How much damage is the internet needlessly doing to the environment

9 risposte [Ultimo contenuto]
andyprough
Offline
Iscritto: 02/12/2015

As a sidebar jumping off from our librejs discussion and how the internet is broken, I'm wondering if anyone has seen any research into how damaging the modern internet is to the environment. I'm talking strictly in terms of the excess use of electricity to handle all these billions of websites loaded up with 50mb+ worth of garbage when they contain at most 5kb worth of meaningful text.

For example, if I look for a recipe to cook baked lasagna, most of the recipe websites will be overloaded monstrosities, with tons of js, auto-playing videos, numerous privacy-invading tracking devices, tons of advertisements, etc. All in order to serve up about 25 lines of actual useful text on what ingredients are needed and what temperature to bake a lasagna. So, whatever vps is serving that website up is using probably 1,000 times more electricity and resources than needed, and my machine is similarly using vastly more resources to view the page than are useful or necessary.

And that's not even to consider the other billions of websites constantly serving up exabytes of and pR0n and other completely worthless uses of energy resources, websites that have 0kb of actual useful data. And it's also not considering all the rare earth metals and mining and the pollution from all the manufacturing required for all the billions of people to have super computers in their back pockets and instant access to all these worthless websites.

Just a question about what is the environmental cost of electronically moving around all these petabytes and exabytes and zottabytes and yottabytes of worthless data. It must be an awful lot. If anyone has seen any research, I'd be interested.

koszkonutek
Offline
Iscritto: 03/19/2020

I have no research but I have an idea. Let's use this to convince ecology enthusiast to support our cause

calher

I am a member!

Offline
Iscritto: 06/19/2015

Someone tried/is trying to make a truly 100% sustainable website.

https://solar.lowtechmagazine.com/

andyprough
Offline
Iscritto: 02/12/2015

I've seen that, it's quite a good idea. I'm thinking also a return to an all-text or mostly-text web without all the animation and incessant privacy violations is a good idea, à la Project Gemini: https://gemini.circumlunar.space/

Just dump html/http altogether, and move to a protocol which does not allow all the abuse.

koszkonutek
Offline
Iscritto: 03/19/2020

Let's think what could possibly replace WWW? Gemini has not been created with this in mind, as stated on its main page. We need something simpler than WWW but we want the simpler thing to still provide some reasonable subset of functionalities of WWW. Log-in/log-out, fill a form, etc. How about standardizing a certain subset of current HTML and CSS, without javascript, and marketizing it as a new, lean platform? If reasonable, we could replace HTTP with something like Google protobuf but without getting rid of HTML+CSS. Then what's left is some good marketing.

I once had a discussion like this on email and someone stated that:

> Banking, computer games, dancing kittens, fancy presentation with
> drop shadows and fading edges - should all be handled by some other
> _application_.

Do some of you also think this way? If so, read on.

Some of those things should be handled separately. The problem is - the need for "applications" enabling simple things like login/logout and data posting would be enormous. I hope you all agree we'd end up with tons of "apps", each of them being specific to single online "service" and most of them being proprietary, not to mention not being properly packaged for distros. Eventually, someone would come up with an idea of serving the apps online, each under its own URL and we'd get back to where we started - with something equivalent to "web" (or worse). The choice is not just between current "web" and something simple. The choice is between "web" and something simple accompanied by tons of nonfree, incompatible, standards-less "apps".
The good think about (perceived) feature overload of HTML+HTTP is that it allows creation of some of such "applications" without requiring the user to run app's own code, thus saving users from the possibility of being given nonfree code.

Also, I am pretty sure typical users value things like fancy-looking pages. Even though I might sometimes prefer minimalism, I understand not everybody does and I am trying to respect people's different preferences, as long as those preferences don't contradict with software freedom principles.
I also often think about ways to encourage people to using free software. If free software is able to meet their preferences, it helps a lot. A very minimal web alternative will - despite all its good sides - remain unappealing to most people. If I am to choose between getting 10 people switch to a minimalist swfreedom-friendly system and getting 100 people switch to a moderately more complex system that's still swfreedom-friendly and more environmentally-friendly than current web - you can probably guess what I would choose.

You might be wondering why I suggest retaining HTML and CSS and not replacing them with something more lean that also provides the necessary features? Because their current popularity would ease adoption of our hypothetical platform. Many tools would be immediately compatible with it

lanun
Offline
Iscritto: 04/01/2021

> what could possibly replace WWW? Gemini has not been created with this in mind, as stated on its main page.

I think you might be somehow mistaken about that. The statement on the Gemini project's main page only shows the humility and realism of its support team. There is no doubt Gemini could possibly replace the web for many people. To achieve that, it does not need to replace the web for everyone. This is how I read that statement. I might be wrong of course, but I see many ways in which I could manage to completely circumvent the bloated web without hurting either my workflow or my personal use of online resources, and Gemini is a very good candidate to become part of the solution.

Similarly, it seems unrealistic to think that free software is going to replace proprietary software for everyone. It most probably will not, as Gemini will most probably not replace the web for everyone. We are still living examples that free software is a viable replacement for many people.

The Gemini people are also right and wise not to strive to remove the web because there is enough room for diversity in this world so usages can vary from person to person and from group to group. Hegemonic behaviors need to be contested with all opposing power available, and for that alternatives are needed. Not necessarily full replacements. It simply does not make sense to try to replace an hegemon by another.

> dancing kittens

What a terrifying hegemon. Have they been neutered yet?

Avron

I am a translator!

Offline
Iscritto: 08/18/2020

> Just a question about what is the environmental cost of electronically moving around all these petabytes and exabytes and zottabytes and yottabytes of worthless data. It must be an awful lot. If anyone has seen any research, I'd be interested.

https://www.researchgate.net/publication/326470455_Evaluating_the_Energy_Consumption_of_Mobile_Data_Transfer-From_Technology_Development_to_Consumer_Behaviour_and_Life_Cycl...

This is not exactly what you are asking for as it is on a cellular network but it is interesting. It shows for instance that the power consumption may not increase with the amount of data.

I guess the Troll Longe would be a better place for this discussion, this is not about Free Software.

lanun
Offline
Iscritto: 04/01/2021

> the Troll Longe

Is that what they use to train trolls?

andyprough
Offline
Iscritto: 02/12/2015

Don't be a spelling nazi - he clearly meant to say it belongs in the "troll luge".

Obviously true, because this discussion needs more speed, power, and a sense of defying death.

troll-luge.jpg
lanun
Offline
Iscritto: 04/01/2021

I see. Mind the troll deluge then.

Now you mention it, I had been wondering if the intended spelling was not in fact the "troll loge", from where this discussion could be followed with the appropriate loftiness.