31 May 2026: Updated this blog's template to be fully HTML5/CSS. Removed the few remaining HTML 3.2 tags that were a carryover from when the blog was manually updated. Creating new templates based on this sample will be much simpler. Find the new file at the Free Apps page.
This template is CSS 2.1 with one CSS3 property (border-radius for the rounded corners).
28 May 2026: I came across a nice link today - AI short films. The page doesn't rate the shorts, just provides some basic information.
I went through a few of them and kinda' liked the following (meh-ok):26 May 2026: Is this the original google? The site udm14.org certainly feels like it. Here's a 2024 write-up with details (comments). I had no idea this existed and honestly... while it makes me shudder to say this because it's google, it feels pretty good to use (nostalgically).
It's interesting to experiment with but I'm quite happy with the established privacy-respecting search alternatives that have emerged over the years. They have long-surpassed both the modern and classic google in everything that matters.
25 May 2026: It seems that Colorado and California will exempt open source operating systems from increased government surveillance known as Age Verification. BigTech - an extension of the state - continues to support these Anti-American bills in a sign of allegiance to the new zio-fascist regime. I wrote a little about this previously on April 3-5, 2026.
@davidgerard@circumstances.run sarcastically posts - "i'm sure systemd and friends will be reverting those age verification system commits promptly". The responses from linux users have been equally on point.
Another discussion at Hacker News.
21 May 2026: As I started to add bookmarks to the Static Bookmark Generator (see 17may post), I quickly found that a long list of tags can visually overwhelm the page. At one point I had 15 bookmarks and ~25 tags.
I still like this script as a solution, but I am finding that I need to limit tag names to broader categories. For example, instead of tagging a specific bookmark as 'comics, marvel, spider-man', I now just use 'comics' as the filtering tag. This keeps the tag list manageable and scannable.
Currently, I have 36 bookmarks and 18 tags. I wonder what the tag list will look like when I have a couple of hundred bookmarks.
21 May 2026: I totally misread what the smolweb HTML spec was trying to do when I wrote about it yesterday. I skimmed it and thought it was aiming for the web version of smolnet - in line with Gemini or Gopher. It's not. It's HTML5 with lots of fat trimmed. It's a great solution for a lighter modern web but extremely heavy as a smolnet experience.
HTML 3.2, while not light compared to the core smolnet protocols, is a much better fit for the light HTML experience I'm trying to create on these pages - deprecated tags and all. I'll keep using it along with a little sprinkling of CSS as needed.
20 May 2026: I came across a proposed HTML specification for the light web. I really like the idea and the guidelines it sets. It's encouraging how much work and thought went into it. They've even built their own validator.
I'm going to do some preliminary work on converting some of my pages (which mostly adhere to HTML 3.2 standards) into these concrete guidelines (with caveats, see below). I'll see where this takes me. Any new pages I create will 100% be using (a portion of) these specifications - the light web needs something like this.
My immediate two concerns: the spec allows different grades of CSS - from Essential & Lightweight (grade A) to Heavy Performance Impact (grade F). That's too fuzzy. There needs to be a clear cut-off after grade C - no compromise. The other concern is the allowance of the <script> tag and JavaScript. If a page needs scripts, it's not the light web (or a light client).
Still, it doesn't mean I can't use a subset of the guidelines and work within those constraints.
A new update at the 'AI Said...' page about Service Workers. Frenemy or Enemy? If you do nothing else, take a look at the (number of) Service Workers running in your browser (how-tos, bottom of linked page).
The fingerverse page saw some minor cleanup and the addition of two clients: chawan and finger (stdout).
Static Bookmark Generator: The line heights finally look right in all places.
Added a couple of links to my ASCII page but honestly... it needs a rewrite. I'm not sure what I want this page to be.
19 May 2026: Beautiful ASCII Animated page. You can really lose yourself in this.
update: It was my locked-down Firefox that had some problems. Other FF variants loaded the page smoothly.
18 May 2026: Managed to get a few tweaks done today - some cosmetic, some fundamental as to how the data is processed. This is the new front face. Relative linking works great, absolute linking to local files, word wrapping, white space issues improved, fonts... all looking good.
RSS testing, a bookmarks file and real world use to follow.
I'm still not exactly sure how I'm going to use this. One large DB of links or separate bookmark files for specific things (browser extensions for example). I've got a ton of links that are wasting away and I think this script is going to help with access and rediscovery issues in my current setup.
17 May 2026: I have been really impressed with postmarks - a single-user bookmarking application. I learned about it by visiting Ben Brown's personal bookmark page. It makes a great demo for postmarks and has made a nice addition to my RSS feed.
Long story, short. I needed to implement something similar for neocities and personal use because my current organizational methods aren't nearly as effective.
I spent the entire day working on a poor man's, static-build version of postmarks. I like the results so far. It's portable (only a single html file with all the bookmarks as arrays; plus an RSS file). Like the Static Blog Generator of last week, it requires a little manual work (running a python script against a text file).
This is a demo screenshot of the Static Bookmark Generator along with the most recent documentation. I need to do further testing (especially with RSS). The generator will serve as both an app (with small demo bookmarks.txt) and likely a new page (or set of pages) on this site. I would love to say I'll have this ready for upload in a week or so.
15 May 2026: Updated a number of random quote files on the Free Apps page today. Minimal updates to Stoicism, John Gray and Dog Facts. Added new quotes to the command-line quotes file - 15 great Agent Smith Quotes.
14 May 2026: Quick note... finally tackled the smolnet page. It's been sitting there unattended for months. Its purpose is to provide links to both the smolnet and the light (text-friendly) web.
There aren't too many browsers that can handle all the different protocols included on the page, so it's really being built specifically for Chawan and Dillo but... with extensions, regular web browsers should handle most protocols with minimal issues.
On a side note: excited to be working with 'Dillo' again as well as testing the 'Chawan' browser (it seems remarkably capable at this early stage).
13 May 2026: I watched the entire film last night for the first time ever. The original silent version never quite clicked with me. I only ever saw a few minutes here and there.
Metropolis wasn't just ahead of its time - it lit the trail for almost every science fiction and fantasy film that followed. The modernized version completely engaged me on so many levels.
If anyone decides to give this a go now, the 2026 MetropolisRemix is a great modernized enhancement of the original. Or you can wait until 2027 for its 100-year anniversary.
13 May 2026: I previously had two random quotes htmlApps relying on third-party APIs and a CORS proxy. While the services were great, the moving parts caused some problems. To fix this, I created a minimalist, self-contained template to handle this kind of data.
The two affected pages - 'Random Dog Facts' and 'Stoicism Quotes' - have now been updated to be fully self-sufficient, with the data stored directly as arrays within the HTML.
I also finally took the opportunity to create the John Gray Quotes page I had mentioned on the Free Apps page many months ago.
12 May 2026: Do I want to watch Metropolis (1927)? The images look amazing! Youtube has a lot of them but I'm leaning toward MetropolisRemix's latest 2026 reworked colorized & dubbed version.
How can I say no? 😀 🍿
11 May 2026: Uploaded a python-based Static Blog Generator w/RSS support written specifically for neocities.org but will work on any web host. It basically takes a bunch of individual posts (written in HTML) and converts them into this paginated blog page. It's on the Free Apps page.
The script comes with documentation that should cover everything. It really is just a simple blog maker that requires a little bit of manual assistance to cover the basics of a blog.
It's written and tested on Linux but should be compatible across the board (macOS, *BSD and Windows).
07 May 2026: Reticulum does one thing and does it well: it connects anything to anything. It's a hardware agnostic network stack bridging various networks like LoRa (very low bandwidth), Wi-Fi, Ethernet, TCP/IP networks (does not use IP addresses) and many more. Reticulum does not care what physical hardware carries the bits (radio, wire, fiber, serial cable).
It's an amazing network stack built on a foundation of encryption with various ways to communicate securely: one-to-one, a general transmission to everyone nearby, and messages to many recipients at once.
YouTuber, Data Slayer, does a good job of demonstrating and testing the Reticulum network in his post called The Internet, Reinvented. It's a lot to take in but it's because there are so many different technologies that Reticulum is handling in the backend.
Not too long ago I wrote that the BBS infrastructure was "the first grassroots built, user-maintained, national digital communications network" and that "I'd go so far as to call it the most powerful people-driven communications technology yet created." It saddened me that it might've ended there. That the magic was over.
I have been deep-diving since, looking for hope in a solid modern alternative. That's when I found Reticulum. I'm now convinced that the same kind of resilient, decentralized, grassroots, standalone network can exist again (if not in this form, than in another).
Kick the tires a bit (browse nomadnet (LXMF protocol), runs on top of Reticulum) with the online Ren Browser (git). On the upper left-hand corner there's a drop-down that lists sites that are online - example screenshots: a chat, a sample of a download site.
Nomadnet: [Github] [Reddit] [Micron Markup Editor] [Why Micron?]05 May 2026: Updated AI Said... with an article on 6G's new Integrated Sensing and Communications (ISAC) feature. It's a portable radar system enabling a cataclysmic end to the little privacy we have left.
This new form of surveillance threatens to will arrive in a few years - after the mandatory online ID verification takes hold, VPNs are outlawed and our one-party government moves the country even further toward fascism.
04 May 2026: These Scientology speedruns are insane! 😂 They have spread everywhere and in every form.
01 May 2026: Another minor update to netfinger (nf) at the Free Apps page. Finally fixed a line that was bothering me for a long time - clarity in as few words as possible. The line now reads as:
-r --render HTML as <pre> with clickable links
Perfect! It's the little things:-)
On the short stories page, I decided to put the better stories on top (8 stories, unranked). The section below (5 stories) may be deleted if I can't improve the stories.
Spent some time fixing 'Shadowking'. It lands in the upper section now. I just wanted to write a DOS story and put it out into the ether.
Related to 'Shadowking', I can verify from personal experience that wardialing for local BBSes did work (*67, eventually you would turn up a BBS or connection) but BBS lists were still a million times better back during the peak period.
The lists were constantly updated, finding their way to BBSes everywhere throughout the nation. Forums were the core feature that tied everyone together, software and pics were bait. The beauty was that it was a niche, underground environment created and maintained by regular people. The first grassroots built, user-maintained, national digital communications network.
(General overview) Many BBSes - each entirely independent - linked together (for forum updates, etc) in a massive ad-hoc, volunteer‑run network relying on nothing but basic infrastructure (telephone lines). I'd go so far as to call it the most powerful people-driven communications technology yet created.
I even ran a one-line BBS for friends, with ANSI and RIP support. It felt more empowering because everything was done in-house. There was no dependence on ISPs, remote servers, domain registration, DNS or remote policies - a stack of interdependent technologies. When you were ready, plug your DOS machine into the existing infrastructure (a telephone line) and you were serving a mostly local community (long-distance calls were $$$).