A protocol is simply a set of standards that two parties agree to. They can be between people (like Alice and Bob using an agreed cipher to encrypt mail) or involve computers (in which case you have the HTTP for delivering websites, IP for connecting devices to the internet, IMAP for email services, etc.). You might have also heard of spies, government officials, etc. in movies and TV shows talk about “following protocol” or “breaking protocol”. Technically, you could create your own protocol for anything, but you would first need to find another party to agree to it (whether that’s additional devices or people you know).

One protocol you are using right now is ActivityPub! This “federates” the various different Lemmy, Mastodon, Piefed, etc. instances to form an interconnected social network that is greater than the sum of its parts!

The internet is full of diverse opinions, and you tend to go towards communities that fit your morals, ideas, etc. On big tech social media platforms, this is done through algorithms that track your activity, while on the Fediverse, content is more curated (so still a sort of bubble, but one that you choose for yourself I guess). Additionally, negative opinions are more likely to gain traction online, whether they are valid or not. People are more likely to share polarising views online rather than those experienced by everyone.

You will also note that in the real world, there is fewer “tech-savvy” people. Most people don’t know about the unethical nature of LLMs and how they are problematic to copyright. The majority of people don’t have a strongly negative opinion on LLMs nor do they usually have a strong positive opinion on it. They see it as something that can make paragraphs using only a few words and “art” with a single prompt. They think it’s neat, but certainly not the saviour to humanity many business execs think it is.

But I would like to highlight that, in real life, there are a lot of people who hate LLMs, text-to-image models, etc. Most artists don’t like that these models are being trained on their creations without their consent or any sort of compensation. Doctors and others in the medical industry don’t like that people are turning to LLMs for health advice (that is usually wrong and/or harmful), particularly when it concerns mental health. Software maintainers hate that “vibe coders” are submitting unreviewed LLM-generated code, taking up the time that could be better spent fixing bugs or developing new features. And I’m sure that there are businesspeople who are worried that the “AI bubble” will pop any moment once investors realise that they are losing money, crashing the economy and bankrupting a lot of people.

It’s not that everybody online hates LLMs and, in the real world, people feel the opposite. That wouldn’t really make sense. It’s more that polarising opinions are amplified on the Internet. See the “AI bros” online for some strongly positive opinions, they are quite ridiculous I think, they somehow treat LLMs to a higher degree than most business execs. And on the Internet, there is a higher proportion of tech-savvy individuals who know the ethical, legal, and moral risks of LLMs.

I highlight that they are LLMs, not AI. To say that they are “AI” would mean that they have intelligence. In my opinion, since these models do not actually understand the prompt given. I believe that intelligence requires being able to understand a problem and figure out a solution. There are plenty of intelligent beings on this planet, us humans, corvids, octopi, certain species of whale, a lot of the primates, ant colonies, etc. But LLMs are not one of them.

Aside from a lot of the ones that are abbreviations (like GNU Image Manipulation Program, or GIMP), most of them are fine I think.

Immich, it sounds like “image”, which makes sense for photo hosting. Inkscape is a landscape of ink, suitable for a vector graphics editor. “Chrono”, the clock app on Android, is named after the embodiment of time. Radicale, the CalDAV self-hosted service, is the word “radical” conjoined with “calendar”. KeePass is a password manager, a master key is used to unlock the vault. KDE likes to put “K” in front of a lot of their app names. KCalc, KGet, Konsole, KOrganizer, KAlarm, KWrite. Their functions are pretty self explanatory. Okular is a PDF reader by KDE, and the name is a play on the word “ocular”, used to describe vision, but with a “K”! MarkText lets you write text in Markdown format. LibreOffice is a free (as in freedom, or libre) open-source office suite. Writer, Impress, and Calc are related to documents, presentations, and spreadsheets. And then there are all the apps that are not unique and are simply what they are. Think “Offline Translator”, “OSS Document Scanner”, etc. (very common with a lot of Android apps I use)

I would imagine Kittygram refers to the vast quantities of cat photos on Instagram.

Not really an insult, but I’ve heard the phrase “how the turns have tabled” as a version of “how the tables have turned”.

The cables under the sea are used to connect the networks across seas and oceans, allowing people on both sides to communicate with one another. A normal wireless connection can’t work at these distances without the use of a very large (and expensive) receiver, so undersea cables are used to do so. Disruptions of these cables (by both natural and human actors) have caused concern for a lot of people as they are critical for communication across the world.

In a nutshell, the “Internet” is actually a mix of fibre optic cables, copper wires, wireless networks, and satellites that all intercommunicate with each other using the Internet Protocol (IP). These different parts are used for different purposes.

Wireless is used for mobile devices, home networks, etc., while wired connections are common when speed is of utmost importance, so things like enterprise networks, servers, that sort of thing uses wires and cables. Fibre optic cables uses light to transmit data while copper wires use the flow of electrons (electricity) to do so, the former is faster but more expensive while the latter is more affordable. Finally, satellites are used when neither is available, and typically used when you are somewhere remote, or when normal connection methods are disrupted. In Ukraine, soldiers use satellites to communicate due to the disruption of mobile networks from the war.

A lot of people have already mentioned that they are too large and expensive for consumer devices. They also need to be cooled to temperatures very close to absolute zero (using liquid helium) for the superconductors, which adds additional bulk and power draw.

These quantum computers would also require a different architecture that isn’t supported by the majority of software and OSs, since they work very differently to standard non-quantum chips. Think how RISC-V isn’t very well supported too well right now, and it’s much worse than that since the logic gates needed for computation, storage, etc. need to be rewritten entirely due to the use of qubits over bits! Aside for some very specific research bits, there is nearly nothing that can currently run on a quantum computer.

As of right now, even the largest quantum computers only has a few hundreds, maybe a few thousand, qubits, which isn’t enough processing power for running a regular desktop OS and consumer applications, even if everything was magically ported to quantum land. Not only that, major hurdle right now is reducing environmental interference from affecting the qubits, meaning there are currently high error rates, i.e. the qubits do not behave.

My point is that, in the modern day with these battery technologies, the difference in battery capacity between two nearly identical devices, one with a replaceable battery and the other with one glued in, is now negligible.

It’s a bit like the push for eSIMs, where Apple recently released phones with slightly more mAH in their eSIM only phones for the U.S. market. Many people argued that the removal of a physical SIM slot (and usually the microSD slot we all love) is needed for larger battery capacities in smartphones, but in reality, the actual difference in the iPhones was very small.

And the existence of Framework, Fairphone, Lenovo, etc. who produce devices with easily replaceable batteries show that it is possible to do so without sacrificing battery life compared to the competition. A lot of the Lenovo Thinkpads have this feature, and compared to laptops with a similar size, weight, and processing power, there is no significant difference in battery life. Framework’s “13 Pro” (note that the screen size is around 13.5") can compete with the 14" MacBook Pro in battery life in a similar form factor, while also offering a replaceable battery and a repairable design. And for smartphones, Fairphone’s latest model can compete in battery life with other phones in the same price range, even with a replaceable battery.

The difference in the battery life of easily replaceable and glued-in batteries is very small, for both large corporations like Lenovo and relatively small companies like Framework or Fairphone. I believe this can be compared to the removal of the physical SIM slot (and the microSD slot), where the actual gain in battery capacity is negligible. In this case, we have a real “identical” product with and without the slot, and the battery capacity difference is very small.

A lot of others have already mentioned the use of trigonometry. I would like to add that, often times, the position of the Sun, stars, etc. were used for navigation. I would imagine their positions would be useful to judge large distances. And many times, map makers asked those living in the area about regions they themselves have not explored in great detail. Lots of maps are also an amalgamation of many other maps, so although it is unlikely one person explored vast swaths of forest by themselves, it is more likely that different parts of it came from different people, including locals.

p.s. Eratosthenes, an ancient Greek polymath from 200 BCE ish (!!), measured the circumference of the Earth to a remarkable accuracy simply by using shadows of a vertical rod cast by the Sun. It’s kind of insane how people were able to do all this without the assistance of modern technology.

Counterpoint, battery chemistries have improved a lot over the years. Even discounting SiC batteries, the density of batteries has improved significantly. Most smartphones will last the day, and newer models (even without SiC!) can last multiple! Laptops now have much larger batteries than before, and their processors are less power-hungry. In the modern day, you can have replaceable batteries while also having good battery life. Additionally, the development of new technologies like sodium ion could improve power density and reliability of batteries! Something to look at too.

Many laptops have replaceable batteries that are not glued to the chassis, and they have great battery life. Framework’s “13 Pro” laptop has a battery that can compete with the (Apple Silicon) Macbook Pro, even beating it in some tasks! This is while making lots of other parts replaceable too, like the motherboard, keyboard, speakers, display, etc. It’s not just Framework too, many other companies have laptops that do not use glue (Lenovo, Microsoft Surface, some Asus ones…), and their batteries are not significantly smaller than the competition.

The Fairphone 6, which has a screw-in replaceable battery, has a very respectable battery life (can be multi-day depending on usage) that can compete with competitors in the same price range of 600 euros. HMD also makes lots of phones are easily repairable, and their battery capacities are just as good as the competition and there are no major complaints about worse battery life compared to the competition.

Oh, and you also have files like .zip or .tar(.gz), which are used to store a compressed version of some amount of digital files. And they can different in compression techniques, how data is arranged, etc.

Not just those. Files are just a method of storing digital data, so it’s not just those four. You can have files storing databases, software (think exe, AppImage, deb, rpm, etc.), design files, projects, and more!

And file extensions are a method of telling different programs how to handle different files, since the data is formatted a bit differently. For instance, a “.txt” file is stored in plain text, while an executable file is compiled code that needs to be run.

For your example, I would like to note that you are comparing a plain text file type to a rich text file type. Plain text file types, like .txt, .md (Markdown), and the different code files (like .json, .py, .rs, etc.), can be viewed and edited with a simple Notepad-style text editor. The data is stored, as the name suggests, in plain text. In comparison, rich text file types, like .odt and .docx, encode additional data like fonts, styles, images, animations, etc., and require a rich text processor (like LibreOffice, MS Office, etc.) to read them. You can’t view them through a notepad-style application, for example.

And for images, video, and audio, you have it take into account compression, codecs, that sort of thing. You might have heard that a PNG can store transparent images and is a lossless format while a JPEG cannot and is a lossy format. “Lossless” means that, after compression, no data has been removed (or “lost”), while “lossy” means that some data is removed after compression. For audio, MP3s are lossy while WAV files are lossless. You might have also heard of “raw” photos and “raw” videos, those mean that the data is directly from the camera in its original quality.

For most file types, you can’t just change the extension to convert them, as the data stored is arranged differently! This is why renaming a .txt file into a .odt will not be a valid rich text document, for example.

Huh, I didn’t know that. Pretty interesting! I wonder how many acronyms have multiple different long forms…

Having a large internet personality like Pewdiepie advocate for privacy, self-hosting, and open-source is always good!

Vikunja is a great option, but the mobile app is still in development (lots of missing features). I like it, but I have found that using a simple markdown editor like Markor (on Android) and Marktext (on Linux, available on Windows + Mac too) works for me, and I use Syncthing to sync with my laptop. Note that Marktext isn’t being actively developed anymore, but it was the best one I could find that respects the folder structure of my notes. Obsidian is great too, but it’s closed source and having everything in vaults is annoying for me. You might also like VSCodium or Kate, which are code editors that also have markdown highlighting (but not Markdown preview).

However, you can’t easily use custom themes with Markor and Marktext. For Markor, you can only change the background, font, and text colour of the editing mode (not the preview mode!) while Marktext does let you change the font universally, but you have to use one of the prebuilt themes (I found One Dark to be the closest to my system theme). One day I would like to try my hand at making my own cross-platform markdown editor, maybe once I have less schoolwork to worry about.

edit: I was wrong, Marktext is actively developed! Additionally, the latest version of Marktexts lets you use custom CSS to create your own theme, and there are far more default themes to choose from now!

As per his video, he states that he wishes that he could magically make AI go away and that he “hates everything in his project”, but since that’s not possible, he would prefer if people using AI did so using their own hardware and not some company’s cloud servers.

Make of that what you will. I think it’s pretty neat, not for me, but I’m sure someone will find it useful.

Note that I have seen a lot of people make some really cool “rehousings” of their laptops to turn them into transparent boxes mounted to the wall, usually made of something like acrylic. They look awesome, but haven’t tried it myself since I just self-host using my laptop in its original chassis

Let’s dissect thus term by term. First, what is a license?

In a nutshell, a license lists what you can and cannot do with something. They can apply to both software and code as well as things like art and music. You might have heard of Creative Common licenses and things that are “free for noncommercial use” or “only for personal use”.

Open source means that others view, modify, and redistribute the source. They can apply to software as well as hardware, art, music, etc. If you want to look at the specific terms, search up the Open Source Initiative. The Free Software Foundation also has their own definition you could look at.

So, an open-source license is a list of things you can do with something, usually but not always software, that aims to allow others to view, modify, and redistribute the source.

There are multiple open-source licenses, and they each have their pros and cons, but there is a general split between permissive and copyleft licenses. Permissive licenses (like Apache, MIT, etc.) try to provide the user more freedom by having no restriction on how the source can be used, while copyleft licenses (like GPL) aim to ensure that freedom will last by restricting how closed-source projects can use open-source licensed bits.

There is a large debate between the two that I won’t get into, there are plenty of articles available discussing points for each side.

Value is inherently subjective. Money, gold, products, services only have value because of people. An ant doesn’t care for a solid brick of gold, it serves no purpose for the ant, but humans see the shiny metal as incredibly valuable. And while the same ant might think that a small insect as valuable food, most humans do not see the same singular insect as all too worthwhile to eat.

What is valuable to you could be worthless to somebody else. In the context of Netflix, most people probably consider it to be decreasing in value as the content (or the quality of it) has not increased enough to compensate. However, there are some people that might be looking forward to a Netflix-exclusive show and believes that it adds a lot of value. In that case, the price increase might be worth it for them.

Something similar could be said for other streaming services. Spotify, Apple Music, etc. could have your favourite artist listed that can’t be listened anywhere, and you might consider that valuable.

No I do not. Why would you need to speak to them? You just create an account on their website, and you can add any kind of code on Codeberg as long as it has a open-source license and it isn’t illegal.