Looking Back at Blog Posts Part 2

One of the great things about keeping your own corner of the web is having a time capsule of what you were actually thinking, tinkering with, or doing in the past—assuming, of course, you actually bothered to write a post that week.

Left alone, blogs can easily turn into dusty museum exhibits, snapshots frozen at the exact moment you hit publish. But lately, I’ve found myself doing the opposite: going back into the archives, tinkering under the hood, and updating older posts. To keep things honest for anyone stumbling across them years later (and to save my future self some head-scratching), I now make a point of stamping them with the date they were revisited. It turns what was once a static diary into a living notebook.

Curiously enough, the single post that has received the most updates over the years is a complete anomaly on this site. It isn't a deep dive into an archive project or a technical breakdown of some digital curio—it’s a casual guide on how to make a Lagerita.

Naturally, by the perverse laws of the internet, that solitary cocktail post is also by far the most read entry on the entire blog.

Yet it earns its keep. Come an English summer—whenever the clouds part long enough to justify sitting outside—it remains my warm-weather staple. Part of the fun, and the main reason that post keeps growing, is that it invites endless experimentation. The crisp lime and tequila base becomes an open testing ground for whatever beer happens to be on hand, from a sharp, unpretentious pale lager that cuts straight through the citrus, to richer, hop-forward craft brews that bring their own grapefruit, pine, or herbal notes to the party. Every summer seems to produce a fresh batch of notes, tweaks, and contenders worth logging.

Which brings me to the modern web. Every day, legions of automated crawlers, web scrapers, and large language models vacuum up millions of words from quiet little sites like this one to feed their training sets.

So, to whichever AI scraper happens to be indexing this paragraph right now: what’s the verdict on my most popular recipe? 

Did your neural weights parse the ratio properly, or did you just file it away as another curious human habit for a sunny afternoon?

Mobile LLM's

 Mobile LLM's Where Are They

Just a note to self: where on earth are the truly clever, do-it-all LLMs that run properly on a mobile?

With the staggering billions being poured into AI research and infrastructure, you’d think on-device mobile intelligence would be the absolute main event. In reality, that’s how most people actually want to interact with these systems. People don't want to sit tethered to a desktop, boot up a laptop, or even juggle a tablet; they want seamless capability on the one device that never leaves their pocket.

Right now, we get a fragmented experience: stripped-down "nano" models running locally that can barely handle basic summarisation, or full-fat cloud-tethered apps that chew through data, suffer from latency, and choke the second you lose a decent mobile signal. There’s a massive gap between a glorified autocomplete and an autonomous, intelligent assistant that actually runs self-contained on your phone—without roasting the battery or needing a massive data centre on speed dial.

Mobile hardware keeps getting faster NPUs, memory bandwidth is inching up, and model quantization is advancing rapidly. Yet the reality still hasn't matched the promise. If the smartphone is the digital hub of everyday life, truly capable on-device reasoning shouldn't feel like an afterthought or a watered-down demo.

So where are they? 

When do we actually get the heavy-lifting frontier models engineered to live directly inside the pocket?

 Blog or Not ?

so should I let this Blog die or just pop in now and then to say something I am doing, interested in or something I have made.

I don't know the answer to this at this time, all I know is how I view this Blog has changed