I don't recommend Tailwind CSS

5 hours ago

Tailwind requires learning dozens of utility classes, slowing initial development.
It breaks the separation of structure and design, bloating HTML and reducing readability.
Class naming is inconsistent, causing confusion (e.g., items-center vs justify-center).
The design system is not enforced due to arbitrary values, relying on discipline.
It is not a good gateway to learning CSS, as it abstracts away fundamentals.
Code readability suffers, especially without @apply, and HTML order does not determine priority.
Debugging in Dev Tools is cumbersome due to class soup.
Modern CSS now offers native features (layers, nesting, variables) that reduce Tailwind's necessity.
The tool has costs that may outweigh benefits for medium/large projects.

<antirez>5 days ago

Complexity is being ignored when adding features or optimizing, harming software quality.
Complex build systems and absurd dependency chains make software bloated and fragile.
Telling new programmers not to 'reinvent the wheel' hinders learning and innovation.

HTML5 semantic tags like <main>, <article>, <nav> help structure pages; less obvious elements like <ul>, <details>, <dl> are also useful.
Avoid excessive wrapper elements; prefer semantic markup and CSS that works with it.
Layout is fundamentally hard in GUI frameworks; understand the layouts a system allows rather than forcing your own.

The author refactored a colleague's repetitive but working code for resizing shapes, removing duplication by grouping behaviors by direction and shape.
The refactoring was reverted by the boss because it reduced collaboration trust and made future requirement changes more complex.
Obsessing over clean code and abstraction is a common phase, but it can lead to over-engineering and neglecting human factors like teamwork.

AI-driven development is breaking traditional code review: faster code generation caps velocity at slow human reviews.
Asynch diff-centered code review, ~20 years old, fails in an era where writing code is cheap but reviewing isn't.
Diffs lack context, requiring reviewers to reconstruct intent and design, adding expensive overhead.

The cost of producing plausible code has collapsed, but broad claims about dramatic speedups, obsolete practices, or halving headcount are unproven.
Management practices should be audited based on their underlying assumptions: those resting on code cost need review; those on human coordination, trust, or verification remain valid.
Evidence of speedups is smaller than noise; honest instrumentation is needed to measure actual acceleration, not felt speed.

<antirez>5 days ago

Complexity is being ignored when adding features or optimizing, harming software quality.
Complex build systems and absurd dependency chains make software bloated and fragile.
Telling new programmers not to 'reinvent the wheel' hinders learning and innovation.

HTML5 semantic tags like <main>, <article>, <nav> help structure pages; less obvious elements like <ul>, <details>, <dl> are also useful.
Avoid excessive wrapper elements; prefer semantic markup and CSS that works with it.
Layout is fundamentally hard in GUI frameworks; understand the layouts a system allows rather than forcing your own.

The author refactored a colleague's repetitive but working code for resizing shapes, removing duplication by grouping behaviors by direction and shape.
The refactoring was reverted by the boss because it reduced collaboration trust and made future requirement changes more complex.
Obsessing over clean code and abstraction is a common phase, but it can lead to over-engineering and neglecting human factors like teamwork.

AI-driven development is breaking traditional code review: faster code generation caps velocity at slow human reviews.
Asynch diff-centered code review, ~20 years old, fails in an era where writing code is cheap but reviewing isn't.
Diffs lack context, requiring reviewers to reconstruct intent and design, adding expensive overhead.

The cost of producing plausible code has collapsed, but broad claims about dramatic speedups, obsolete practices, or halving headcount are unproven.
Management practices should be audited based on their underlying assumptions: those resting on code cost need review; those on human coordination, trust, or verification remain valid.
Evidence of speedups is smaller than noise; honest instrumentation is needed to measure actual acceleration, not felt speed.

Hasty Briefsbeta