U-Zyn's Nodes

ntfy: Push notifications without signups

2026-03-15T00:00:00+00:00

Just discovered ntfy.sh</a>, a very handy push notification service that does not require any signups. Very useful for:

Setting up push notifications from a server to your phone to notify on any events.</li>

Long running LLM tasks such as Claude Code or OpenClaw</a>.</li> </ol>
Grateful that it's not only free, it does not require any accounts. Just install the app on your phone, subscribe to a topic, and start sending notifications to it!

Who is OpenClaw for? 🦞

2026-03-06T00:00:00+00:00

🦞 OpenClaw</a> is all over my feeds. Viral testimonials, threads about productivity gains, people swearing it changed their workflow. I installed it, ran it for a bit, and came away underwhelmed. Not because the tech is bad, but because I cannot figure out who it is actually for.
What is OpenClaw actually</a></h2>
Strip away the demo videos and you get a fairly straightforward architecture. OpenClaw runs a long-lived daemon that listens on messaging channels: WhatsApp, Telegram, Slack, Signal, take your pick. When you send it a message, it passes your text to an LLM, loads your memory files into context, and invokes whatever tools it needs. There's also a heartbeat: a configurable cron job (default 30 minutes) that triggers the agent even without a message, so it can check in, surface reminders, or run scheduled tasks.
The heartbeat is the interesting part. The design addresses one of the more popular wishlist items, instead of prompting to get an output, AI would be able to perform tasks without you first prompting it. Memory on the other hand: it's markdown files on disk, loaded into context each invocation. ChatGPT's memory works the same way under the hood. Skills are shell scripts paired with prompt files, essentially wrapped CLI commands with some LLM scaffolding.
The architecture is clean and straightforward.

Raspberry Pi prices have gone through the roof

2026-02-21T00:00:00+00:00


Raspberry Pi prices have been rising aggressively, driven by surging memory chip demand</a>.
What makes it worse is how quickly the hikes have come — two rounds within just two months. The most recent price increase</a> pushed some models to more than double their original MSRP.

Web browser discrimination

2026-02-19T00:00:00+00:00

An unfortunate reality of the modern web is that many websites check your User-Agent string and serve different content (or no content at all) based on what browser they think you are. With our previous UA string, we were getting degraded UIs, "your browser is not supported" pages, network throttling, and outright HTTP 403 errors from many prominent websites.
From: Ladybird Browser Newsletter: This Month in Ladybird — January 2026</a> </blockquote>

A sad case of browser discrimination. I had assumed that with the death of Internet Explorer, such practices had largely faded away. User-Agent (UA) exists to enable graceful degradation, allowing older or less capable browsers to receive simplified content, not to serve as a whitelist that reserves the full web experience for a handful of approved browsers.
When you throttle or block based on UA, you are breaking the ethos of the open web, the very reason the World Wide Web was created in the first place. It is discrimination. From the article, this appears to be fairly common across the websites and services of Big Tech companies, such as Google and Meta. When those same companies also control the lion's share of the browser market, this begins to look uncomfortably close to anti-competitive behavior.

Bambu RFID scanner on Flipper Zero

2026-01-17T00:00:00+00:00


3D printing is fun. One common situation for anyone who owns a 3D printer is that you eventually end up with a growing stack of filaments at home.
These spools come in many combinations of colors and materials, and sooner or later you'll need to top up a spool that's running low. The problem is that Bambu filament labels only show the material and variant—such as PLA Basic, PLA Wood, PLA Matte, or ABS—but not the color code or color name. This makes it surprisingly tedious to figure out which exact filament to reorder. You either dig through old purchase receipts or manually try to match colors on the website.
If you have a Flipper Zero with the flipper-bambu plugin installed, you can now easily scan a Bambu filament spool via its RFID tag to retrieve the color code, along with other useful details such as temperature configurations, specifications, and production date—information that isn't normally visible.

Passkey Security CTF Challenge for TISC

2025-10-25T00:00:00+00:00

</a>
Back in June 2025, I submitted a security Capture-The-Flag (CTF) challenge for the annual The InfoSecurity Challenge (TISC) 2025</a>, organized by Centre for Strategic Infocomm Technologies (CSIT)</a>, an agency in the Ministry of Defense of Singapore, and it revolves around Passkey.
I was lucky to have my challenge selected for TISC. TISC 2025 consists of 12 CTF challenges. My Passkey CTF challenge is 1 of the 3 community challenges that were accepted — with the remaining either developed internally by CSIT or developed/commissioned by CSIT-related entities. This nets me a cool $1000! :)

Inconsistent keyboard UX on macOS

2025-10-20T00:00:00+00:00

Remember back in the pre-GUI OS days when everything was done with commands typed on keyboard?
Remember how fast and efficient one can get with just keyboard commands and shortcuts? Remember your Lotus-123 or WordPerfect days? That's efficiency.
Before world wide web (WWW) took over everything, many systems were designed with purely keyboard-operatable shortcuts or commands.

DEF CON 33

2025-08-11T00:00:00+00:00

It has been a dream of mine for 20 years to one day attend DEF CON, and it has finally happened.
It was an amazing experience — way larger than I had imagined. What I love most is seeing passionate, like-minded people gather in one spot and show off beautiful things and projects they care about. The scale is massive: it occupied all three levels of the West Hall of the Las Vegas Convention Center, with many villages, talk areas, merch booths, vendors, party/chillout areas, and more. At any one time, there were probably 15 talks and 30 activities happening simultaneously — it was truly overwhelming. Just look at the schedule</a>.

Passkey has a theft detection feature, but Apple, Google and Microsoft broke it

2025-05-30T00:00:00+00:00

TL;DR: Passkey has a clever design that allows users to know if your keys are silently stolen and cloned, but the feature is currently not very useful today because its major providers such as Apple, Google and Microsoft do not play ball.

Passcay: Zig library for passkey auth

2025-05-04T00:00:00+00:00

</a>
Released Passcay</a> over the weekend. Passcay is an open source Zig library for Passkey relying party, i.e. websites that want to support secure Passkey authentications. Passcay is pronounced exactly the same as "passkey". In hindsight, I probably should have named it Passquay, still pronounced the exact same way, to add a touch of Singaporean flair to it.
I've been using Zig</a> as my primary language for about 1.5 month now</a>, and I'm feeling increasingly confident with it. I do miss a few things from Rust — especially ownership and the memory safety it enforces — but overall, I’m happy with the switch. One downside: LLMs don’t seem to work as well for Zig as they do for Rust or JavaScript, likely due to the smaller corpus of example code available.
I made Passcay to solve my own problem. I am working on a web service that would be having passkey as native authentication rather than as 2nd factor authentication, above the insecure and privacy-affecting email-password login.

OpenAI Codex

2025-04-17T00:00:00+00:00

Another day, another AI coding agent. OpenAI just announced Codex</a>, modeled after Anthropic's Claude Code</a>.
I have just found my happy place only a few days ago with Claude Code and Visual Studio Code Agent Mode</a> + GitHub Copilot</a>. Nonetheless, happy to be evaluating Codex. OpenAI's UX are usually polished and what I like about it already is that it's open source. This means it will get better so much faster than many of their competitors.
Tried to peek under the hood of Claude Code last week but it's all obfuscated.

Fun project 16 years ago

2025-04-08T00:00:00+00:00

Fun project. 16 years ago today, in 2009, I built this dashboard for IMDA (then Media Development Authority of Singapore [MDA]) to showcase the traffic to MDA-funded startups.

It was a "simulated" live data of all visitors to all the tracked sites, with the data source being Google Analytics. At least one of the startups then is now a unicorn. You can probably guess which one - Sea, then Garena.

Guest Lecture at NUS Business School: The Open Web

2025-04-08T00:00:00+00:00

Thank you Bernard Leong</a>, a friend, fellow entrepreneur & technologist for the kind invitation.
It was an honor to return to my alma mater and speak at National University of Singapore (NUS) Business School</a> for the Master of Science in Venture Creation students about a long-time passion of mine: the open web. In my talk, I covered the inception of the web in the 1990s, the destructive influence of social media algorithms, and the promise of the fediverse and open protocols in shaping the next phase of the web. I also explained why Web3 is not part of that evolution.
Tech companies today should look into integrating open source into their business strategies and contribute to building open protocols. I shared case studies on companies that have successfully done so, Anthropic's Model Context Procol (MCP), Meta's Llama & Singapore's own Menlo Research</a>.

A series of rabbit holes

2025-03-21T00:00:00+00:00

My journey started casually enough—working on Tixie</a>, I was looking into simpler ways to deploy Rust-based Lambda functions when I came across Cargo Lambda</a>. What got my attention was that it is so easy to cross-compile with it, mainly due to Zig</a>'s powerful build system. Even though it was built for Rust developers, it utilizes Zig for cross-compilation, allowing me to compile ARM binaries</a> that worked on AWS Graviton on my Apple Silicon Mac.
That was the start of a series of rabbit holes that consumed me for more than a week.

The Roadmap Pitfall: Companies Should Not Sell Tomorrow's Features Today

2025-03-13T00:00:00+00:00

The Roadmap Pitfall</a></h2>
If you run a company, do not release roadmaps or make future promises. You owe your customers nothing beyond what they're paying for right now. Your customers pay for what you sell today, not what you will be selling tomorrow.
I call this the Roadmap Pitfall. It is a common mistake and trap that many companies, especially startups, fall into. The tech landscape is littered with companies that have overpromised and underdelivered through their public roadmaps, hurting their reputations. Some infamous examples are Tesla's Robotaxi</a>, Meta's Libra</a> and Metaverse</a>, Google Stadia</a>. They are selling products that might become revolutionary tomorrow but not today.

Introducing Tixie

2025-03-09T00:00:00+00:00


Spelling, or 听写, is a very familiar process for parents of Singaporeans with primary-school-going children, or in fact, for parents worldwide. Every week, each of our children is to learn two sets of words and phrases, one in English and one in Chinese, to be tested in school. Each set consists of 15-20 phrases and sentences.
A usual process for us would be first having each of our children study, on their own, the words or phrases for the upcoming spelling test, and once they are ready, passing us parents the list for us to read aloud to test them. I'm sure many parents in Singapore are very familiar with it. Rinse and repeat every week.
I created Tixie to facilitate the latter read-aloud process, allowing it to be conducted solo by the students only, without involving parents or another person.

Digg is coming back

2025-03-06T00:00:00+00:00

 </div>
Digg is coming back</a>!
Digg was one of the very first social news aggregators back in the 2005-ish. It's one of the inspiration that led to me creating and launching Ping.sg in 2006</a>. Ping.sg was a social blog aggregator for mainly Singpaore blogs.
To me, that's a sign of revival of free and open web independent of big techs.

Hidden messages from OpenAI GPT 4.5 announcement

2025-02-28T00:00:00+00:00

Watch the OpenAI GPT 4.5 announcement video</a>.
Messages from the announcement, some obvious, some not so:

GPT 4.5 is going to be awesome. </li>
We got a glimpse of GPT 1 (08:34</a>), circa-2018. Spoiler alert: it sucked. </li>

The engineering team that worked at OpenAI is just as awkward in social settings as your company's engineering team. </li> </ol>

Claude Code is impressive

2025-02-27T00:00:00+00:00

My workflow with generative AI has been changing quite rapidly over the past two weeks. DeepSeek's release really sped up innovation and release cycle among its competitors.
I may draft a longer note covering my workflow next time; I'm writing a quick one to share how impressive Claude Code</a> is.
Claude Code hit a homerun</a></h2>

Thoughts on Perplexity Deep Research, after 24 hours of heavy use

2025-02-16T00:00:00+00:00

Since the introduction of Perplexity's Deep Research</a> on Feb 14, 2025, now the 3rd major deep research AI offerings after OpenAI ChatGPT's, and Google Gemini's, I have been using it rather extensively. It helps greatly on a research project I am working on.
Early findings on Perplexity Deep Research, it often takes 5-20 minutes, not quite 2-4 minutes as advertised. Perhaps it is due to the spike in usage since the introduction, or perhaps it is a tweak to make it appear as if it's working comparatively hard as the other 2 Deep Research offerings. Overall, I do not have an issue with the response time. I can also see its thought process throughout the thinking time.
I am a subscriber of ChatGPT Plus ($20/month) and Perplexity Pro ($20/month). As such, I do not yet have access to ChatGPT Deep Research, which, as of today, is only available to ChatGPT Pro ($200/month) subscribers.
My preliminary findings on Perplexity Deep Research after about 24 hours of rather extensive use.

OpenAI on Open AI Strategy

2025-02-02T00:00:00+00:00

Q: Would you consider releasing some model weights, and publishing some research?
Sam Altman: yes, we are discussing. i personally think we have been on the wrong side of history here and need to figure out a different open source strategy; not everyone at openai shares this view, and it's also not our current highest priority.
Source: AMA with OpenAI's team</a> </blockquote>
It is interesting that Sam Altman said what he said, considering the ongoing legal battle between Elon Musk and OpenAI</a> on the similar topic. It is even more intriguing that DeepSeek appears to be a catalyst for this shift in thought.

Switching cost for AI models is $0

2025-01-27T00:00:00+00:00

It must have been be a challenging week for the folks at OpenAI and other similar closed-source foundational AI model companies. The release of DeepSeek</a>, which is not only better than OpenAI's o1 model but also free, has sent shockwaves through the industry. The fact that DeepSeek has managed to achieve this feat with only a fraction of the resources that it took for other comparable models to be trained is even more surprising.
Foundational closed-source AI companies are usually making a loss on the consumer products through the likes of ChatGPT Plus, Claude Pro. It is also commonly believed that these consumer subscription-based AI products do not bring in much profits. It is not surprising to find that some of these subscription plans are even loss-making, such as that of $200 a month of ChatGPT Pro plan</a>.
The foundational AI companies are still figuring out ways to turn profitable. The revenue model today is largely relying on the API and enterprise sales, i.e. the "AI wrappers" and existing apps racing to support AI feature.

Sam Altman: The honest answer is we have no idea [on how to bring in revenue]. We have never made any revenue. We have no current plans to make revenue. We have no idea how we may one day generate revenue. We have made a soft promise to investors that once we've built this sort of generally intelligent system, basically we will ask it to figure out a way to generate an investment return for you. [The Journal by WSJ</a>] </blockquote>

The Evolution and Decline of The Open Web

2025-01-26T00:00:00+00:00

Remember a freer web?</h2>
Remember when the World Wide Web was once much more open and freer?
Individuals used to maintained control over their own websites, curated their own content, and engaged with others without the influence of big tech social media, who have gotten way too big with the network effect.
Remember when content creators were publishing either directly at their own websites and not "microblog" at social media? The web was much freer then. We did not have to worry about our content getting removed, hard censored, or soft-censored with content not favored by "the algotirhm".
A part of the decline is due to the content creators themselves, attracted by the network effect and the near-instant gratification that social network platforms provide, another part is due to the mistreatment of social network platforms with the contents that they are being entrusted with by creators, to have them syndicated and pushed to their respective followers but failing these trust.
The recent controversies surrounding TikTok bans (The New York Times, 2023; Reuters, 2024), Meta’s decision to block news content in Canada (BBC, 2023; Financial Times, 2024), and Twitter/X’s continuous policy shifts (The Guardian, 2023; Wired, 2024) underscore how a select few corporations determine the parameters of digital communication.
This shift has significant consequences. Independent creators and smaller platforms struggle to gain visibility as algorithmic filtering prioritizes engagement-driven, monetized content. Where the web was once an egalitarian space, today’s digital environment increasingly favors corporate interests over user autonomy. The web, originally envisioned as a decentralized and democratized medium, now reflects the business priorities of a few dominant firms that set and alter the rules at will.
The decentralized web before social media</h2>
Before social media monopolized content distribution, users had significant agency over what they consumed and shared. Websites such as Blogger.com allowed individuals to self-publish, while content management systems like WordPress, Movable Type, Joomla, and Drupal facilitated independent website management without reliance on major tech firms while allowing free permissionless publishing onto the Web.
Prior to the dominance of algorithmic recommendations, content discovery operated through direct, user-driven mechanisms:

Engaging in discussions via blog comments and threaded replies. </li>

Utilizing pingbacks and trackbacks to connect related content across different platforms. </li>

Curating and sharing blogrolls—lists of recommended sites maintained by individual website owners. </li> </ol>
Website administrators maintained complete control over their platforms, from advertising decisions to interface design and content structure. This diversity fostered a highly dynamic internet, where creative expression and niche communities thrived without external constraints.
Content discovery before "The Algorithm"</h2>
Content published on social media are now at the mercy of the algorithm that decides who should be seeing your content, and for how long. It would be easy to think that your content will be pushed to all your subscribers or followers on social media platforms, it was indeed done that way, but as of about a decade ago, it has stopped being done that way. For easy reference, let's call it "The Algorithm"
If you are wondering why YouTubers now specifically requesting their subscribers to "click the Bell" button, or Twitter users asking followers to "Click the Bell" icon, this is precisely the reason. It is no longer enough to simply be subscribing or following a friend to be getting all updates from the friend as The Algorithm dictates what gets syndicated and federated.
I will write more about "The Algorithm" in the future.
Before The Algorithm, we manage syndication, federation and discovery through:

Web Aggregators and Directories – Platforms such as Technorati, Ping.sg (which I launched in 2006), Tomorrow.sg, and Project Petaling Street curated and promoted blog content, allowing users to explore diverse perspectives. Many of these platforms have since become defunct due to the decline of open web and independent blogging. </li>

RSS Feeds – Users subscribed to preferred websites using Google Reader and other RSS aggregators, ensuring direct and unfiltered content access. Many of RSS readers such as Google Reader have also gone defunct. RSS, which was quite the norm of any content-focused websites back in and around 2010 have also started to be quietly dropped or not feature them prominently anymore in header. </li>

Manually Curated Networks – Users discovered content through personal blogrolls and link exchanges, fostering a decentralized and intentional browsing experience. </li> </ol>
In this model, users actively shaped their digital experiences rather than relying on profit-driven algorithms to determine what they saw. Content quality and subject relevance took precedence over viral trends and engagement metrics. This also explains the limited quantity of popular at that time. We had Badger Badger, Star Wars Kids and not too many of them.
The rise of "The Algorithm"</h2>
Initially, social media functioned as a supplementary tool for content creators, offering microblogging capabilities that linked back to full-length articles. Platforms such as Twitter, Facebook, and MySpace were originally designed to complement, rather than replace, independent publishing. Early users were using these tools to post links and snippets to their full articles, often published on their own websites. However, as user behavior adapted to shorter, immediate content formats, social media began to supplant traditional blogging entirely.
Initially, microblogging was rather awkward. Twitter tried to make it less so by labeling the content area with "What are you doing?", encouraging you to share less thoughtful content on Twitter rather than publishing on your site. The dopamine associated with the instant gratification has made microblog becoming more common and less akward. We stopped posting only what we were doing but more thoughts and later starting to find 140 chars terribly limiting (it was due to SMS char limit).
We started posting less and less on our websites and most services that were fascilitating content creation, hosting, federation and syndication started to falter (cite more examples, such as Google Reader shutting down, Technorati losing traffic, RSS losing trend). Big social media started to gain momentum, and with that more power, and started to play a role in deciding what content we see, or should not be seeing. This is also known as "the algorithm". "The algorithm" is not merely referring to censorship, but generally presenting content to readers not in any logical order, reverse chronological or any other form, but in order to get readers to be hooked and start doom scrolling.

Aggregator sites like Technorati lost relevance, limiting independent publishers’ reach. </li>

Chronological feeds were replaced by algorithmic rankings, curating content based on platform priorities rather than user preferences. </li> </ul>
Path of destruction</h2>
"The Algorithm" is engineered by the big tech social media in order to maximize viewership time. It started with an innocent article/video recommendations, and eventually resulting in a blackhole-like mechanics such as that of TikTok or YouTube Shorts where it's so effective that viewers, akin to a zombie, wired to their screen continues, without any pause, to be fed with videos after videos that are brief and designed to fully suck you in.
With "The Algorithm", users of social media then try to beat it, in order to get more views or interactions. This is similar to Search Engine Optimization (SEO) practice that was used by websites, usually commercial, that tried to exploit popular keywords to gain more organic visits resulting from search engine hits.
These changes led to passive consumption habits, where users primarily engaged with whatever was surfaced by the algorithm rather than actively seeking out diverse sources. This shift also facilitated the creation of filter bubbles and echo chambers, wherein users predominantly encounter viewpoints aligned with their existing beliefs.
Additionally, social media platforms now monetize visibility, requiring creators to pay for audience reach through promoted posts and advertising. This model disadvantages independent voices while reinforcing corporate hegemony over online discourse. At the same time, platforms collect and exploit user data, leveraging it to fine-tune engagement-maximizing algorithms at the expense of informational diversity.
Reclaiming the Open Web</h2>
This discussion is not merely retrospective. This is a call to action. The web was conceived as an open forum for information exchange and self-expression, yet the endless viewership optimization of the algorithm has robbed the original intent away.
Concrete steps towards this goal:

Reviving Independent Publishing – Supporting personal blogs and self-hosted websites to reduce reliance on social media platforms. </li>

Using self-controlled Discovery Tools – Encouraging the use of RSS feeds, newsletters, and independent aggregators to follow content without algorithmic interference. </li>

Challenging Algorithmic Domination – Promoting content distribution models that do not rely on opaque ranking systems designed to maximize engagement rather than inform users. </li>

Advocating for Open Web Standards – Endorsing policies that ensure equal access to web infrastructure and reduce corporate influence over digital spaces. </li>

Educating Users on Digital Autonomy – Raising awareness about the implications of centralized control and encouraging critical engagement with digital platforms. </li> </ol>
The web should facilitate intellectual exploration, creative freedom, and meaningful connections—not merely function as a revenue-generating apparatus for tech giants. As users, we possess the agency to redefine our relationship with digital content by prioritizing independent platforms and open web practices.
I remain committed to this cause through advocacy, public discourse, and the development of technological solutions aimed at preserving internet openness. By collectively working toward a decentralized web ecosystem, we can reclaim the web’s foundational ethos—a free, open, and user-governed space.

Self-publishing

2025-01-13T00:00:00+00:00

A proud moment that started with a simple catch-up with a good friend.
I briefly shared that I am working on a project related to self-publishing. My friend, Chandra, shared that he has been considered blogging and asked me to set one up for him on the spot. He also shared that he would prefer self-publishing than contributing content on social media platforms. That was music to my ears!
I am on a mission to bring self-publishing back. More on that later.

It is 2025, build like it's 2005!

2025-01-10T00:00:00+00:00

Let’s Build Like It’s 2005
Software engineering has become overwhelmingly complex.
Want to build something today? Good luck. First, you’ll need to wade through an ocean of decisions: which tech stack to choose, how to set it up, which JS/CSS/TS/TSX frameworks to use, which build tools, linters, and testing environments to adopt, how to structure your project, organize your repository, and configure CI/CD pipelines. The list goes on. If this is something to built by a product + engineering team, despite more headcount, the complexity is often compounded even more.
Even after spending days setting everything up, you’ll never quite feel satisfied. You’ll tinker, tweak, and second-guess every decision for weeks. You’ll obsess over improving your stack, striving for a perfection that always seems just out of reach.
Meanwhile, only 5% of your time goes into building your actual product. The other 95%? That’s spent on the stack.
It's exhausting. It's frustrating.

Hello again

2024-11-28T00:00:00+00:00

Hello again.
I started blogging in 2003 on b2, the precursor to WordPress. But over time, the updates slowed down, and the blog eventually faded away.