This is then probably better as a wiki, but I do understand why people choose github repositories for the curated nature of them.

uBlock Origin is just required for me to block ads and other annoyances. Generally I also use it to block the annoying cookie popups instead of clicking any of their buttons instead of opting for a different extension, like I don’t care about cookies.

I also almost always have Tree-Style Tab, though I’ve found that it can cause Firefox to be non-performant on memory-constrained devices. This extension gives you a tab sidebar (using a similar interface element like the bookmarks sidebar) that organizes your tabs in a tree. I will often have hundreds of tabs open normally, and this makes them more manageable, as I middle-click everything to open things in a new tab, particularly when I’m researching. It gives a nice rudimentary tree view of what content related to what, like if I’m on a Wikipedia dive or TV Tropes dive.

I also usually have Tampermonkey, as I usually want to make a website more accessible. A lot of the time, it’s simply so I can autofill usernames in a text box, because the website might have the login flow split between pages, and Firefox does not recognize or allow me to select an autofill the username for me. Other times, it’s to automatically click through that annoying Microsoft login flow “Keep me signed in? (Don’t ask as often)”.

For watching YouTube, I use the SponsorBlock addon, because sponsored segments have become very annoying, though some creators have been able to make them actually fulfilling to watch.

On Android Firefox, I also get Disable Page Visibility API. This allows me to use YouTube in Firefox and even background the app and listen to music that way. Plus, with uBlock Origin, I am also able to block ads. I also sometimes install this on different profiles for situations where I believe the web app may try to detect my visibility, such as an online coding test for job opportunities where it will not want me to change windows to look up answers or type code into a compiler (this saved me during a C++ multiple choice exam which asked many times “Which of these is invalid C++?” and “What is the output of [this complicated code block]?”).

The below is for my work computers only, because I generally don’t need them otherwise.

I will additionally have Firefox Multi-Account Containers and couple that with Simple Tab Groups as I have 6 logins, with one particular site requiring 3 of those logins. I essentially configure the URLs I need to be in certain groups, define one container for each, and then have a couple of default tab groups set as sticky groups that will allow all containers to stay as those containers so I can use multiple sites side-by-side. The interaction with Tree-Style Tab can be a bit janky, especially when configuring new catch tab regular expressions, but usually disabling and re-enabling both Tree-Style Tab and Simple Tab Groups will fix that. Mostly I do this, because my company refuses to fix their SSO breaking due to weird cookie issues, and it’s pretty nice to be able to clear cookies for a particular tab group easily.

To clear cookies, and because I do that very often, I use Cookie Quick Manager. I can clear a tab group’s entire cookie set by having a tab selected in that tab group, then using “Delete current Context Cookies” and usually that will fix my login issues, and also keep my logins to other sites. For my personal use, I generally don’t need to clear cookies, or using a private window, clearing per-site via the web console, or clearing in settings is good enough.

Since I do a bit of web troubleshooting, Modify Header Value is pretty nice. I can call API endpoints that require a subscription key from my browser, even using the web console, and I don’t need to worry about figuring out if my headers are correct. I can also get my team to get this add-on so they can do their basic troubleshooting. For my personal computer, I’m much more free to create ad-hoc scripts to test things that I can save somewhere, as sending others scripts in my organization also comes with implicit hours of training and coaching on how to use them (and our leadership has been very sensitive to explicit training hours to get everyone up to speed).

Finally, User-Agent Switcher and Manager for the sites that are built for Chrome, but have worse performance or broken features on Firefox, yet they work fine when Firefox sends them a Chrome user agent. Thankfully, I rarely have a use for this on my personal computer.

Take a survey of all open source projects. Then find the proportion that have a company behind them trying to sell an enterprise solution. To make this easier, only look on something like the npm repository.

You’ll find “generally” is accurate.

I personally think the best way to do open source is to do it as a hobby, and not hope for profits off of it. Open source is fundamentally programmers taking control of their field’s means of production, and the last thing I want to see is corporations co-opting that moreso than they have.

This is the main reason everything I release is AGPL unless there is a strong reason against it: Corporations won’t use it.

Open source software doesn’t generally have a company behind it that you can obtain support from via a contract. Some do, but a small, but dedicated library that your entire company relies on? Probably not.

Additionally, there’s some perception that paying for software results in a better product than paying zero, which is an intuition from the adage “you pay for what you get”. Programmers and users of open source software generally believe the opposite, but executives and middle managers are in a completely different headspace from the workers that produce and use these products.

There is one aspect of that which is true: If upstream breaks your product, you have to figure it out. You can’t (or at least shouldn’t) just yell at some company upstream and hope they unbreak things. So, the support costs become the company’s costs, and who knows how much those costs actually are if you aren’t ready to track such thing?

There are so many places that games decide to put their save files on Windows.

I’ve got save directories in:

• %HOMEDRIVE%%HOMEPATH%\Saved Games (only Elite Dangerous)
• Documents (Mass Effect Andromeda, X4: Foundations, EVE Online, GreedFall, Wo Long: Fallen Dynasty, SteamVR)
• Documents\My Games
• %APPDATA%
• %LOCALAPPDATA%
• %APPDATA%..\LocalLow (does not seem to have an environment variable defined for this one)
• Various game install directories

There’s probably other places. Not sure how much the registry is used for saves, either, but that would complicate backups more than they already are.

I’d love if they just unify save and config data for games to %APPDATA%. Documents should never be touched by software without the user’s explicit consent, though, and because of the situation, the Documents directory is the last place I ever put actual documents.

I’ve had 5k+ tabs open at some points, because I just don’t close any of them, and I often middle click as I want to navigate back to the page I was at. Additionally, a lot of sites break the back button, like collapsing comments re-expanding, or it loads slowly and I wanted to look at it quick. Organization is pretty nice with Tree-Style Tab for Firefox.

Every few months I purge all of my tabs, but for the most part, I just don’t care when I have 32 GB of RAM.