Lemmy Project Priorities Observations

5 readers

1 users here now

I've raised my voice loudly on meta communities, github, and created new [email protected] and [email protected] communities.

I feel like the performance problems are being ignored for over 30 days when there are a half-dozen solutions that could be coded in 5 to 10 hours of labor by one person.

I've been developing client/server messaging apps professionally since 1984, and I firmly believe that Lemmy is currently suffering from a lack of testing by the developers and lack of concern for data loss. A basic e-mail MTA in 1993 would send a "did not deliver" message back to message sender, but Lemmy just drops delivery and there is no mention of this in the release notes//introduction on GitHub. I also find that the Lemmy developers do not like to "eat their own dog food" and actually use Lemmy's communities to discuss the ongoing development and priorities of Lemmy coding. They are not testing the code and sampling the data very much, and I am posting here, using Lemmy code, as part of my personal testing! I spent over 100 hours in June 2023 testing Lemmy technical problems, especially with performance and lost data delivery.

I'll toss it into this echo chamber.

founded 1 year ago

MODERATORS

[email protected]

2023-08-26 data, database (bulletintree.com)

submitted 1 year ago by [email protected] to c/[email protected]

2 comments fedilink hide all child comments

It isn't just the constant database crashes in Lemmy, GitHub issue 2910...

https://github.com/LemmyNet/lemmy/pull/3708

On July 24, sanitation of HTML was added to the code. But the testing was not called for and it broken titles of postings, link parameter ampersands, discussion of programming code in code blocks

It's data... and now it's very difficult to undo all the damaged data that has been put into the database for weeks now.

Lemmy is a Link aggregator, and it damages Links now... the ampersand parameter deliminator in URL links is now broken because of this code not being tested. Why wasn't there a call for testing to something that was going to alter every new post and comment from both federation and Lemmy itself? how did such obvious things such as a ? parameter list in URL get overlooked... and then new bugfix release comes out after this was known as an issue - and still not fixed.

Database crashing that results in lost data from unsaved post and comments, failure to deliver Federation data without any way client or server operators are notified, and damaged data as fundamental as URL website links...

I'm all for code changes gong in fast, but the lack of actually testing things and spot-checking on Lemmy instead of just changing Rust code without really realizing that a link aggregator uses ampersand in URL links... and not asking people to help think of side-effects...

Development process could even ask just a couple sites with more attentive operators to try out the code for a few days and ask people to report any problems before advising all sites to upgrade and break their URL links.

top 2 comments

sorted by: hot top controversial new old

[–] [email protected] 1 points 1 year ago

Broken URL links because untested HTML sanitation... that then goes into a release and for weeks has been corrupting data.

At the same time, another change went in with how deleted and removed comments are delivered to clients. And if a person deletes a comment, now all the replies below that comment - even from other people - are no longer shown.

This is a fundamental data issue of the application! A change is made with the internal Rust code logic on how comments are deleted - and no client-side testing is done to make sure there are no side effects.

If B and C reply to A, and A deletes their comment, they all 3 have their comments black-holed by Lemmy. But an admin logged in would not see the change. But the change was made in the Rust code... the developers did not test it as normal end-users of the website... and they do not seem to login on Beehaw or other sites just to check the health and use of the application. It's like a business that never visits their customers and see what their engine is being used for. And it seems it has been that way for years...

Data doesn't matter... that's the attitude. People's comments just disappearing isn't noticed.

[–] [email protected] 1 points 1 year ago

Managing data in 4 year old lemmy code...

There is an option in admin to purge data in a community. The UL isn't built for long-running SQL statements, so it is likely to not return after pressing the button if there are more than a couple thousands posts in the community... which is likely the reason one would purge it.

But the whole separation of local from non-local that Lemmy uses in the database. The purge has no option to preserve locally created data. So comments and posts from your local users get purged along with the community you followed from a remote server.

So if you follow [email protected] - a very active community, one of the highest number of posts and comments in the Lemmy network - and Lemmy is crashing because of the poorly performing SQL code - you might opt to purge your local instance copy of that community to reduce the PostgreSQL crashes and storage consumption. A perfectly natural thing to need to do to manage a server running for years like lemmy.ml or Beehaw has....

Without an option to preserve posts and comments with local content, that means you are deleting what a user on your instance has created. it will disappear from their profile, they may go back looking for a copy of a post the created a couple weeks ago - and it's just gone! Your local instance may have only created 300 of of the 10,000 posts in that community - there is no option to preserve them or even not remove posts that your users had commented on.

This is a data-centered application, but it shows no learning of the kind of things a 4.5 year old Internet website and development project would show. And the developers who create the site do not seem to use Lemmy to even discuss moderation, tips for new instance operators, programming issues or bugs... Lemmy as a data and information system is not something they feel is up to the task or useful.