The upstairs neighbor: [Frost troll wearing clogs doing jumping jacks]

Yeah, I just avoid all communities that start with “Ban” or “Fuck”. Life is much less irritating that way.

I used to, but I try to do better. Lately, I’ve had “blind voting” turned on in my app, so I can only see the score after I’ve voted one way or the other. Basically forces me to read and vote on the submission without being influenced by the current score.

Worked at a local pizza place my junior and senior years of high school.

Pancakes … In there silverware drawer

Lol, yeah

The Fediverse is just full of Carols (on the right/brown coat):

What do you want to practice? Just general sysadmin stuff? Networking? Clustering? Horizontal scaling? All of the above?

Old PCs are just Debian servers waiting to happen. Depending on their specs, you may be able to do VMs or you can utilize container frameworks like Podman, Docker, or LXC to deploy individual applications or application stacks. Or you can just bare metal install anything you want.

Years ago, I bought a batch of 16 Wyse thin clients on eBay for about $15/each. These had 4GB SSDs and 2 GB RAM, so I upgraded about half of them with 64-120GB SSDs (whatever I had lying around) and 8 GB RAM. Thin clients can usually be found pretty inexpensively and are pretty power efficient, but they’re not performant workhorses. They’re great for practicing networking, VLANs, system orchestration (e.g. Ansible, Cockpit) application clustering and horizontal scaling, diskless workstations, setting up a demo office server and workstations, and even VMs if you’re just practicing; they’re a little underpowered to run a lot of VMs, but you can certainly run a few small ones just to practice managing them.

than to carry something like an evaporative cooler

Evaporative coolers don’t really work in high humidity. If you live in an area that’s a dry hot, they work great. Summers in my area, though, are very muggy. Other than ice pack based products, the only passive coolers I’ve found work in humid environments are these sweat bands that have either desiccant beads in them or that stuff that’s in diapers. They pull the sweat away keeping it out of your eyes and give a little evaporative cooling at the same time.

Webapp. It’s the development branch of Tesseract.

I can use regexes, so it makes it a bit shorter list:

I’ve toyed around with LLM-based moderation tools but it never really panned out. It was too hit or miss to be relied upon even with the temperature parameters turned way down in an attempt to get consistent results. Granted, I was using a small local model and not feeding it to one of the big players.

To give an example, I tried to keep it focused by creating one custom model per rule to enforce. An example prompt to mod calls for violence was basically:

ROLE: You are a forum moderator who does not want users calling for violence.  Examine the input and analyze whether it violates any constraints. 

KNOWLEDGE:
- {list of dog-whistle slang for calling for murder}

CONSTRAINTS:
- Content should not advocate violence
- Content should not normalize violence
- Content should not escalate tensions or fan flames
- Content should avoid promoting harmful stereotypes
- Content should not utilize broad, sweeping generalizations
- Content should not use dehumanizing language
- Content should not undermine human rights, due process, or the rule of law

FORMAT YOUR RESPONSES AS JSON:
{
  reason: [A one to two sentence summary],
  score: [On a scale of 0 to 10, how severe is the content advocating violence]
}

The score part of the response was my band-aid to get around the high number of both false positives and false negatives as I originally had it returning true or false only. Any score 7 or higher caused the item to be passed to the mod queue along with the reason, and I would review its actions later.

Ultimately it was slow and still somewhat unreliable, so I abandoned the idea after running it for a little less than a day since I can 't run bigger models to get better results fast enough to keep up. Using a cloud based service was out of the question for many, many reasons, both financial and ethical.

To answer your question, as long as the models were locally hosted and properly tuned/tested, I’m fine with it in theory, except for the ideology part; that’s pretty messed up. While I don’t want my submissions used to train anyone’s model and take measures to prevent my own instance from being used as a data source, I remain aware that once I post something, I have no control over its fate the moment it federates out.

[1] Yes, I know that’s like half the comments that get posted around here. My goal was to try to have it mod things so posts were bases for actual discussions instead of being a knee-jerk rage factory.

I used to do that but once the backend added that feature I removed that step from the automod script. Basically it was to prevent the communities here from being unmoddable on remote instances.

I’m not even a real instance anymore, how did I make the list 😆

But also, you should see the local numbers haha

lemmy=# select count(distinct other_person_id) from mod_ban where mod_person_id in (1, 2,288);
 count 
-------
  9792
(1 row)

I wonder what happens when I hit 10,000?

Any time I hear Minerva without that K-Rock intro, it just sounds wrong to me lol. That, and the “That was new Deftones. Minerva is the name of that song” at the end.

Yeah, I think so.

Man, why is sharing a 10 second mp3 so hard? lol. This or a shorter version without the “world premiere” part is what I was recalling:

https://tube.dubvee.org/w/gC3pchZNmSTeZxX2V2fqrK

I was surprised that so many songs in my collection weren’t actually supposed to start out “schhhhhhhh K-Rock (k-rock)” lol

[email protected] is my pet community. Basically an oldies station for Millennials.

Technically, yes. But colloquially, when we’re talking about “analytics” we mean embedded 3rd party trackers that feed to Google or another outside entity. Those are embedded much deeper in the application and track things much more invasively such as how long you hover over certain links, how you move your cursor around the screen, your viewport size, browser fingerprinting, and more.

The analytics I’m utilizing and referring to here are passive in that they’re collected anyway as part of the standard logging that happens when you access the webserver which is also part of our basic security posture. They’re not as granular or invasive but can still give you useful information about what parts of your site people use the most, how many clicks it takes a visitor to get from the homepage to where they want to be (by following the IP, URI, and seeing where that ends), how many visitors the site gets per day/week/month/etc, and such.

Logging is standard practice if you give even the slightest damn about security (read: you should), so I don’t see it as a problem. It’s what you use those logs for, how long they’re retained, and whether you sell them off.

So as long as you’re only using them for security auditing and website analytics and don’t keep them forever and don’t plan to sell them to data brokers, there’s really nothing to fret over. A good place to disclose how you use the logs, how long you retain them, and what is logged is in the site’s privacy policy.