When I ask people why they hate Comic Sans, I generally hear "It's an ugly font" and nothing more than that; here’s a more objective critique of comic sans (for later). The hurriedly designed font, however was ill-fated to be in desktops all around the globe while the desktop publishing boom happened. In the limited selection of fonts in early Microsoft software, it stood out as the most accessible option, fitting a variety of purposes. This accessibility caused it to be overused to the point people started detesting it — the irony.

History repeats itself with the coveted and celebrated “Ghibli” style being just a prompt away. I will not wade into the murky waters of legality or morality; that discourse has been chewed to pulp on the endless scroll of social media. A multi-billion dollar company has very well factored it in and counts the virality as a win in the popularity contest. The line’s gotta go brr.

Yet let us be clear: style is not substance. Like any trend, it will fade, much like how overexposure turned Comic Sans into an object of disdain. But Ghibli or Miyazaki won’t lose anything because what truly matters is story and immersion — substance beyond aesthetics. This isn’t new. China has an entire industry dedicated to replicating popular products—watches, fashion, paintings—because there’s demand for cheap imitations. But the fakes, in their very existence, reinforce the value of the real. AI-generated content operates similarly, flooding the internet with low-value replicas that, paradoxically, make authentic works more valuable.

What’s been also fascinating is how artists and designers have been fiercely defending Ghibli’s craftsmanship while celebrating AI’s ability to write mid code. The models have been trained on countless Github repos, stack overflow threads and forums, the legality of which is still a blur.

The hypocrisy is evident: AI-generated art is sacrilege, but AI-generated code is liberating. Good code, after all, is an art form—its beauty lying in elegance, efficiency, and the quiet mastery of years of experience. And if craft is to be revered, why should it not be so across disciplines?

What I still don't understand is Big Tech’s endgame. AI models churn out terabytes of meaningless content daily. This essentially seems like tech giants are cannibalising their own storage platforms for essentially meaningless garbage. Sneakers of the trend cycle—Sambas, New Balances, yesterday’s hype—at least have the dignity of disintegration. Unlike the sneakers though, I wonder what value (beyond shareholder value that is) is generated by years of GPU cycles dedicated to mediocrity? Yet it perfectly fits in the society’s narrative of overconsumption of things.

Despite AI’s ability to mimic well, when it comes to art originality isn’t just probability—it’s wit, nuance, and intended unpredictability. The AI-content hangover is already creeping in, a dull nausea of sameness and I don’t think Miyazaki is worried.

I don't believe that the tooling for UX designers is mature and most designers aren't even aware of it. UX design, by its nature, involves managing numerous screens, multiple iterations, and a wide array of features.

The Need for Robust Version Control - For All Plans

A version control system is a tool that tracks changes to files over time, allowing users to save, compare, and revert to previous versions. For designers, this means never losing work, easily managing multiple iterations, and collaborating without confusion over file versions. Much like Git helps developers coordinate code changes, a version control system for design would streamline workflows, reduce redundant work, and ensure that teams always have access to the most up-to-date designs.

Currently, version control exists on Figma's enterprise plan and that's my gripe with it. Like it's counterpart git, it should be accessible to even the free plan to develop a habit of keeping track of changes, specially when working in a team. Learning it early is super important, like any other design skill.

The Performance Bottleneck of Web-Based Design Tools

Another key issue is that most modern design tools, including Figma, operate as web applications within a browser container. While this offers benefits like cross-platform accessibility, it comes with considerable performance trade-offs.

The 'native' apps rely on Electron.js, which runs on Chromium, causing performance bottlenecks. This leads to noticeable lag in interactions, slow file loading, and high memory usage, making it difficult to work efficiently with complex design files. This setup severely limits the number of processing threads the application can utilize and the amount of RAM it can consume. As a result, working with large files—especially those exceeding 2GB—becomes a gamble, with frequent slowdowns and potential crashes.

A key issue is that designers often lack exposure to better tooling, particularly in version control. Unlike developers, who learn versioning systems early, designers rarely encounter such tools in formal education. This lack of familiarity contributes to a lower demand for such solutions, even though they are crucial for managing complex design projects.

I recently stumbled upon an economic times article: India's shiny new metros are costly white elephants.

I have notes.

The author presents three key arguments: Metros vs. suburban rail, Metros vs. buses and electric three-wheelers, and the claim that a metro system is a failure if it doesn’t achieve optimal ridership immediately.

Metros vs. Suburban Rail

While expanding suburban rail networks is a valid point, metros serve a different purpose—they are more surgical in their reach. Metro stations can be built directly over or under busy roads and squares, offering closer proximity to where people need to go. In contrast, suburban rail stations require larger land acquisitions, making them costlier and more challenging to develop.

Metros vs. Buses and Electric Three-Wheelers

1. Flexibility vs. Reliability: Bus networks are flexible and offer better accessibility, but they still rely ply on the same roads, shared with hundreds of private and commercial vehicles. Till ther's no cap on private vehicle ownership, increasing the number of busses would just congest the existing roads further.
2. Traffic and Speed: Buses are essentially slower, fragmented trains operating on already choked roads. Frequent stops further worsen traffic congestion. Dedicated bus lanes? Not an option in most Indian cities.
3. Efficiency and Maintenance: Metros outperform buses in ease of use, efficiency, comfort, security, and maintenance costs. Standardized trains, carriages, and spare parts lead to economies of scale.
4. Controlled Environment: Metro tracks operate in a regulated setting, leading to fewer accidents and lower maintenance costs.
5. Public-Private Model: Metro systems often involve public-private partnerships, resulting in better service quality, monetization opportunities, and easier system upgrades.
6. Navigation: For non-locals, metros are easier to navigate since they are designed with structured planning, unlike unpredictable bus routes.
7. Three wheelers: Plagued with chaotic road behavior they follow the agility of two-wheelers while occupying space equivalent to a four-wheeler – not the best option for long commutes.

The Low Ridership Argument

New metro projects often struggle with ridership initially due to limited coverage. However, the fundamental issue even for scaled metro networks perhaps is walkability and last-mile connectivity— often ignored during such over-simplified discussions. Whether it’s a bus, train, or metro, accessibility depends on how easy it is to reach the station. Poor connectivity discourages many from using public transport. Three-wheelers here, are particularly effective as last-mile commute options and not end to end solutions.

Despite the figure of speech, the comparison of metros to "white elephants" is ironic. A more fitting comparison would have been SUVs—heavy, often carrying just two passengers, and crawling through average weekday traffic like a sluggish elephant.