Nostalgic Business: Selling Websites on CDs in the Early 2000s | August 18 2026, 16:31

In the early 2000s, I had a “business” selling the internet at retail – on CDs. I would make agreements with content providers – websites like algolist, openbsd, and cooking recipes to place a “buy this site on CD” banner on every page, and in return I invested time in creating a static version with search functionality and recording it onto discs. Links from the pages led to .nadiske.ru, where there was a large order button and upsells from other projects, related and not. A percentage from each disc went to the content owner. Delivery was done by cash on delivery through the Russian Post (meaning, I sent the discs without prior payment), and I collected the money at the post office. The site effectively had two pages – a product list with checkboxes and an order form. However, there were as many sub-sites as there were partners – on each of them, the main product was listed first, and the rest were upsells. The idea was that since we’re already ordering, let’s get something else too”.

So, several interesting problems arose back then. How to match the money received with the orders posted? For each order, I dynamically formed a price – adding pennies so the price was unique. For the end customer, it made no difference whether they agreed to a price of 150 rubles or 150 rubles and 12 kopecks, but for me, it was convenient for matching. Together with the date, it was quite reliable.

The second problem was where to manage orders with statuses and the like. In the end, I chose Microsoft Access. It pulled XML with new orders from the website. I developed the whole system on Access in a couple of hours. It was ideal. By the way, has Access died out yet?

Interestingly, the top sellers were discs that I had produced just to expand the selection – Unixes and Linux. Understandably, there weren’t any links from partner websites to these, but they were bought as add-ons, and ultimately they topped the sales and were the easiest and cheapest to produce.

Also, I developed a search engine for the CD to search through this content. I don’t remember how, but back then it was possible to embed executable code in the browser that opened index.html from the disc. Either there were ASP inserts and DLLs, or something of that nature.

I closed the project when it became clear that decent internet would eventually reach the regions, and the demand for discs would someday wane. Plus, everything was heading in that direction. Additionally, the disks required updates, and it was a pity to invest more effort and time, especially considering the first point.

Challenging the Limits of AI with ARC-AGI-3 Tests | August 17 2026, 04:51

Today I came across ARC-AGI-3 tests for AI models. That is, if you think that AI can do anything for you, consider that you have already earned a large part of the $850,000 prize fund – just submit your beautiful code, and let it be the best. But it doesn’t work that way. Humans handle 100% of tasks, while programs have not yet surpassed the 2.5-3% mark. What are those tasks?

The program is given not a textual task but a mini-game. There are no instructions or descriptions of the rules. At each step, the program receives a “frame” — a JSON object with the current state of the field. It is a grid of up to 64×64 cells, where the cells are encoded with numbers from 0 to 15 (representing different colors or types of blocks). The agent has a standardized interface (actions 1–7), including basic steps and the ability to specify specific coordinates (X, Y). But AI does not know beforehand what these actions do! In one task “Action 1” might shift a block to the right, in another repaint it, in a third turn on gravity. To understand the logic, AI must actively “poke with a stick” at the environment. It performs an experimental action, the environment reacts and returns a new frame. The agent must analyze the changes, update its theory about the physics of this particular world, and make the next deliberate step. The game can end with the status WIN, GAME_OVER, or continue further.

Again – humans score 100%, machines – 2-3%. You can try it at the ARC PRIZE website, there are sample tasks available.

Exploring Unpublished Social Sharing Previews: Solutions and Simulations | August 14 2026, 14:49

It’s interesting that there seem to be no tools available to check how Social Sharing Previews for Facebook, LinkedIn, Twitter work for a site that is not yet public. I can think of two solutions for this: simulating social network behavior and creating a page with stripped content but correct headers, which is opened to the public for a minute, then preview generation is performed through the actual tools of FB, LI, Twitter, conclusions are drawn on what to adjust, and the page disappears. The simulation could also be done, seems not too complicated.

Automating Disk Analysis with AI: Towards Smarter File Management | July 23 2026, 12:24

I contemplate creating a small program that would scan the disk and analyze all folders and files, peering inside the files only if necessary – primarily focusing on their names, and using LLM to tag and briefly describe entire hierarchies, and more importantly, classify, and save the size of files in the hierarchy. It’s assumed that such a tool runs in the evening and the results are discovered in the morning, because naturally all this is slow. But in the morning, I receive a detailed, structured, relatively small comprehensive report on what needs to be backed up, what can probably be deleted (for instance, a clone of some repository from the public domain), sorted by impact (freeing up space, less clutter, etc.).

Of course, throwing tens of thousands of files at an LLM – is an unreasonable waste of tokens. An agency scheme is assumed – the system takes the top level, makes assumptions, checks them, looks at the time files were modified, checks if there have been changes since the creation of the entire hierarchy and asks the LLM very compactly, setting a percentage of confidence in the judgement about the necessity/unnecessity of any given folder. Then the user can manually check something, or ask to take a closer look at something else.

I wonder, would such a thing be useful to anyone else?

Yuka App Reveals Surprising Health Scores: The Uncommon Bread That Scores Perfectly | July 14 2026, 19:18

There’s an app called Yuka, and it shows how healthy food is based on its barcode from 0 to 100. You’ll never guess what scored a 100 out of 100 from the photo. It’s not berries and not salmon. It is – bread.

Ezekiel 4:9. It’s truly not ordinary bread. There’s no flour in it. It’s made entirely of sprouted grains, seeds (like flax, sunflower, and chia) and psyllium as a binding agent. Sprouted Grain Bread. Well, in terms of texture, it’s bread as usual. Speaking of which, the regular white and yellow sandwich bread from the same store scored a 4 out of 100.

How Yuka evaluates products. Three criteria – nutritional value according to the Nutri-score system (60%), presence of additives (30%), organic status (10%). The algorithm adores products high in fiber and protein but severely penalizes for caloric content, saturated fats, sugar, and salt. Salmon, due to fats and high caloric content, scores 78/100. Meanwhile, ready-to-eat French fries score 66/100, and BBQ chips – 4/100.

Parking Woes at Cocoa Beach: A Costly Oversight | July 12 2026, 13:09

Florida, Cocoa Beach. Parked the car not far from a roadside restaurant. The sign says paid parking, here’s the parking number, here’s a QR code to download the app. Took photos of the QR code, my car’s number, and the parking spot number and thought as I walked the hundred meters to the restaurant, I’d install the app and pay then. Installed it, entered the parking number, and now it’s asking for the parking spot number! Darn, have to go back. And there I was, already greeted by a parking ticket… Talk about efficiency

Mind-Blowing Facts About SQLite: From Naval Beginnings to Mars | June 18 2026, 12:48

Today I learned some mind-blowing and brain-blasting facts about SQLite — the most widely used database in the world (A trillion installations. In every smartphone, browser, vehicle, A350 aircraft, even on Mars). So, it was born on the military destroyer USS Oscar Austin. It’s developed by JUST THREE people. Open source. But. You can’t just walk into this open source – it’s invitation-only and through an affidavit. The company is called Hwaci (“Hipp, Wyrick & Company”). Also involved in music (founder’s wife is a musician). Check out the website. Office — in a residential house in Charlotte. 600+ lines of tests for every line of code. 100% branch coverage and MC/DC. That is, they simulate OS crashes, power outages, I/O errors, and memory shortages. The main test suite is proprietary and closed. Imagine that, open source with paid private tests. Want access — join the consortium for $120,000 a year.

And the strangest thing — the spirit of the project is almost monastic. Instead of a Code of Conduct, they have a Code of Ethics, taken from chapter 4 of the Rule of Saint Benedict (literally 1500-year-old “tools for good deeds”). At the beginning of each source file instead of a legal notice — a blessing: “May you do good and not evil…”.

(They have not yet found a suitable version control system and wrote their own for themselves — Fossil (based on SQLite, of course). And their parser-generator Lemon is also homegrown. Just like Linus with Git.)

Innovative DIY Program for Live Transcription and Screen Capture Analysis | June 18 2026, 04:47

I made a really cool thing for myself. I launch a program, it turns on the microphone and listens. I switch to, say, a browser, comment on what I see on the screen, periodically pressing a hotkey to take a screenshot. Meanwhile, my program makes a time-stamped transcript of my comments, saves the screenshots with time stamps, then it recognizes the screenshots, extracting therefrom the spellings of various words, brands, identifiers, people’s names, so as to then transform the transcript of my speech into correct text. And all this – local models, running on my laptop, which means, absolutely free.

After I finish talking to the computer, I start processing the transcript, which takes the raw transcript and text-recognized screenshots as input and outputs a processed transcript, which now looks presentable (Gemini API is used here). One could even go a step further and automatically cut out fragments from the screenshots that were discussed, and insert them in the text exactly where they were mentioned.

Or here’s another thing I can do: just turn on a video on the speakers and the program immediately makes such a transcript for me. Google on YouTube the video “Angular HttpClient Under The Hood. Design Patterns & Source Code Overview” starting at 3:51 – I just put it on autopilot for a couple of minutes, then stopped my script.

Transforming Image Proportions with Generative AI: Smart Redesign Solutions | June 16 2026, 10:08

I published an article about how to transform images with changing proportions. Using generative AI, of course, because transforming a square into a rectangle can either result in data loss, their extrapolation, or by stretching and compressing the image itself. Here, I describe a method where smart extrapolation is performed. When processing hundreds and thousands of images, this approach is not without errors, but their number is relatively small, and it turns out to be much more advantageous to focus on manually correcting the erroneous ones than to do all the work manually right away.

This is specifically necessary during a redesign, when it turns out that the new design slightly mismatches the old one in size, for instance with banners, and the number of these banners is measured in hundreds and thousands.

Automating Banner Crop/Resize Across Breakpoints with Generative AI

Migrating SAP Commerce Content with Graph Databases: A Neo4j and Memgraph Guide | June 10 2026, 03:12

Published a new article on Hybrismart after a long hiatus. It’s about how to migrate data from an old site to a new one using a graph db (specifically, I used neo4j and memgraph). The case is as follows: there is an old site and a new site, and you need to transfer CMS data – components, pages, layout from the old to the new, while along the way making various transformations – for example, in the new site the styles are different, the layout is different, some of the components are different. For this task, I used a graph db.

It’s been a while since I wrote on my blog about SAP Commerce Cloud. I worked at SAP for two years, and thought it inappropriate to write about their products while formally having access to internal documents. Currently, I am working on two projects simultaneously – one about migrating SAP Commerce Cloud, and the other significantly about graph databases. At the junction of these worlds, the article was born.

https://hybrismart.com/2026/06/10/migrating-sap-commerce-content-with-a-graph-database/

Migrating SAP Commerce Content with a Graph Database