I learned writing code for free, by looking at other code and reading online manuals.
I also learned how to play the piano by listening to people playing it and getting online sheet music for free.
Do you have a bot army that scrapes websites and repositories? Do you not give credit when credit is due? Do you poorly plagiarize work and claim it to be your own?
While I appreciate the learning aspect, the abuse of these LLM’s is blatantly clear.
Who cares?
The people who actually spent the time writing the code and documenting it, remember these LLM’s have to learn from somewhere.
A lot of repositories use the GPL license.
The GNU General Public Licenses are a series of widely used free software licenses. The GPL is a copyleft license, which means that it guarantees end users the freedom to run, study, share, or modify the software, but if you distribute a derivative work or modification, you must provide the source code to those recipients under the same or equivalent license terms — there is no requirement to publish anything to the public at large.
I doubt many LLM’s are sighting their sources or giving proper credit when they spit out code snippets.
What distro are you using? There are 3 big distro’s you need to keep in mind for, Debian, Arch and Fedora, every other distro typically bases themselves off of one of these big distro’s.
Documentation is your friend, if you’re using a Debian based distros the stress ng repository linked above has this in their documentation -
Debian packages for Ubuntu
Recent versions of stress-ng are available in the Ubuntu stress-ng ppa for various Ubuntu releases:
(run the commands below to install the package)
sudo add-apt-repository ppa:colin-king/stress-ng
sudo apt update
sudo apt install stress-ng
Now that is installed, they have an examples section.
Examples
Run 8 CPU stressors for 60 seconds:
stress-ng --cpu 8 --timeout 60
stress-ng: info: [184401] setting to a 1 min run per stressor
stress-ng: info: [184401] dispatching hogs: 8 cpu
stress-ng: info: [184401] skipped: 0
stress-ng: info: [184401] passed: 8: cpu (8)
stress-ng: info: [184401] failed: 0
stress-ng: info: [184401] metrics untrustworthy: 0
stress-ng: info: [184401] successful run completed in 1 min
Couple of definitions and examples that will help you along the way.
If you really need to expose something to the internet set yourself up an SSO/OIDC provider and have it sit in front of the service and have a deny-first policy in place, whitelist the known few IP’s that need to connect regularly or alternatively geoblock.
Also suggest setting up some security headers, however headers can be confusing for beginners.
Virtualize or containerize everything.
Somebody watched the backrooms recently.
It’s clear that Redlibs maintainer is burnt out, as were the LibReddit folks who developed and maintained it initially before it was forked. Software like this is an endless cat and mouse chase, Reddit break something, the community works around it, rinse and repeat.
It’s why we don’t see Bibliogram around anymore, Instagram made their APi unbearable to work with.
In regard to the difficulty of keeping a Redlib instance alive, currently most of the discussion occurs in one issues post on GitHub, and people have suggested other docker images that currently work.
If you host a SearXNG instance you can replace Reddit links with Redlib links, just add this to your settings.yaml.
hostnames:
replace:
'(.*\.)?reddit\.com$': 'lib.your.domain'
'(.*\.)?redd\.it$': 'lib. your.domain'
Ubuntu is based off of Debian, just use Debian instead.
Got that forehead tattoo like DC Young Fly
Matrix is nice but it’s clear they’re appeasing more towards shareholders than the community, self-hosting via Docker Compose was a PIA as there are 4-5 services that need to talk to one another.
Guess you won’t be making issues there for a while eh