He/him. Chinese born, Canadian citizen. University student studying environmental science, hobbyist programmer. Marxist-Leninist.
- 591 Posts
- 424 Comments
The most popular non-Canonical derivatives, Linux Mint and POP OS, have both totally rejected and vocally criticize Canonical’s bullshit, Snap or otherwise. This isn’t going to make the fall in line, this is going to make them finally get serious about ditching Ununtu and switching directly to the upstream Debian base.
Without changing anything else about how the code is managed, which, doubtful considering Musk (at least not for the better), a rewrite will end up just as dysfunctional as the original codebase by the time it’s reimplemented all the features.
And if you were committed to changing your coding practices, a rewrite would almost invariably be unnecessary as slower incremental revisions will invariably cause the codebase to turn over and shed the problematic parts while keeping the working stuff.
When larger codebases than Twitter have managed to completely shift languages without a full rewrite, this idea is coming from ego and Elon’s savior complex, and not a place of logic and actual necessity. Not even shift languages like Java to Kotlin (which, Twitter is written in Scala which is another primarily JVM language) I’m talking full ecosystem shifts like PHP to Python or JavaScript to Rust while keeping the codebase continuous. Not saying it’s easy, but it can at least be mostly painless if and only if it’s managed correctly. For context, Google has switched from Python to Go for its core infrastructure, Firefox is switching from C to Rust and Tor is following the same route, Patreon changed from PHP to Python a few years ago, and Discord is also switching its core infrastructure code from (IIRC) Node.js to Rust.
I’m still manually doing HTML includes for jQuery and Bootstrap. Not from CDNs either, I download the files to my repository with the correct license and attribution notices and host them on the same static file server as all my custom assets. It’s really not hard to do and also means your website has one less tracker for users to worry about (yes CDNs track you, even the ones that swear they deliver files anonymously because how exactly do you plan on proving that they actually deliver files anonymously).
Also, never really found PWA frameworks any better than good old jQuery and Bootstrap, so yeah I still use those two. This also mean my webpages do not require JS to load, making them lighter, more compatible with legacy browsers, as well as working most of the way with JS disabled if the user is not comfortable with allowing JS from some rando’s blog (which, as a rule, users shouldn’t be).
I haven’t checked but I am 99% sure that is licensed under MIT which is the darling license of the node ecosystem. When you do that you are basically opening yourself to being abused by corporations.
To be fair, if they’re just distributing the source code, not even AGPL can stop them, since they’re distributing the entire codebase, unchanged, under the same license. Plenty of other reasons not to use MIT, like you said it’s easy for corporations to exploit, but I don’t think this would have helped.
If I had to do something like that I would most likely copy paste the code from a stack overflow answer. Having a whole module for one small function seems ridiculous to me.
Moreover, the JS ecosystem is notorious for its use of helper libraries with a ton of primitives that you then use in your code so you don’t even need to deal with the standard library. The most famous and infamous being jQuery. This couldn’t have been rolled into one of those?
I’m honestly just surprised that people are putting up with their horrible redesigned webclient and app. I use Reddit a lot less than I used to specifically because of Lemmy, pretty much only for more niche tech/programming stuff not yet found on Lemmy, but when I do it’s strictly old.reddit.com and Slide for Reddit.
Being able to only see two or three comments deep in a forum specifically designed around nested comments is unacceptable.
Make sure you never connect it to the internet either.
TVs can record snapshots of what’s being displayed on screen and send it for analytics. They’re supposedly only recording a scattering of pixels throughout a screen and trying to match it to those same pixel values at the same positions generated by scenes in known media properties, which would in theory mean they can’t really recreate what is actually on screen or identify any media personal to you that’s not on their media database. (Honestly even that is creepy as fuck.)
But since the code is proprietary, who’s to say they’re not just taking full blown screenshots of literally what’s on screen every now and then? If they sent a full screenshot and compressed it with LZMA or something on the highest compression power, every hour or so and slooowly sent it a few bits at a time over the course of that hour, you’d most likely never notice since it would likely be encrypted with SSL and not be so much data that would be easily discernible from other random network activity from someone who was monitoring their home network traffic. They could totally say it’s simple HTTP requests for software updates or grabbing the latest Netflix listings or whatever. (And even then very few people actually monitor what their devices are sending. Even companies that eventually had scandals where they sending unauthorized analytics frequently and in plaintext, as in you only had to hook it up to Wireshark a single time to realize what they’re doing, still manage to get away with it for years before someone noticed.) Or, the TV could be built with a trigger where it normally doesn’t record your screen, but if you were a person of interest, they could start monitoring you whenever they want by sending a signal to your TV.
And I’m sure if you at any point connect your smart TV to the internet, it’s definitely been caching all those past analytics to send in one burst. So don’t do it.
Except direct energy input into photosynthesis is only one part of the equation with farming. Yes, if you had a certain land area, you can only do so much photosynthesis with the sun. “Coincidentally,” that’s about equal to a single layer of plants on that land area, not that groundbreaking of a conclusion. Then again, open field farming does not come even close to the maximum density of plants that can be energized by the sun. Think about how barren a farm field looks compared to a thick rainforest. Most of the sun’s energy is still hitting the dirt of the farmland, not the plants. So, if you put solar panels on the roof of a vertical farm, even with the losses in energy conversion, you can probably power grow lights for at least several layers of a vertical farm, so it’s not like all the energy needs to be piped in.
But also, consider the other factors in growing large amounts of plant biomass. Fertilizer is one: using chemical fertilizer in an unconfined open field is really bad because it can and will run off and contaminate the ecosystem. Leading to oversaturation of nutrients in the surrounding soil and especially rivers, streams and lakes, causing uncontrolled microbe growth and quite likely collapsing the ecosystem. This happens all the time. Not to mention, nutrients that run off are wasted. So using said fertilizer in a confined system where it can’t easily escape is much better for the environment and much more efficient. Same with pesticides (actually, if you contain your vertical farm well, you really shouldn’t need to use much pesticides or at all), as well as GMO crops, where the biggest concern with them is those genes spreading into the ecosystem.
Second, consider land use change. Conventional farming is the single largest contributor to deforestation, which is one of the major contributors to climate change. So, if you reduce the land area required to grow your crops, yes you get less energy from sunlight to help you along, but get this, that sunlight can then be used to recover forests, which, you know, is important on an existential level.
Finally, a huge amount of energy use associated with farming goes into moving stuff. Moving in fertilizer, water, etc, moving out the grown crops, moving around giant machines to tend to the fields, moving workers as they commute, etc. So if we can increase the density of farming and can move it closer to or even inside of population centers where the food needs to be distributed, it will go a long way toward reducing the energy use for the non-photosynthesis parts.
Energy may be an Achilles Heel for vertical farming, but just like how Achilles was still a powerful warrior despite his heel, vertical farming has many more benefits beyond this one drawback. It’s not like we can only get clean energy from solar. Wind, hydro, geothermal, hell even nuclear. We will have enough clean energy for vertical farming if we play our cards right. As @yogthos@lemmygrad.ml mentions, obviously in an energy crisis this energy requirement isn’t ideal, but do we plan on being in energy crisis forever? I hope not.
Most of my knowledge about it is from course readings and lectures in university conservation classes, but I found some open access links, from less scholarly to more scholarly (but also more layman unfriendly):
https://yewtu.be/watch?v=xzXF9usswYo
http://www.gogreen.org/blog/the-environmental-impact-of-lawns
https://www.tandfonline.com/doi/abs/10.1111/j.1944-8287.2003.tb00222.x
Cost has no bearing on environmental impact. Fossil fuels were until very recently the cheapest energy option. Disposable plastic is still by far the cheapest packaging option. Crops grown on the land of previously burned down forests is cheaper than food grown by permaculture or even half responsible land management. I thought after the massive droughts in Europe last summer y’all would take water conservation more seriously.
What companies pay