At the beginning of 2026, I didn’t even know what vibe coding meant. So much has happened in the time I was busy carving wooden spoons, that I felt it simpler to just ignore the hype. One morning, in this seemingly never ending winter, a friend told me excitedly how his employer got them Claude subscriptions and encouraged them to try it for work. And his reaction was something along the lines: “it’s crazy! I love it! we’re doomed…” .inner-shadow { position: absolute; width: 100%; height: 100%; box-shadow: inset 0px 0px 10px 3px rgba(1,0,5,0.35); top: 0; left: 0; border-radius: 12px; } I felt that was an overreaction, I tried LLMs for coding before and aside from small snippets, it couldn’t do more than 50 lines without stumbling on itself. But it was a really gray hazy winter morning, and I didn’t feel like splitting wood in the quiet of the house, so I made some coffee and got my laptop out. Stages I gave it the usual task that I really want to see done but I’m always dreading to start: Stages in rcmd rcmd is a macOS app switcher that allows you to hold Right Command, and press the first letter of the app name. So press V for VSCode, G for Ghostty, C for Chrome: instant app switching without cycling through app icons. Stages was dreamed up as a way to switch workspaces with the same paradigm. So press C for Coding where VSCode and Ghostty would focus and everything else was hidden, D for Design and so on. But the functionality is way harder than it seems, the app needs to record the on-screen windows, and be able to restore them when the computer is restarted. The task was complex, just to outline some things Claude needed to do: understand the pre-existing app codebase with all its UI quirks refactor the UI to be able to present sets of apps and windows find ways to record custom window data like: the project opened in VSCode the working directory in Ghostty the current URL in Chrome files/folders/projects opened in other apps like Finder, Pixelmator, Sketch, Xcode etc. find ways to re-open windows at that specific project/file/folder figure out a way to hide individual windows on macOS It’s weird, but there’s no instant hide/show for windows on macOS. You can minimize them, but that has slow animations and unwanted side effects. I mentioned to Claude that I knew a window manager called Aerospace which had a workaround for this. I gave Claude a 200 line stages.md, let it loose on the rcmd code, and stared at it crunching stuff for a few minutes. I went to make another coffee, and when I came back it was… done? I hit Build in Xcode and was prepared to see a barrage of errors, that’s how it always was. I got ready to screenshot the errors and the stupid code diff and show my friend how “no, we’re not doomed, it just got lucky on that specific task”. No errors though, the app compiled and launched like nothing happened. Uhh.. now what? The code diff was immense, I guess I should just see if anything actually works anymore. I hid every app except VSCode and Ghostty, pressed rcmd + lalt - C and it said Stage C was assigned. No way this works… I brought back all apps on screen, quit VSCode, pressed rcmd - C and instantly VSCode and Ghostty were the only apps visible and focused, in the correct folder and project. I looked through its thinking process and it even did a web search for Aerospace macOS window manager, found its Github, reverse engineered its workings until it found the window hiding workaround I mentioned in passing and implemented it from scratch in rcmd. Astonished is a small word for what I felt. Of course, this was not a feature ready for release. There was so much code review and so much polishing work to do (not all apps have ways to restore their state after quitting, some focus flickering happened, and Stage creation should be more intuitive). But the fact that an LLM could do such complex work, changed everything. Picture-in-picture 🔗 Pipiri is a macOS app that can show any window in a floating picture-in-picture window I had this barebones app I did as a standalone Swift file, for creating a Picture-in-Picture view of a window where there’s data I’d like to keep an eye on. A Picture-in-Picture view of a Claude session floating over Xcode I mostly used it to watch long-running terminal commands while I continued coding, and for keeping Xcode logs visible while I tested parts of my app. I thought it was too niche to release, so I didn’t have the motivation to do the work on creating an UI, licensing, webpage, icon etc. I created a blank Xcode macOS app project and gave Claude minimal instructions: adapt pip.swift to a menubar app with a SwiftUI architecture add a Settings window where I can adjust the trigger hotkey, the position of the PiP panel etc. implement auto-updates with Sparkle add license verification through Gumroad link to a Contact form and a privacy policy, similar to how I do it in my other apps What can I say? It did everything. Granted, the main PiP functionality had already been written manually by me years ago, this was just doing the things that every macOS dev hates to do but has to repeat on every app. Then onto the webpage. Manually wrote a list of features with explanations in a README.md because I prefer to address future users of my apps personally, then told it to build the landing page by taking inspiration from my other app pages, using what I wrote in the README. I was more than relieved to see 95% of the webpage look exactly as I would have done it myself. Not a single word changed, only HTML structure, colors, image paths etc. Adapting old webpage structure for a new app was not something I enjoyed doing. I also got a little carried away and asked Claude to implement a VM bytecode interpreter and write the offline license validation in that bytecode. I didn’t expect it to actually do it! It even added a license generator CLI on top. I launched a v1.0 and it was received well enough. Got 2 feature requests from the first day: Multi-window mode (allow keeping multiple PiP panels visible from different) Idle/Change detection (get notified when the window contents haven’t changed in a while, or if they did change after being still for a while) These are features I would have liked myself, but was too busy to start working on them before. Now, Claude was not able to do these properly by simply mentioning the features. I had to explain the features in detail, how I want the multi-window PiPs to appear on top of each other, what APIs it should use, describe how the idle notification should look and that it’s not a system notification. In the end, the code needed a lot of review, testing and manual fixing, but just getting me started on the features was enough to motivate me to finish and release them. Getting a semi-functional feature in readable code that I can polish is much more motivating than getting a huge refactor and full implementation that doesn’t even compile. After releasing the change detection feature, I got a nice message from someone working at a hospital telling me how they use it to know when a patient has been added to the admission list without having to check it very second. This is always humbling to me when it happens, reminds me how naive and limited we are as devs thinking that we considered all use cases. When in fact we should probably just build the tools, and people will find use for them in their own way. Event-based automation 🔗 Crank is a macOS app that can run scripts and Shortcuts on event triggers My little brother was telling me about how he’s trying to find ways to make money without having to work long hours in coffeeshops and restaurants. He has already started a few ideas like websites for small local shops that want an online presence, copywriting, product photo shoots. can you believe those product shots are generated from a bland photo of a jar of honey? So I thought why not also throw in a simple macOS app. I can bring in the ideas that I’ve been sitting on for the past 5 years, help with code review and fixing, and he can just prompt Claude and test the app until it’s working as expected. The main theme was simple, low maintenance apps only. I had this idea for an event-based automation app: just the basic parts of what has already been done in Keyboard Maestro or Shortery, at a much cheaper price and free for basic needs. The UI of the Crank macOS app In support emails and chats, I’ve had a lot of people ask for some very niche functionality in my other apps. Usually, it’s some kind of automation, things like: “can I disable the MacBook screen only when I connect my Dell monitor at home?” “how can I make all screens black when I want to audition a piece in my recording studio?” “can optimised images in ~/Blog be renamed to YYYY-MM-dd-name and converted to webp?” “please add an option to toggle HDR when I open IINA or any other movie app” So I wanted to give those people the ability to resolve their frustrations without having to wait for a dev to do it. I hopped on a screen sharing session with my brother and let him ask Claude to build this app bit by bit. The first thing that jumped out is: you have to know the terminology to get quality results. He’s not a dev, he doesn’t know what a debounce or throttle is, so incidentally, he doesn’t know to ask Claude to implement it for events that arrive in bursts. He has no idea what a private API is so he can’t guide Claude to use one where we need to access low-level system APIs. And obviously concepts like dry run or event log are not familiar, but definitely useful in an app like this. Interestingly though, he was able to get Gemini to write a prompt for Claude that included a lot of these features that you’d find in an automation app. I wouldn’t have thought of that. But dev intervention was still necessary. Some added features were just dangling in the UI without any effect internally, and it was impossible for a non-dev to know what they were for. We started that way, then he continued on his own while I worked on my own apps. Even without my assistance, the result was still a very functional app that could run scripts or launch apps automatically on event triggers. And the thing is, people could have used this as is, it was useful enough to be released. But I knew it wouldn’t be received well, people expect a certain polish for apps that have paid tiers. The dev touch To be really useful, the app needed some triggers that Claude did not know to implement on its own because such code uses obscure private APIs and IOKit data. Things like: ambient light data through the MacBook’s integrated sensor (ALS.swift) if camera is turned on or off (IsCameraOn.swift) Focus mode or Do Not Disturb changes Thankfully I had already implemented those in mac-utils, but adapting them to Crank’s event trigger architecture would be quite the work. By now, I learned that this is where Claude shines: once it has working code to reference, it does a really good job at adapting that code to another project’s structure. After I pointed Claude to the files and told it to create triggers out of them, I needed less than half an hour of testing and fixing logic bugs on my part to get everything in good order. It can write quite readable code when it wants. I remembered seeing this crazy app that told you the MacBook lid angle, so I thought what the heck, someone might find a use for that. I told Claude to implement a lid angle trigger as well based on that code. I’m thinking one might want to disconnect stubborn Bluetooth peripherals when the lid is closing, and you only have a chance to do that before the lid was fully closed. Open source becomes incredibly more useful with an LLM to understand the code in seconds instead of days. It also feels like stealing, not gonna lie. If I did the research myself, it wouldn’t have felt like that, and I can’t figure out why. Cherri Some actions simply can’t be done through scripts. One example of that is enabling Do Not Disturb (or changing the Focus Mode). There’s no CLI that can do it because the system does not expose those APIs. A cherri script on the left and the compiled Shortcut on the right But it can easily be done through Shortcuts, which has a Turn Do Not Disturb On/Off action built-in. But (big but…) a named Shortcut containing that action needs to already be created by the user, we can’t call individual actions. And I can’t have Crank ask the user to create Shortcuts and drag actions around for such basic things. Shortcuts can be pre-created as file.shortcut, and installed by simply opening them, so I could provide those Shortcuts bundled with the app. But doing a rough calculation, I would have to create about 100 individual Shortcuts by hand, using mind numbingly slow drag and drop. Programmer laziness got the best of me and I went looking for a quicker way. Somehow I stumbled on this gem: Cherri - a programming language that compiles directly to a signed Shortcut I tried to get Claude to write a few basic Shortcuts with Cherri, but it clearly didn’t know the syntax. So I told Claude to just learn it from the website, and gave it the whole list of actions I need. In less than 10 minutes it: learned the Cherri programing language syntax learned to use the cherri compiler got action definitions with cherri --action wrote and compiled 98 individual Shortcuts, fully functional with inputs and outputs damn The scripts problem The problem with creating an automation app, is that you usually have to spend most of your time building a huge library of built-in actions. I didn’t want to go that route. That’s why Crank’s actions seem pretty limited: Shell script AppleScript Shortcuts Launch App Open URL Open File Notification But in my opinion, this set of actions allows you to do most common things on a Mac, if you know what to use. However, not all users of this app will know what a script is, or how to write one, so I thought I should lower the barrier of entry as much as I can. If LLMs can write such good code, why not have them write the automation scripts for us? (well, because they could delete all our files with a misplaced semicolon, that’s why…) But non-devs would try an LLM anyway, so I thought I could at least make their life easier and add some safety guards. In this test, I wanted to see if an LLM understands how to make another less capable LLM get good results. So I told Claude to create a context.md describing common CLI utilities that would help in automating a Mac, and express that as a prompt for Gemini. I added a Generate Code button in the script editor and tried to make the free Gemini tier write short useful scripts, instead of hundred-lines hallucinations on CLIs that don’t exist. It’s not perfect, but when it works, it feels magical. Generating a script to compress invoices and print them Turns out, LLMs know how to talk to other LLMs. The generated context.md tells Gemini things like: You are a macOS shell automation expert. Claude even went as far as to customize the context based on the Mac it runs on: Assume the script will run on \(ARCH) Macs with macOS \(MACOS_VERSION) Use the event variables set by Crank if relevant to the task Tool Availability on this System: basic stuff like ffmpeg, jpegoptim etc. … It’s not complex code or writing, but not very creative either. And I like that I can delegate the boring parts to a smart text generator so I can focus on the harder reverse engineering and making Macs do things they weren’t supposed to do. At least until it learns to work with Frida and a disassembler and I’m fully replaced. Consequences I started this exploration as an AI skeptic, or better said, ignorant. But things are progressing faster than I can adapt to them. So while I’m no longer a skeptic, I’ll probably stay ignorant in the long term. I see Claude Code as a valuable tool for experienced devs, just like I saw syntax highlighting, autocomplete, IDEs and other dev tools when they appeared. It enables me to spend less time in front of a screen without neglecting the people using my app, which is my end goal. But it is frightening to think; what would my mind go through if I was a junior dev in college and seeing a $20/month algorithm do what I’d need to learn in years? And now I’m thinking “I’m fine, Claude is smart but it’s still far from being able to do all the complex work I do for creating a truly useful app”. But far is relative, and such a different measure in the world of AI. I’m probably in denial, I remember laughing at LLMs hallucinating non-existent Swift syntax less than 3 months ago. To calm some of the people using my apps who might also read this: I’m not becoming a vibe coder and giving Claude free reign on Lunar and Clop. At least at the moment, the codebases of those apps are still too messy and complex for an LLM, I don’t want even the remote possibility of introducing a critical bug. rcmd still has a small codebase and doesn’t work with hardware and sensitive files, so it’s safer to try it there, but every line is still reviewed and validated by hand. I’ll probably try to get Claude to help me on things I started but never had the time to finish, like: building a custom filesystem index and fuzzy search algorithm for Cling adding natural-language calendar event scheduling in Grila fixing long-standing bugs in StartupFolder and IsThereNet But it will probably function more as a talking rubber duck, than as an employee I can trust to do my work for me.

You know what they say in the world of products: fast, cheap, good; pick two Which is very similar to my experience when trying to find the most suitable wood finish for my hand carved wooden spoons and coffee cups. hand carved wooden spoon It’s hard to find a finish that: cures fast (in less than 2 days) is food safe (can be used on things you eat or drink from, like spoons and cups) is free of solvents (doesn’t smell of nasty chemicals that make you dizzy every time you apply it) and with the hard requirement of making wood hydrophobic and leaving a visually pleasant sheen without creating a plasticky layer Natural drying oils Sure, we have pure tung oil which doesn’t need any solvent, has a pleasant nutty smell, and cures into a food-safe hydrophobic polymer inside the wood fibers. But it takes 2 to 4 weeks for the curing to happen depending on oxygen intake, and the finish is quite matte and boring. While tung oil is natural, that doesn’t mean you can ingest it. It is a powerful irritant in its liquid form, it is only food-safe after curing into its inert polymer state. There’s also the more rare polymerized (or heat-bodied) tung oil, which is a honey-like viscous oil that cures in a few days because it was heated to very high temperatures in the absence of oxygen. That would be ideal except I can’t find it anywhere around me to buy, and it would most likely need thinning with citrus oil solvent (or D-limonene). Citrus oil has a nice but way too potent smell of orange peel. It might seem pleasant at first compared to the chemical smell of naphtha and white spirits, but it can be too much if you finish small wooden objects often like I do. ash cup, in the process of being oiled with tung oil And there are plenty other drying oils, the most well-known being linseed oil, which is indeed edible but has the unfortunate disadvantages of yellowing over time, taking ages to cure and having a persistent oily-grassy flavor. Polymerized linseed oil (aka stand oil) cures faster but still retains the nasty flavor. Careful not to confuse it with boiled linseed oil or BLO, which uses metallic drying agents like cobalt or manganese salts. The boiled name comes from the old practice of heating the linseed oil with lead oxide inside, effectively shortening the oxidation process. But nowadays there’s no heat involved in BLO. More details in The lowdown on BLO The oil painting world has experience with many drying oils like safflower oil, walnut oil, poppy seed oil, hemp seed oil. Unfortunately they cure slower and into a less durable film while also being rare and expensive so they aren’t good alternatives for wooden utensils. Hardwax oils For a good few months I used Osmo Polyx Oil which is a solvent-based hardwax oil. Its safety datasheet and full declaration of ingredients mention about 50% of it being white spirit solvent, and the rest is composed of natural oils processed to become drying oils, and a small amount of hard carnauba wax. It is fast curing in less than a day, leaves a beautiful shimmer on the wood surface, it is food safe after curing, but smells awful because of the solvents. In the winter months I carve indoors and have to finish the pieces indoors as well, and the horrible solvent smell fills my house for a whole day. I seem to have become sensitized to the smell and now I have almost a full can of Osmo that I can’t use without a mask, even outdoors. Rubio Monocoat two-component hardwax oil on a half finished wooden board There’s also Rubio Monocoat and other two-component hardwax oils where the base component is usually a solvent-free blend of drying oils and waxes, and the accelerator component is Hexamethylene diisocyanate or HDI. The base component can cure on its own in about 3 weeks and the accelerator shortens the curing time to less than a day. This alternative cures fast into an inert and food safe polymer, smells like apple pie, and the finish has a slight satin effect. But it’s really hard to mix properly and apply on small wooden objects like spoons and cups. I almost always use too much accelerator, the mixture ends up too viscous and hard to spread on the wood and the constant opening of the bottles and cans makes the components dry prematurely in their containers. However, this idea of blending drying oils with hard wax seems to be promising and it’s the path I continued on developing. Notable mentions Food-safe epoxy is another alternative, but I hate both the idea of using synthetic plastics and their plasticky look takes away from the experience of using wooden utensils. Why even use wood if you’re going to cover it in a layer of clear plastic? Many people also recommend other edible oils like coconut oil, olive oil etc. But these don’t cure and don’t provide any protection for hot liquids. They either go rancid inside the wood fiber or get washed out after using the wooden object. Some recommend non-edible petrol-based mineral oil (aka liquid parrafin) because it doesn’t go rancid, but has the same effect of not actually doing much for protection and will leak into hot liquids. Beeswax is often used by kuksa makers (traditional Scandinavian wooden cups) but wax molecules are too large to penetrate wood fibers well, and it has a low melting point of 60℃ (140°F) so hotter liquids are out of the question. Carnauba wax has a higher melting point at 80℃ (176°F) and is harder than beeswax, but it has a glossy finish, having a similar look to epoxy resin after buffing. The ideal finish should combine the durable hydrophobic properties of fiber-penetrating tung oil with the fast drying properties of a top wax layer, in an easy to apply but solvent-free blend. Urushi wooden spoons finished with urushi lacquer Some carvers use urushi lacquer which is the sap from a tree common to Japan. The result is a hard and glossy finish that can resist temperatures of over 300℃ (572°F), but the look and process of application is far from what I’m looking for. To ensure curing is complete, it needs 10 to 20 very thin coats over many months in a warm and humid environment. The wood then takes this dark brown color, regardless of what wood you started with, which is the opposite of what I want. I work with fruit and flower woods that are less common in woodworking (apple, fig, jasmine etc.) and I want people to know that and experience what that specific wood looks and feels like. I want to keep their original hues and tints and accentuate the wood grain of each piece. fig wood spoon between other hand carved spoons Someone from Israel bought a spoon I made from fig tree wood, but still asked about it a bit incredulously. It had greenish light wood grain, with a bit of untouched bark kept on the handle and a wonky curve that followed the shape of the tree branch it was made from. They have many fig trees there but had never seen such a thing made from their wood, so they felt a connection to their motherland in that usual object. lilac coooking spoon, front and back My mother-in-law cut a large lilac tree and gave me some thick branches. It’s a wood that cracks very easily while drying, and I couldn’t use most of it. But the few things I made from it, I gifted back to my wife’s mother and sisters in the form of cooking spoons and are still cherished by them today. Its pastel purple grain accents remind them clearly of the tree that used to flower every spring in their yard. A brown lacquer would never be able to create this effect, and that is why I’m still trying to find the perfect natural oil based finish. Natural hardwax oil My experiments started simple, with melting about 10g of carnauba wax flakes into 40g of pure tung oil. Tung oil and carnauba wax flakes Most people recommend a double-boiler or bain-marie approach to melting the wax. That’s because the heat source would be a container of boiling water which can never get above 100℃ (212°F), a safe temperature to avoid reaching the flash point of these oils. I just drop a spoonful of carnauba flakes into a glass jar of tung oil and place it in the microwave for a minute until everything becomes liquid. For small quantities, the microwave works wonders, and there’s no danger of overheating the oil or gelifying it because it cools off rapidly. experiment #1 Ingredients: 40g pure tung oil 10g carnauba wax flakes Process: pour the tung oil into a glass jar or ceramic container drop the carnauba wax flakes into the oil and stir to disperse the flakes place in the microwave for 30 seconds then stir again for a few seconds repeat until the wax melts and everything is liquid let it cool down until the mix is solid Result: Brittle paste that’s hard to scoop from the jar and apply on the wood. Doesn’t spread easily, feels sticky. After the whole wood surface was covered and left for 10 minutes, the excess was wiped with a paper towel. After wiping, the finish is very matte, which means that not much of the wax remained, and the oil didn’t penetrate the fiber enough because of the consistency. Smells nice though. Improving the consistency The consistency changes a lot based on the oil-to-wax ratio. Too much wax and the blend cools into a brittle paste that’s very hard to scoop from the jar. Too little and the blend is too liquid with not enough wax remaining on the wood surface to form a film. A 4:1 oil-to-wax ratio is good, having enough wax for the top layer, but in this simple formulation it is still too hard and brittle. Instead of increasing the oil quantity which could defeat the purpose of the wax layer, I added a bit of beeswax to soften the blend. This works, the paste becomes softer but it makes the top layer of wax not durable enough because of its lowered melting point, so I needed an alternative. My idea is to have the wood absorb the tung oil from the blend while leaving the wax molecules on the wood surface. The wax should be soft enough to not create a glossy layer and not be brittle, but hard enough to protect the wood for the time it takes the tung oil to cure. experiment #2 Ingredients: 40g pure tung oil 7g carnauba wax flakes 2g beeswax 1g lanolin To soften the blend without having to add too much beeswax, I found lanolin to be a good addition. It is a highly hydrophobic wax extracted from sheep wool that even in tiny amounts can make the paste consistency softer and easier to apply. Process: Similar to #1, just immerse the waxes into the oil in a glass jar and microwave in increments of 30 seconds until everything is liquid. Leave to cool until solid. Result: Softer and more homogenous paste that can be scooped easily from the jar with a finger or rag. Spreads relatively easy onto the wood surface, feels like a more viscous hand cream. Smells amazing, nutty, vanilla and honey scent, faint enough to not bother anyone. After covering the whole wood surface, I used a heat gun to liquify the finish and enhance wood fiber penetration. The paste melts very easily and the oil in it gets absorbed quickly in the wood pores, leaving the wax to solidify. After the wood has cooled, I applied another layer because it seemed that some parts of the wood had little to no wax remaining. I removed the excess with a paper towel and buffed the wax to a satin look. small issues curing speed This finish is not fast curing by any means. It is fast drying, as the oil gets absorbed and the wax on top solidifies into a very thin medium-hard layer. But the melting point of the wax is about 75℃ (167°F) so still not perfect for very hot liquids. You definitely can’t brew your tea directly into a cup finished this way, not until the oil cures completely after 2-4 weeks. I tested it by brewing a cup of espresso which seems to be fine, the temperature of the coffee is far below 75℃ (167°F) when it reaches the cup. Using a spoon for eating soup also works fine, because the soup needs to be below 45℃ (113°F) for me to be able to eat it. olive wood spoon with hydrophobic layer sycamore coffee cup with hydrophobic layer After usage, the wooden spoon and cup could be easily washed with warm water and soap, the wood didn’t stain at all. The wax still worked as I could see the water droplets sliding off the wood. sheen and appearance Visually, it doesn’t leave a very strong sheen. It has a soft waxy look, instead of the shimmer of oiled wood. That’s because wax doesn’t follow the micro porous surface of the wood, it solidifies into a smooth surface. So instead of scattering light you get mirror-like reflections creating that glossy wax look. Commercial finishes use flattening agents like fumed silica to get around this, but I’m trying to use as little additives as possible for now. The look also comes from the carnauba wax being softened by the beeswax and lanolin which also translates into a softer top layer. natural resins The woodworking world has moved on almost completely from natural resins. Almost every finish is either based on polyurethane or epoxy resin if you’re looking for a harder top layer. At best, you’re looking at a shellac layer which doesn’t mix with oils. Thankfully oil painting forums and blogs still have a trove of information on tree resins, drying oils and siccatives. damar resin crystals That’s how I found out about damar resin, a semi-hard resin extracted from Dipterocarpaceae trees growing in southeast Asia. It is a clear resin, melts at about 120℃ (248°F) which is easily attainable in a microwave, and has a bit of flexibility so that it doesn’t create a cracking layer on wood. It can be found in the form of crystals or pellets in art shops and it’s pretty cheap. People dissolve it in turpentine to create varnishes, but I wanted to incorporate it into my oil-wax blend through heat instead of solvents. experiment #3 Ingredients: 40g pure tung oil 7g carnauba wax flakes 2g beeswax 1g lanolin 2g damar resin 5g coconut oil 2 drops of MnZrCa drier 2 drops of vitamin E To increase sheen and make the top layer stronger, I melted damar resin crystals directly into oil, without any solvent. I also added a very small amount of manganese, zirconium and calcium drier to make the curing faster, and some vitamin E to increase shelf life. Process: Add the carnauba wax, beeswax and lanolin into the tung oil in a glass jar and microwave in increments of 30 seconds until everything is liquid. Add the damar resin crystals into the coconut oil and microwave in increments of 10 seconds until everything is liquid. Let the mix cool down for about 30 seconds then add the drops of MnZrCa drier and vitamin E. Stir until the mix is homogenous. Pour the damar resin-oil blend into the tung oil blend and stir until homogenous. Let cool down until solid. Result: Medium-soft paste that can be scooped relatively easily from the jar with a finger or rag. Spreads easy enough onto the wood surface. Smells the same as the last experiment: nutty, vanilla and honey scent, faint enough to not bother anyone. The nasty smell of the drier doesn’t transfer into the blend because its quantity is too small. The resin has no smell in its solid form. After covering the whole wood surface, I used a heat gun to liquify the finish and enhance wood fiber penetration. The paste melts very easily and the oil in it gets absorbed quickly in the wood pores, leaving the wax and resin to solidify into a very thin layer on top. After that there was almost no excess left and I buffed the top layer to a satin look using a paper towel. Curing: The tung oil inside the wood cured in about 48 hours. This can be verified by melting the wax layer with a heat gun then removing the liquid wax with a rag. The bare wood then can be sanded with 240 grit to check if the wood dust falls off easily. If the wood dust seems wet and gums on the sandpaper, the oil is still not completely cured. Look: The satin sheen looks better because of the resin. Regions with medullary rays look wavy and move in the light, regions with straight fibers look exactly between glossy and matte. Osmo Polyx Oil still has a prettier shimmer that this finish doesn’t achieve, I’ll look into adding trace amounts of fumed silica to see if it makes the finish settle better into the wood fiber microstructure and scatter light instead of being more on the smooth glossy side. The fumed silica might also help with adding shear thinning capability so that the paste is even easier to scoop from the jar without softening it. notes on added ingredients metallic driers The metallic driers are probably the first thing that jump out, because we always read about how bad they are when seeing people talk about natural wood finishes. Not all driers are bad, and it depends on things like quantity, leaching probability and type of metal used. Sure, lead and cobalt driers are out of the question, I don’t want them anywhere near my food and beverages no matter how small the risk of leaching. They contain heavy metals that can have catastrophic effects if they accumulate in your body. I use a drier containing manganese neodecanoate to speed up surface-drying, and zirconium and calcium salts for through-drying. Zirconium and calcium are known to be safe. Zirconium is the same metal that dental implants are usually made of, even ceramic knives are made using zirconium, it seems to be a pretty inert metal when it comes to contact with our body. Manganese is toxic if inhaled or ingested in high amounts, but the quantity of manganese in 2 drops of drier is so incredibly low you could take them orally and not feel any effect other than the horrible solvent taste. That, combined with the fact that the drier disperses throughout the oil makes the quantity too low to matter. The strictest Tolerable Upper Intake Level for manganese is set to 8 mg/day. In fact we get some of that quantity daily through food that is enriched with vitamins and minerals. If the drier were to ever leach completely from the finish, one would get less than 0.1mg per lifetime of object. You’d probably have to eat the spoon completely for that to happen, in which case the quantity of ingested wood might become more of a problem. But the whole food-safe standard on drier based finishes is actually based on the fact that a cured finish traps that metal into an inert polymer which can’t leach into food. Just like Teflon pans or aluminum and copper pots, everything is safe until the layer chips off somehow. I compromised on having a quantity small enough to not pose any safety risk and that wouldn’t shorten shelf life by making the oil cure inside the jar, but high enough to make the oil cure in 48 hours instead of 2-4 weeks. This, in my opinion, makes the finish safer by being more predictable in curing time, and minimizing the probability of food contacting uncured tung oil. coconut oil and vitamin E Because damar resin melts at 120℃ (248°F), tung oil is not suitable for melting the resin directly into it because tung oil can gelify at high temperatures, hindering its curing capabilities. That’s why I melt the resin crystals in a small amount of coconut oil, which has a high enough smoking point of 180℃ (356°F), smells amazing and doesn’t go rancid easily. oil smoke point chart I added 2 drops of vitamin E (aka tocopherol) because it is a powerful antioxidant and will lower the chance of ingredients like coconut oil going rancid. It can also lower the chance of the tung oil curing inside the container by inhibiting the autoxidation reaction. This is the current formulation that I’m using for finishing my wooden spoons and cups, and which I’m selling on my online shop in Romania. experiment #4 to be continued…

When I started wood carving, the only sharpening method I remembered was from seeing my mother use some kind of smooth broken stone that she passed over the length of the knife blade before sacrificing a chicken. I also remember seeing my father use a very coarse stone wheel placed on a motor shaft which threw many sparks when he sharpened some large axe for splitting wood. I had neither of those around anymore in my rented place in the city so I jumped headfirst in the mind numbing and sometimes esoteric art of getting a sharp blade. crude sharpening methods that my parents used Carving knives The first blade type I had to sharpen was for my BeaverCraft carving knives. They fortunately came with a strop, basically a plywood base in the form of a paddle, with leather stuck to it on both sides, and a green waxy bar. Unfortunately I had no idea what to do with it. BeaverCraft carving knive set Stropping is, at the most basic level, dragging the blade back and forth on a semi-hard surface (like leather). Clean leather won’t make your blade sharper though, and that’s why you have that green bar which you need to rub onto the leather. It contains a fine abrasive that slowly removes tiny amounts of steel from the very tip of your blade to make it sharper. Reading about stropping and watching videos about it gets into a lot of debate on: how often to strop on what angle should the blade be held on the leather should you move from the heel of the blade to the tip, or the reverse can you strop too much and round the apex what is the green compound for and how should it be applied And so on. It’s hard to find much exact science on this, everyone figures it out as they go, and experts share their beliefs based on their extensive experience. I settled on stropping after every 15 minutes of constant carving, but I had no idea what I was doing or if I got any result. That’s one thing I learned that would have relieved me of much frustration at the start: find a way to test the outcome of your motions, otherwise you’ll just keep repeating things that don’t work And it also applies to programming: so many times I’ve seen colleagues writing code but with fear and uncertainty as they didn’t know that what they did would have the expected result. And only because they didn’t know how to test how the code would work. For example they were backend programmers, that only knew how to run their backend python server, but had no idea how to run the whole stack to check if the front end did work correctly with their changes. Or app devs who didn’t know how to use a local database, and instead always feared release day because of how their changes will impact the production db. Nowadays I know how to test for sharpness. If I want a basic working blade I test if the knife cuts printer paper easily and without tearing it. If I want a carving blade, I check if it pops hairs off my arm easily. And I check that often so I don’t do hundreds of blade passes for nothing. Anyway, back to my carving knives. For a long time I did a pretty bad job at keeping them sharp. I would carve for a while then feel the need to push too much into the blade or get tear out in the wood, then I would move the blade back and forth on the strop a few dozens of times, then getting back to carving I would notice a slight improvement which went away after just a few cuts, then back to stropping a thousand times for nothing. Stropping is not sharpening, and after so many cuts those blades needed serious sharpening which a piece of leather doesn’t do. Well, actually, stropping is sharpening, but it’s a very very fine kind of sharpening. It’s similar to using a super fine grit stone, something between 14000 and 100000 grit. The green paste is a super fine abrasive that does remove metal from the apex of the blade, but so little at a time that I would need to do thousands of passes to get the same result that a coarse grit sharpening stone would do in a few passes. I started looking into sharpeners, and because “stones” felt like something old which only my mother used because she didn’t have any better method, I would look into “modern” sharpeners like: pull-through V-blades honing rods powered belt grinders precision sharpening rigs rolling sharpeners modern sharpening methods I tried I tried a few of those expensive methods, ruined a few blades with them, wasted days on reading reviews and watching tutorials… none of those methods felt like a deterministic way to get a sharp blade every time. The precision sharpening rig is probably the closest to that, but I don’t want to waste precious workbench space on that. Also, I can’t take it with me when traveling. Wood planes and chisels Once I got into larger projects, I discovered how to use wood planes (which seems to be thought of as an antiquated method in Romania, manual woodworking is dead here). I also discovered they need frequent sharpening and their blade shape seemed to be perfect for using a flat surface sharpening method like, you know, a stone. tiny block plane I use for chamfering edges Every woodworker on YouTube showed how to sharpen planes on stones so I thought I should probably get one eventually. But which type? There are whetstones, natural stones, ceramic stones, oiled stones, diamond stones, you name it. After trying a ceramic stone which needed to be soaked in water for 5 minutes before doing any sharpening, and constantly wetted between every few passes, and after making a wet mess and chipping the stone and dreading to sharpen because I had to keep the stone in water, I started looking into drier methods. Diamond stones seemed to be more up my alley: a modern method based on an old, tried and true idea. Whetstones and oiled stones are great as well, I just wanted something with less maintenance and more resistant to abuse. At this point I got sick of wasting money on this hobby, so I got the cheapest set of Chinese diamond plates online. It came in a set of 4 thin steel plates with foam backing, each plate having a different grit: 400, 600, 1000 and 1200. Let’s not get into grit standards, it’s probably FEPA but who knows. And what do you know, it worked, it was easy to use, cheap and no special instructions needed. cheap diamond sharpening plates I also got lucky in finding out about the OUTDOORS55 YouTube channel and the science of sharp website who do microscope analysis of blades and methods of sharpening. They’re getting as close as possible to actually knowing what the heck happens when you move the blade on top of a stone, leather, denim or any type of material people have tried sharpening on. This helped me cut through the bullshit fast and get to a simple working method of getting any blade from rusty to shaving sharp in a few minutes. My own sharpening blocks At this point I had these diamond plates scattered on my workbench, I had a leather strop, a green stropping compound bar that painted my fingers green every time I used it, a round diamond file for my hook knives (because you can’t sharpen the inside of curved blades on a flat stone) and a piece of flexible leather for stropping those hook knives. It worked, but it was a mess. And I also work on the go a lot, I always do some wood project at my parents house, carve some branch at a walk in the woods or need to sharpen both a kitchen knife indoors and a chisel outdoors. I wanted to simplify this, but could not find any ready made product I wanted. So I got to doing what I know best, half-assing an improvised product that works for me but would be ashamed of showing it to anyone. First iteration I cut a piece of 18mm thick beech wood into the shape and size of a diamond plate, then stuck a 600 grit plate on one side, and the leather on the other side. beech block with leather strop This not only got the stone and strop higher up from the table which helped a lot with sharpening long knives, but also put them in the same package which I could carry with me. I called this a sharpening block in my mind. To keep the block from slipping while I moved the blade over it, I cut a piece from a non-slip silicone baking mat. Even the small pressure of the blade makes the silicone adhere well to both the block and whatever surface it’s sitting on. non-slip silicone baking mat piece I still didn’t know what to do if I needed a coarser/finer grit, what to do about the green waxy mess of the stropping compound, and how to carry the hook knife sharpeners. The first sharpening block, beech wood base, diamond plate, leather strop Second iteration This summer, my sister-in-law visited from Italy and asked if I could make a coffee table for her. Timing was tight because I had to make it in a week so she could transport it from Romania back to Italy by car. I had no time to find wood slabs so I got two oak wood panels from a big-box store, glued them one on top of the other to make the table-top thicker, and figured I’ll find some table legs afterwards. I found some 15cm diameter smooth beech logs at a firewood seller nearby. Turns out they were leftovers from a veneer factory that can take a really long and thin log slice automatically, leaving the log core as smooth as a turned piece. Making veneer out of tree logs The table turned out pretty nice and solid, her children love sitting on top of it or hiding under it with their toys. I guess it could also be used for holding coffee cups, they didn’t get a chance to try that yet. oak coffee table with beech wood legs So this left me with some 36mm thick oak wood that I thought could make for a better sharpening block. This time, I glued two coarse and fine grit plates back to back (either 400/1000 or 600/1200) and embedded some neodymium magnets inside the wood to keep the diamond steel plates firmly attached but able to flip easily. oak block with embedded magnets to hold the diamond steel plate For the strop, I didn’t have any leather left and it was crazy expensive, so I looked into alternatives. You can basically use any semi-hard porous surface for a strop, even plain thick cardboard. It just needs to hold the fine particles of the compound and not flex too much when pressing the blade into it. People seem to use leather, denim, cotton, felt, cardboard, balsa wood etc. After building my workbench, and the leg vise for it which needed some rubberized cork for the vise faces, I now found myself with 3m² of that rubberized cork on my hands because I could only buy it in bulk here. I tested the cork for stropping and I was amazed to see it’s even better than leather: it holds more compound which makes it last more the bars of waxy compound break down easier when dragging them across cork, almost like drawing with wax crayons. On leather I needed to heat the bars often. it doesn’t get cut as easily. I would often drag the blade wrongly across the leather, cutting into it. The cork is like a self-healing material where cuts disappear it’s cheaper it can be shaped by sanding, which allows me to give the strop rounded edges that are great for stropping recurve blades (those knives which curve to the inside and annoyingly only make contact with the edge of the strop) it doesn’t dry and curl like leather So I cut and glued a 3mm thick piece of that cork on the other side of the oak base and loaded it with pink corundum stropping compound. Yes, this is the time I discovered the even more esoteric world of stropping, lapping and honing compounds which are not green bars of waxy stuff. We’ll get to that. rubberized cork strop with pink compound The 36mm thick oak gave me enough space to drill 20mm diameter holes with a forstner bit, where I could place a wooden dowel lined with that same cork for honing the hook knives. Another hole was for the stropping compound. stropping compound and round strop embedded in the block I finally reached an all-in-one sharpening block, super stable, with coarse enough grit for getting a bad blade into shape fast, fine grit for sharpening, enough strop for a few years and a way to sharpen recurves and hook knives. Presentation of the oak sharpening block It was a bit too thick though… Stropping compounds The green stuff Most people get a green bar of compound and use it all their life and never think about it. But nooo, I had to do research and see what that compound contains and why is it so waxy and is there better stuff? My understanding is that the green stuff is made of very fine particles of chromium oxide, which is also used as green pigment in cosmetics and painting, and is why it makes everything in your life green when touched. Dialux Vert green stropping compound I had a Dialux Vert bar, and searching for its material safety data sheet surfaces the following: Composition/information on ingredients: So it is made of very fine particles of chromium and aluminum oxide suspended in paraffin wax. Does that mean it can be melted and poured into cylindrical molds? Yes it does! Adding some green flakes in a small cup with walnut oil in it allows it to be melted in a microwave oven. Any type of oil works, I tested sunflower oil, olive oil, coconut oil, mineral oil etc. you just need some kind of non-solid fat to make the paste less viscous when melted. After lining the 20mm hole with wax paper and pouring the melted mix, it solidified as a green cylinder with a chapstick-like consistency, that I could easily get out and apply to the cork. Diamonds In the last decade, diamond pastes and sprays started becoming popular for stropping. diamond powder, paste and spray Diamond powder was widely used for lapping gemstones for a long time, and diamond paste was being used in the dental industry for polishing implants. Eventually people figured they can suspend the very fine diamonds in a sprayable emulsion, advertise it for stropping and ask a ton of money for it. Diamond sprays are not accessible here in Romania, but I was able to get some lapping pastes of 3 microns and 0.25 microns particle size to test. For reference, chromium oxide green polishing compound is usually formulated with 0.5 microns particle size, but that doesn’t mean that all particles are of that size. That’s like the average particle size which gives you an idea of what kind of polish you can expect from it. I admit, I like diamonds. They’re easy to apply by squirting the paste from a syringe, I can throw it in a bag without making a green mess, it does seem to cut metal very slightly faster and lasts a bit longer between applications. But a tiny syringe that lasts 3 months costs more than a green bar that lasts me years. I can’t make this compromise. There are people that bought the diamond powder directly and made their own pastes and emulsions but it all seems too complicated for little gain. I did try some cheap Chinese AliExpress diamond pastes, but for the life of me I can’t figure out how they calculate the grit and particle size. I bought the finest I could find which is listed as W0.5 micron mesh. I don’t know what means but it doesn’t get to a shaving sharp blade easily. Cubic boron nitride Looking into other abrasives, there’s another loved stropping compound: cubic boron nitride or CBN emulsions. It’s impossible to buy that here, but it’s way too expensive anyway and I’m pretty sure the difference would be marginal. People tout that CBN and diamonds are actually necessary for harder steels, but I’ve yet to see that tested. My guess is that it’s just people’s way of justifying the purchase of a new and expensive toy, we’re all guilty of that. Wikipedia says that Cr₂O₃ has a hardness of 8 to 8.5 Mohs which is far higher than plain steel at 4 Mohs and even higher than tungsten at 7.5 Mohs. If it can scratch tungsten, I’d say it can hone a steel knife. Corundum I eventually found a pale pink compound that has the same fine grit as the green one, but without the color problem. 800g brick of pink corundum stropping compound It’s made from corundum, which is aluminum oxide without the chromium. It’s harder than the green stuff (at 9 Mohs), doesn’t leave its color everywhere, and it’s dirt cheap. I bought the Lea Chromax brand and I’m very happy with it so far. I paid €10 for a huge 800g brick that will last me a lifetime. I break small chalk-like pieces from it that I pocket or leave throughout the house to ensure I always hame some on hand. It applies easily to both cork and leather and doesn’t flake off or stick to the blade. I gifted sharpening blocks along with pieces of stropping compound to many friends and relatives and the pink Chromax block still looks as large as when I bought it. Third iteration I am working on a way to merge the all-in-one quality of the thick oak block with the pocketability of the thin beech block. I also got my hands on some narrow 20mm width diamond plates of very fine 3000grit. I find them useful for when I need a very sharp blade for doing finishing cuts on a spoon or when planing dense wood. I’d like to integrate this plate somehow in the block. fine diamond plate of 3000 grit It could maybe be glued to the side of a 20mm thick block of wood, and used hand held instead of on a table. I did try that on the beech block I already had, and it seems to work nicely. It’s not that hard to sharpen freehand and handheld as long as it’s just for doing the last fine honing on an already sharp blade. beech block with the 3000 grit diamond plate on the side The wax paper method for pouring melted compound is also not ideal. I’m thinking that pouring the compound into a chapstick tube would make it easier to use, and be thinner so it can fit inside a hole in the wood block. I also have these round diamond files which are great for sharpening round blades, serrated knifes, drill bits, even some rip cut saws. They come as a double-ended rod with a conical file on one end, and a round + flat file on the other end, and they’re meant to be placed in a pencil-like holder. round diamond file, cut in half I cut the rod in half with an angle grinder as I can’t fit the whole length of the rod inside the wood block. With only half of it, I can drill a 6mm hole in the block to embed the file. I can’t fit the cork lined wooden rod for honing though. I only have 3mm thick cork for now which is way too thick for this use case. There’s this thing called Nanocloth which is like a thin microfiber cloth specifically for CBN and diamond emulsions. It’s hella expensive and I would never buy such a thing, but it gave me an idea: I could try wrapping some thin felt cloth around a 10mm diameter wooden dowel. It should hopefully fit in the block and hold enough compound for stropping. And I guess that’s it, that would be my ideal sharpening method: a compact block of solid wood with 2 wide diamond plates on one face (coarse and fine grit) a cork strop on the other face a narrow diamond plate on the side (very fine grit) an embedded round diamond file an embedded strop rod and a stropping compound, inside the wood block somehow, that’s easy to get out and apply And all that in a cheap package that should cost less than $30 to assemble. I’m still working on it, I prepared some new 18mm oak blocks and I’ll update this post with the results.

It was the spring of 2014, over 9 years ago, just 6 months into my first year of college, when my Computer Architecture teacher stopped in the middle of an assembly exercise to tell us that Bitdefender is hiring juniors for Malware Researcher positions. I had no idea what that is, but boy, did it sound cool?… I fondly remember how at that time we weren’t chasing high salaries and filtering jobs by programming languages and frameworks. We just wanted to learn something. As students, we needed money as well of course, but when I got the job for 1750 lei (~€350), I suddenly became the richest 18 year old in my home town, so it wasn’t the top priority. And we learnt so much in 2 years.. obscure things like AOP, a lot of x86 assembly, reverse engineering techniques which dumped us head first into languages like Java, .NET, ActionScript? (malware authors were creative). But most of all, we did tons of Python scripting, and we loved every minute of it. It was my first time getting acquainted with fast tools like Sublime Text and FAR Manager. Coming from Notepad++ and Windows Explorer, I felt like a mad hacker with the world at my fingertips. I’m known as a macOS app dev nowadays, but 9 years ago, I actually started by writing petty Python scripts which spurred the obsessive love I have nowadays for clean accolade-free code and indentation based languages. What does all that have to do with static websites though? Pythonic HTML Well, 5 years ago, when I launched my first macOS app, I found myself needing to create a simple webpage to showcase the app and at the very least, provide a way to download it. And HTML I did not want to write. The XML like syntax is something I always dreaded, so overfilled with unnecessary </> symbols that make both writing and reading much more cumbersome. I wanted Python syntax for HTML so I went looking for it. I went through pug… 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 doctype html html head title Lunar - The defacto app for controlling monitor brightness meta(itemprop='description' content='...') style. a.button { background: bisque; padding: 0.5rem 1rem; color: black; border-radius: 0.5rem; } body { display: flex; flex-direction: column; align-items: center; justify-content: center; text-align: center; } body h1(style='color: white; font: bold 3rem monospace') Lunar img(src='https://files.lunar.fyi/display-page.png' style='width: 80%') a.button(href='https://files.lunar.fyi/releases/Lunar.dmg') Download pretty, but still needs () for attributes, and I still need accolades in CSS and JS then haml… 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 !!! %html{lang: "en"} %head %meta{content: "text/html; charset=UTF-8", "http-equiv" => "Content-Type"}/ %title Lunar - The defacto app for controlling monitor brightness %meta{content: "...", itemprop: "description"}/ :css a.button { background: bisque; padding: 0.5rem 1rem; color: black; border-radius: 0.5rem; } body { display: flex; flex-direction: column; align-items: center; justify-content: center; text-align: center; } %body{style: "background: #2e2431; min-height: 90vh"} %h1{style: "color: white; font: bold 3rem monospace"} Lunar %img{src: "https://files.lunar.fyi/display-page.png", style: "width: 80%"}/ %a.button{href: "https://files.lunar.fyi/releases/Lunar.dmg"} Download even more symbols: %, :, => and / for self-closing tags …and eventually stumbled upon Slim and its Python counterpart: Plim 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 doctype html html lang="en" head title Lunar - The defacto app for controlling monitor brightness meta itemprop="description" content="..." -stylus a.button background bisque padding 0.5rem 1rem color black border-radius 0.5rem body display flex flex-direction column align-items center justify-content center text-align center body style="background: #2e2431; min-height: 90vh" h1 style="color: white; font: bold 3rem monospace" Lunar img src="https://files.lunar.fyi/display-page.png" style="width: 80%" a.button href="https://files.lunar.fyi/releases/Lunar.dmg" Download ahhh.. so clean! Here’s how that example looks like if I would have to write it as HTML: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 <!DOCTYPE html> <html> <head> <title>Lunar - The defacto app for controlling monitor brightness</title> <meta itemprop="description" content="..."> <style> a.button { background: bisque; padding: 0.5rem 1rem; color: black; border-radius: 0.5rem; } body { display: flex; flex-direction: column; align-items: center; justify-content: center; text-align: center; } </style> </head> <body> <h1 style="color: white; font: bold 3rem monospace">Lunar</h1> <img src="https://files.lunar.fyi/display-page.png" style="width: 80%"> <a class="button" href="https://files.lunar.fyi/releases/Lunar.dmg">Download</a> </body> </html> not particulary hard to read, but writing would need a lot of Shift-holding and repeating tags The thing I like most about Plim, and why I stuck with it, is that it can parse my other favorite symbol-hating languages without additional configuration: Python for abstracting away repeating structures Stylus for writing style tags CoffeeScript for the script tags Markdown for long text content Here’s a more complex example to showcase the above features (might require sunglasses): example of writing a HDR page section, similar to the one on lunar.fyi 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 ---! # use Python to generate the dynamic image sizes for the srcset attr WIDTHS = [1920, 1280, 1024, 768, 640, 320] def srcset(image, ext, page_fraction=1.0): return ','.join( f'/img/{image}/{width}.{ext} {width // page_fraction:.0f}w' for width in WIDTHS ) doctype html html lang="en" head -stylus # use Stylus to do a readable media query that checks for wide color gamut @media screen and (color-gamut: p3) @supports (-webkit-backdrop-filter: brightness(1.5)) section#xdr -webkit-backdrop-filter: brightness(1) filter: brightness(1.5) body section#xdr picture source type="image/webp" srcset=${srcset('xdr', 'webp', 0.3)} source type="image/png" srcset=${srcset('xdr', 'png', 0.3)} -md # write markdown that renders as inline HTML Unlock the full brightness of your XDR display The **2021 MacBook Pro** and the **Pro Display XDR** feature an incredibly bright panel *(1600 nits!)*, but which is locked by macOS to a third of its potential *(500 nits...)*. Lunar can **remove the brightness lock** and allow you to increase the brightness past that limit. -coffee # use CoffeeScript to detect if the browser might not support HDR $ = document.querySelector safari = /^((?!chrome|android).)*safari/i.test navigator.userAgent window.onload = () -> if not safari $('#xdr')?.style.filter = "none" a[href="https://youtu.be/cVAcRH9b44w?t=55"] { display: none; } @media screen and (color-gamut: p3) { video#xdr { display: block !important; } a[href="https://youtu.be/cVAcRH9b44w?t=55"] { display: inline !important; } } And best of all, there is no crazy toolchain, bundler or dependency hell involved. No project structure needed, no configuration files. I can just write a contact.plim file, compile it with plimc to a readable contact.html and have a /contact page ready! So that’s how it went with my app: I wrote a simple index.plim, dropped it on Netlify and went on with my day. Complexity Cost 1 pip install for getting the Plim CLI 1 npm install for getting stylus and coffeescript (optional) 1 build command for generating the HTML files Complex simplicity The app managed to get quite a bit of attention, and while I kept developing it, for the next 4 years the website remained the same heading - image - download button single page. It was only a side project after all. Working for US companies from Romania made good money, but it was so tiring to get through 3h of video meetings daily, standups, syntax nitpicking in PR review, SCRUM bullshit, JIRA, task writing, task assigning, estimating task time in T-shirt sizes?? In April 2021 I finally got tired of writing useless code and selling my time like it was some grain silo I could always fill back up with even more work… I bet on developing my app further. Since my college days, I always chose the work that helps me learn new concepts. At some point I had to understand that I learnt enough and had to start sharing. This time I really wanted to write software that helped people, and was willing to spend my savings on it. Comically Stuffed Stylesheets A more complete app also required a more complete presentation website, but the styling was getting out of hand. You would think that with flexbox and grids, you can just write vanilla CSS these days, but just adding a bit of variation requires constant jumping between the CSS and HTML files. A presentation page is usually only 10% HTML markup. The rest is a ton of styling and copy text, so I wanted to optimize my dev experience for that. There’s no “go to definition” on HTML .classes or #ids because their styles can be defined ✨anywhere✨. So you have to Cmd-F like a madman and be very rigorous on your CSS structure. The controversial but very clever solution to this was Tailwind CSS: a large collection of short predefined classes that mostly style just the property they hint at. For example in the first code block I had to write a non-reusable 5-line style to center the body contents. 1 2 3 4 5 6 7 body { display: flex; flex-direction: column; align-items: center; justify-content: center; text-align: center; } With Tailwind, I would have written the body tag like so: 1 body.flex.flex-col.justify-center.items-center.text-center That might not seem like much, some would argue that it’s even a lot less readable than the CSS one. Can’t I just define a .center class that I can reuse? Well, think about a few things: this might repeat on many sections of the page, but with slight variations (what if I want a centered row, or longer paragraphs of text aligned to the left) responsive sections might need to alter layout (e.g. vertical on mobile, horizontal on desktop) and media queries will quickly blow up the style size .md:flex-row.flex-col is what you would write in Tailwind adding dark/light mode support is yet another media query .dark:bg-white.bg-black looks simple enough interactions like hover effects, complex shadows and filters like blur and brightness is syntax that’s often forgotten .shadow.hover:shadow-xl creates a lift off the page effect on hover by making the shadow larger .blur.active:blur-none un-blurs an element when you click on it choosing colors and reusing them needs a lot of attention .bg-red-500.text-white sets white text on saturated red red-100 is less saturated, towards white red-900 is darker, towards black li:nth-child(4) > ul > li > code { transition: box-shadow 0.55s ease-in-out, filter 0.55s ease-in-out; } #tailwind-usefulness + ul > li:nth-child(4) > ul > li > code:hover { transition: box-shadow 0.25s ease-in-out, filter 0.15s ease-out; } #tailwind-usefulness + ul > li:nth-child(3) > ul > li:nth-of-type(1) > code { background: black !important; color: rgb(255, 200, 149) !important; } @media (prefers-color-scheme: dark) { #tailwind-usefulness + ul > li:nth-child(3) > ul > li:nth-of-type(1) > code { background: white !important; color: hsl(19, 72%, 32%) !important; } } #tailwind-usefulness + ul > li:nth-child(4) > ul > li:nth-of-type(1) > code { box-shadow: 0 1px 3px 0 rgb(0 0 0 / 0.1), 0 1px 2px -1px rgb(0 0 0 / 0.1); } #tailwind-usefulness + ul > li:nth-child(4) > ul > li:nth-of-type(1) > code:hover { box-shadow: 0 20px 25px -5px rgb(0 0 0 / 0.1), 0 8px 10px -6px rgb(0 0 0 / 0.1); } @media (prefers-color-scheme: dark) { #tailwind-usefulness + ul > li:nth-child(4) > ul > li:nth-of-type(1) > code:hover { box-shadow: 0 20px 25px -5px rgb(0 0 0 / 0.5), 0 8px 10px -6px rgb(0 0 0 / 0.7); } } #tailwind-usefulness + ul > li:nth-child(4) > ul > li:nth-of-type(2) > code { filter: blur(4px); } #tailwind-usefulness + ul > li:nth-child(4) > ul > li:nth-of-type(2) > code:active { filter: blur(0) !important; } #tailwind-usefulness + ul > li:nth-child(5) > ul > li:nth-of-type(1) > code { background: #ef4444 !important; color: white !important; } #tailwind-usefulness + ul > li:nth-child(5) > ul > li:nth-of-type(2) > code { background: #fee2e2 !important; color: black !important; } #tailwind-usefulness + ul > li:nth-child(5) > ul > li:nth-of-type(3) > code { background: #7f1d1d !important; color: white !important; } Sure, long lines of classes might not be so readable, but neither are long files of CSS styling. At least the Tailwind classes are right there at your fingertips, and you can replace a -lg with a -xl to quickly fine tune your style. Complexity Cost: 1 command added for building the minimal CSS from the classes used 1 npm install for getting the Tailwind CLI 1 config file for defining custom colors, animations etc. (optional) Responsive images So many people obsess over the size of their JS or CSS, but fail to realize that the bulk of their page is unnecessarily large and not well compressed images. Of course, I was one of those people. For years, my app’s website had a screenshot of its window as an uncompressed PNG, loading slowly from top to bottom and chugging the user’s bandwidth. I had no idea, but screenshots and screen recordings are most of the time up to 10x larger than their visually indistinguishable compressed counterparts. I even wrote an app to fix that since I’m constantly sending screenshots to people and was tired of waiting for 5MB images to upload in rapid chats. It’s called Clop if you want to check it out. Yes, just like that famous ransomware, it wasn’t that famous at the time of naming the app. I needed a lot more images to showcase the features of an app controlling monitor brightness and colors, so I had to improve on this. Delivering the smallest image necessary to the user is quite a complex endeavour: Optimize the image using ImageOptim Resize it to fit multiple screen sizes using vipsthumbnail Figure out what fraction of the page width will be occupied by the image Write a suitable srcset attribute to load the suitable image Optional: convert the image to formats like webp, avif or JPEG XL for smallest file size I did so much of that work manually in the past… thankfully nowadays I have imgproxy to do the encoding, optimization and resizing for me. I just have to write the srcset, for which I defined Plim and Python functions to do the string wrangling for me. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 -def image(img, ext='png', factor=0.4, mobile_factor=1) picture -call img=${img} ext=${ext} factor=${factor} mobile_factor=${mobile_factor} self:sources img srcset=${srcset(img, ext, factor)} -def sources(img, ext='png', factor=0.4, mobile_factor=1) source type="image/avif" srcset=${srcset(img, ext, mobile_factor, convert_to="avif")} media="(max-width: 767px)" source type="image/avif" srcset=${srcset(img, ext, factor, convert_to="avif")} media="(min-width: 768px)" source type="image/webp" srcset=${srcset(img, ext, mobile_factor, convert_to="webp")} media="(max-width: 767px)" source type="image/webp" srcset=${srcset(img, ext, factor, convert_to="webp")} media="(min-width: 768px)" source type="image/${ext}" srcset=${srcset(img, ext, mobile_factor)} media="(max-width: 767px)" source type="image/${ext}" srcset=${srcset(img, ext, factor)} media="(min-width: 768px)" 1 2 3 4 5 6 7 8 9 10 11 12 WIDTHS = [1920, 1280, 1024, 768, 640, 320] def imgurl(image, width, ext="png", convert_to=""): conversion = f"@{convert_to}" if convert_to else "" return f"https://img.panaitiu.com/_/{width}/plain/https://lunar.fyi/img/{urllib.parse.quote(image)}.{ext}{conversion}" def srcset(image, ext="png", factor=1.0, convert_to=""): return ",".join( f"{imgurl(image, width, ext, convert_to)} {width // factor:.0f}w" for width in WIDTHS ) Complexity Cost 1 imgproxy server that needs to run somewhere publicly available, be kept alive and secure some Python and Plim code for generating srcsets Hot reloading After 2 weeks of editing the page, Cmd-Tab to the browser, Cmd-R to refresh, I got really tired of this routine. I worked with Next.js before on Noiseblend and loved how each file change automatically gets refreshed in the browser. Instantly and in-place as well, not a full page refresh. I got the same experience when I worked with React Native. There should be something for static pages too, I thought. Well it turns out there is, it’s called LiveReload and I had to slap my forehead for not searching for it sooner. After installing the browser extension, and running the livereloadx --static file watcher, I got my hot reloading dev experience back. Actually now that I think about it, Hugo has super fast hot reloading, how does it accomplish that? Yep, turns out Hugo uses LiveReload as well. Complexity Cost 1 more command to run in 1 more terminal panel, multiplex helps with that 1 browser extension to install and hope it’s not compromised or sold to a data thief 1 npm install for getting the livereloadx CLI Contact pages After releasing the new app version, many things were broken, expectedly. People tried to reach me in so many ways: Github issues, personal email, through the app licensing provider, even Facebook Messenger. I had no idea that including an official way of contact would be so vital. And I had no idea how to even do it. A contact form needs, like, a server to POST to, right? And that server needs to notify me in some way, and then I have to respond to the user in some other way… sigh I thought about those chat bubbles that a lot of sites have, but I used them on Noiseblend and did not like the experience. Plus I dislike seeing them myself, they’re an eyesore and a nuisance obscuring page content and possibly violating privacy. After long searches, not sure why it took so long, I stumbled upon Formspark: a service that gives you a link to POST your form to, and they send you an email with the form contents. The email will contain the user email in ReplyTo so I can just reply normally from my mail client. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 form action="https://submit-form.com/some-random-id" label for="name" Name input#from hidden="true" name="_email.from" type="text" input#name name="name" placeholder="John Doe" required="" type="text" label for="email" Email input#email name="email" placeholder="[email protected]" required="" type="email" label for="subject" Subject input#email-subject hidden="true" name="_email.subject" type="text" input#subject name="subject" placeholder="What's this message about?" required="" type="text" label for="message" Message textarea#message name="message" placeholder="Something about our apps perhaps" required="" type="text" rows="6" -coffee # Custom subject and name: https://documentation.formspark.io/customization/notification-email.html#subject nameInput = document.getElementById("name") fromInput = document.getElementById("from") nameInput.addEventListener 'input', (event) -> fromInput.value = event.target.value subjectInput = document.getElementById("subject") emailSubjectInput = document.getElementById("email-subject") subjectInput.addEventListener 'input', (event) -> emailSubjectInput.value = event.target.value Complexity Cost None, I guess. I just hope that the prolific but unique Formspark dev doesn’t die or get kidnapped or something. And you call this “simple”? It’s not. Really. It’s crazy what I had to go through to get to a productive setup that fits my needs. One could say I could have spent all that time on writing vanilla HTML, CSS and JS and I would have had the same result in the same amount of time. I agree, if time would be all that mattered. But for some people (like me), feeling productive, seeing how easy it is to test my ideas and how code seems to flow from my fingertips at the speed of thought, is what decides if I’ll ever finish and publish something, or if I l’ll lose my patience and fallback to comfort zones. Having to write the same boilerplate code over and over again, constant context switching between files, jumping back into a project after a few days and not knowing where everything was in those thousand-lines files.. these are all detractors that will eventually make me say ”f••k this! at least my day job brings money”. Reusability So many JS frameworks were created in the name of reusable components, but they all failed for me. I mean sure, I can “npm install” a React calendar, and I am now “reusing” and not “reimplementing” the hard work of someone better than me at calendar UIs. But just try to stray away a little from the happy path that the component creator envisioned, and you will find it is mind-bendingly hard to bend the component to your specific needs. You might raise a Github issue and the creator will add a few params so you can customize that specific thing, but so will others with different and maybe clashing needs. Soon enough, that component is declared unwieldy and too complex to use, the dev will say “f••k this! I’d rather do furniture” and someone else will come out and say: here’s the next best thing in React calendar libraries, so much simpler to use than those behemoths! I never had this goal in mind but unexpectedly, the above setup is generic enough that I was able to extract it into a set of files for starting a new website. I can now duplicate that folder and start changing site-specific bits to get a new website. Here are the websites I’ve done using this method: lowtechguys.com / where I publish my apps subsol.one / app I made for my brother to publish parties on robert.panaitiu.com / my brother’s personal website lunar.fyi / the app I have been talking about above And the best thing I remember is that for each website I published a working version, good looking enough, with a contact page and small bandwidth requirements, in less than a day. How does this solve the problem of straying away from the happy path? Well, this is not an immutable library residing in node_modules, or a JS script on a CDN. It is a set of files I can modify to the site’s needs. There is no high wall to jump (having to fork a library, figuring out its unique build system etc.) or need to stick to a specific structure. Once the folder is duplicated, it has its own life. For those interested, here is the repo containing the current state of my templates: github.com/alin23/plim-website I don’t recommend using it, it’s possible that I’m the only one who finds it simple because I know what went into it. But if you do, I’d love to hear your thoughts. Gatsby? Jekyll? Hugo? Weirdly, this website I’m writing on is not made with Plim. At some point I decided to start a personal website, and I thought it probably needs a blog-aware site builder. At the time, I didn’t know that RSS is an easily templatable XML file, and that all I need for a blog is to write Markdown. I remember trying Gatsby and not liking the JS ecosystem around it. Jekyll was my second choice with Github Pages, but I think I fumbled too much with ruby and bundle to get it working and lost patience. Both problems stemmed from my lack of familiarity with their ecosystems, but my goal was to write a blog, not learn Ruby and JS. Hugo seemed much simpler, and it was also written in Go and distributed as a standalone binary, which I always like for my tools. I marveled at Hugo’s speed, loved the fact that it supports theming (although it’s not as simple as it sounds) and that it has a lot of useful stuff built-in like syntax highlighting, image processing, RSS generator etc. But it took me sooo long to understand its structure. There are many foreign words (to me) in Hugo: archetypes, taxonomies, shortcodes, partials, layouts, categories, series. Unfortunately, by the time I realized that I don’t need the flexibility that this structure provides, I had already finished this website and written my first article. I also used a theme that uses the Tachyons CSS framework, for which I can never remember the right class to use. I thought about rewriting the website in Plim but converting everything to Tailwind or simple CSS would have been a lot of work. I eventually started writing simple Markdown files for my notes, and have Caddy convert and serve them on the fly. Helps me write from my phone and not have to deal with Git and Hugo. I still keep this for longform content, where a laptop is usually needed anyway.