

I just found this in your buy me a coffee page:
promotion of the livingrimoire AGI software design pattern
If you think that whatever your creation is is a “AGI software pattern”, then it should become obvious why you got banned from everywhere.


I just found this in your buy me a coffee page:
promotion of the livingrimoire AGI software design pattern
If you think that whatever your creation is is a “AGI software pattern”, then it should become obvious why you got banned from everywhere.


I think Cinny has a pretty similar interface.


That is a bad idea. Cross posts aren’t a special type of post. Your instance simply checks the URLs of posts and finds other posts with the same URL. That is how cross posting works in Lemmy.
[...]
// Fetch the cross_posts
let cross_posts = if let Some(url) = &post_view.post.url {
SearchCombinedQuery {
search_term: Some(url.inner().as_str().into()),
post_url_only: Some(true),
type_: Some(SearchType::Posts),
..Default::default()
}
[...]
Piefed seems to do the same thing
[...]
new_cross_posts = db.session.query(Post).filter(Post.id != self.id, Post.url == self.url, Post.deleted == False,
Post.status > POST_STATUS_REVIEWING).order_by(desc(Post.id)).limit(limit)
[...]
I also found this weird hardcoded ap id in the function that “calculates” cross posts in piefed for some reason:
[...]
if self.community.ap_profile_id == 'https://lemmy.zip/c/dailygames':
# daily posts to this community (e.g. to https://travle.earth/usa or https://www.nytimes.com/games/wordle/index.html) shouldn't be treated as cross-posts
return
[...]


oh man thats great. thanks lmao


Does anyone have an archive of the no poop challenge?


We say “What email provider do you use?”.
Not much difference.
Yes. The rule is “no US politics”, not “no politics”.
As long as discussion of Venezuelan politics does not turn into or knowingly cause US political discussion, it will be fine. Let’s be honest here though. Given the current events, it will surely turn into one.
A while ago, we made rule 6 a temporary rule that was initially meant to last until the US elections. Later, through a community vote, it was made permanent.
Community vote: https://lemmy.world/post/23032034
Temporary rule 6 announcement: https://programming.dev/post/21409464


Yes. ForgeFed is an ActivityPub extension just for repository hubs like Forgejo.
For example someone on instance codeberg.org will be able to interact with repos hosted on instance gitea.angry.im.


Peertube. I’d like to see more channels from YouTube at least mirror their content there. It’s fine if they don’t want to deal with another platform, but let us at least do the mirroring.
They used to do affiliate link injections
Once it approaches maturity, surely it could switch to e.g. moar reliability at the expense of agility?
Unfortunately, you can’t just “switch” to more reliability. That’s not how it works. Reliability is affected by the design and implementation of the software. If your software wasn’t reliable in the beginning, you’ll have lots of “fun time” rewriting and even more “fun time” debugging components to try and make it more reliable.
At some point, this is no longer practically viable and the unreliable parts stay as they are. Especially if the unreliability is a design flaw. This is known as “technical debt”:
While an expedited solution can accelerate development in the short term, the resulting low quality may increase future costs if left unresolved.
However, failure to prioritize and address the debt can result in reduced maintainability, increased development costs, and risk to production systems.
https://en.wikipedia.org/wiki/Technical_debt
As for the 0 priority of testability, I found a test directory in the repo: https://codeberg.org/rimu/pyfedi/src/branch/main/tests
Though I haven’t looked at how much the tests cover, there seems to be some sort of testing. So perhaps this importance table is a bit outdated indeed.
although tbf that is only one of several, others being “it’s too complex”
As I explained in my original comment, Rust has a few strict rules and concepts that force a particular discipline (I guess you could also call it style of code). And since it has those safety guarantees, you need to have things like lifetimes and trait bounds and such. These things naturally make the code harder to read and/or complicated. But as I said, they are worth it and will save you many bugs that might occur at runtime (especially ones you do not want to debug at runtime. like concurrency bugs).
You’ll appreciate Rust the more you use it. Especially if you come from languages that do not have the same luxuries like Rust.
The type system and the safety guarantees that are enforced by the compiler are so pleasant to work with, I actually do not ever want to use any other language.
This is also one of the reasons why many companies are now choosing to use Rust. It is not just because it is new and “shiny”, but because it has real world benefits. Memory safety, for example. The tooling is also superb.
Yes, you will not be able to write at the speed a Python dev will write a project, but the rules and concepts Rust introduces are very worth it.
I think the importance table of Piefed demonstrates this pretty well:
- Performance 4/5
- Scalability 2/5
- Agility 5/5
- Reliability 1/5
- Security 2/5
- Testability 0/5
- Modifiability 5/5
- Affordability 5/5
- Manageability 5/5
https://codeberg.org/rimu/pyfedi/src/branch/main/docs/ARCHITECTURE.md#desired-quality-attributes
You simply cannot have good reliability or security if you prioritize agility over everything. Same goes for scalability
Also why does testability have a priority of 0?
If you run your own instance of Lemmy, you still can view all deleted and removed posts.
chromium gets jpeg xl support, a more superior alternative to jpeg and webp