Rain Radars

A few days ago, ThoughtWorks released its Technology Radar for the 30th time. Kudos to the experts' team for consistently providing a general overview of the IT landscape!

Have you ever been in a home improvement store to buy a single light bulb, only to spend two hours wandering through the aisles, looking for an excuse to buy a nail gun, concrete vibrator, or a drywall cutter? Every shelf and every corner tempts you with a product to accomplish your work quicker, greener, or more efficiently.

The Technology Radar is the same but for your IT projects. In the Radar, you will find many interesting tools, frameworks, languages, and techniques to fill the gaps you sense you might have. Or wish to.

This time, for example, it sent me on a weekend-long research trip on CloudEvents. CloudEvents, managed by the Cloud Native Computing Foundation, standardizes event formats to enhance compatibility across various tech platforms. It sounds like a missing link.

Or take LinearB, a platform that aims to help engineering leaders use data to improve their teams' work by offering tools for measuring important metrics (DORA) and automating processes.

The Radar is a 44-page snapshot of today's software products and best practices. It offers many ideas for accomplishing your work quicker, greener, or more efficiently. And this is the riskiest part. It's similar to "conference-driven development" or "air travel decision-making"—where you go somewhere, see (or read) something, and make a spontaneous strategic decision.

The Radar has been chasing Clouds ever since. The term "Cloud" appeared in the first edition of the Radar in 2010. Now, the clouds pour rain (Random AI News). The abbreviation "LLM" was repeated 107 times. Today, the IT sector is a perfect customer for AI-based code completion systems, editors, validators, checkers, generators, and all other ways to clutter and confuse your project. You can use AI to get rid of AI to simplify and scale, too.

We have come a long way since LAMPS. Today, the landscape of tools and products is becoming increasingly complex, proportional to the promise of simplifying things. Ten years ago, the decision to build a database backend was quite simple: you went for PostgreSQL, Oracle, MySQL, or MongoDB. Maybe some Kafka or Neo4J. Today, you can choose a specific database for tensors, financial transactions, genome cracking, blockchain, and whatnot. PHP for the backend? Come on, boring; you can make your project dependent on some one-of-a-kind language and framework, trending on X today to be forgotten tomorrow.

So, if you are a software VP, architect, developer, QA, and every other role, read the Radar and know the stuff, but remember to simplify today's choices. You will thank yourself later. You can engage your team to try out stuff, but be clear about the long-term commitments of your technological landscape. Boring is good.

Oh, and the last thing for the Radar team: look at the pricing models of the products you mention. The tree-tier (free for non-commercial use, annual licensing, and "call us, thy wealthy corporations") has become an industry standard. And the middle layer got more expensive. I remember paying fees of 10-20 USD per seat per month. Now, it is rarely below 50. Is this because we spend IT budgets wandering through the aisles, looking for an excuse to buy a nail gun, concrete vibrator, or a drywall cutter?