default LLM prompt in a file. You want to store a D&D play log in a file, or a bunch of markdown files for D&D monsters.
What does Agile Smell Like? I tried to answer that question back in 2006 when I worked at RedMonk in the form of a PDF.
This guide helps you sniff test how Agile an organization is. A “sniff test” is a quick way to establish a gut-feel about something. It helps you determine what to do next.
RedMonk doesn’t really publish “reports” like this, but they were still doing some way back then, especially when a client asked for one. We were so progressive back then and published it under the Creative Commons Attribution- NonCommercial-NoDerivs License.
What is a platform and how do you do platform engineering at scale? Find out in this interview on tales from Home Depot and beyond.
What is platform engineering? What is a platform? How do you show platform ROI?
Tony was on the the team that lead platform engineering at Home Depot for many years. Here’s what he thinks platform engineering:
This is a livestream Tony and I have started. Tuesdays at 10am US eastern time/4pm Amsterdam time.
There’s probably around 50 million developers world-wide, and 20.5m professional developers. That’s what I’d start saying after seeing a chart from IDC’s September 2025 developer study update. They forecast it to be ~59 million in 2029. If you want to be very precise, a breakdown of the survey puts “professional developers” at around 20.8m in 2025. But even that includes “DevOps professionals” and “platform engineers.” So, to me, that’s something more like 20.5m or even 20m developers.
Compare that to GitHub’s claim that there are at least 180 million developers, just GitHub account holders.
Also, see more numbers in previous round up, from July, 2023
Managing the eels
A combined Google family bundle costing $120 to $200 a month could replace existing subscriptions and services like Apple One, though it may require changes to photo storage and streaming preferences.
The platform that needs a platform
Despite its status as a critical tool for cloud-native applications, Kubernetes remains complex and challenging to manage, with experts advising caution before adoption.