redjard

I’ve asked the author and they responded

[my question, translated]
Hi,
just a quick question about the 2nd latest blog-article “Five daemons or one binary” “Postfix and friends against Stalwart, tested side by side”:
[…] and it came into discussion if/in which way it was written/translated using AI/LLMs?

[response, german, translated by me]
Hi,

yes I use LLMs as support. But I test everything properly/carefully, [and then also(?)] by hand. I only trust the output after tests and when I understand it.

I released the tests for Stalwart today. https://www.behindtheconsole.com/2026/11/11/testing-stalwart-before-trusting-it-with-mail-the-test-catalogue/

Cheers, -mat-.

There’s no way that’s a german accent. There may be some human some ai parts, but that does still call into question if they made sure everything came out correct. Just a human translation from someone not familiar with every detail can already imply something wrong due to slight changes in wording if you are unlucky.

And the main original issue was “if they used ai for the writing, what else did they use it for?”. AI is a sign of low quality, and with lots of 100% slop creations posted on this com I wouldn’t read that sort of thing without someone vouching for the quality of the source (Which I take it you do?).

tldr Because of the ai slopocalypse I just don’t like being hit in the face by ai text from opening a random post.

PS: would a human let this graphical issue slip through?

I noticed that together with the presumed-ai text earlier, but chocked it up to mobile layouting. But it’s the exact same on desktop.

Can you give this a slop disclaimer at least?

I’ve found brave to simply not be good enough to use as a main search engine. Depends on your needs of course.
Iirc it didn’t even respect quotes.

This is extremely work in progress and to my knowledge only currently available for basic certificates not wildcard ones. I assume since @doomsdayrs@lemmy.ml is dealing with dns stuff, she needs a wildcard cert.
Might be good for the future tho when (if) that is added.

The working features should allow to use the device in most common use cases. A phone for example would typically have calls, SMS, mobile data, Wi-Fi, audio, battery charging, Bluetooth and camera.

What’s the battery life like these days?
That seems like it’d be one of my main blockers, if it can’t last a day with low use.

I once soldered a usb wire into the terminals where the battery of my moto g4 used to be, and it kept crashing on boot. So I added a large capacitor right across the terminals, and even with that and being at 5V (on a 5 amp supply), some apps still overloaded it and got it to crash.

Would love if there was some package that just attached like a battery with a usb wire coming out the bottom, that could supply any spikes if the usb provided enough average power.

even a very similar logo …

Oh true. And some gap between the disks? Or is it sufficient as is?

Hows the hdd cooling situation?

Sorry for wrongly accusing you then

You ran a classifier on a stranger’s paragraphs, compared the score to your own, and diagnosed a mental illness from sentence rhythm.

I felt something might be off, and asked. Got a hostile and weird response, and only then ran a classifier to have something that’s not just my own judgement. Of course this isn’t perfect.
I also don’t claim you have ai psychosis, I explained why I care if you sound like ai or use ai.

If writing in full sentences …

It was a few specific phrases and the style of headers. “The “off-topic” defence has a hole in it.”, “What happened when this was reported”, “Three things to take from that” (from xda). Things like that.

I wrote his best defence for him. You didn’t read it, but you did run a detector on it.

I checked the last paragraphs of your post here on lemmy (which I read in entirety), nothing more. I didn’t read the xda post when I wasn’t sure this was worth spending more time on.
I have now read it, I don’t see you mention the potential of a private agreement there either.

Now the two questions you asked as if they were checkmate, both answered in the document before you showed up.

Not sure which document you mean. But I asked to make sure you know what you are talking about (and because I was missing that info), and your answer makes sense, so that’s fine. That’s why I asked.
󠀀
󠀀
Anyway, so the plus “fork” could have a private agreement, and telegram would fully legally be allowed to do that since they own all contributions? And at the same time contributers can’t sue, only telegram can, if the “fork” doesn’t have an agreement? Meaning either way Telegram lets this happen?

As for downstream by itself, I don’t see any legal problem. If the “fork” dev wants to no longer publish code and only publishes 3 documents of text, that’s fully in their right. You can keep copies of the old code, but if there are no upstream requirements then nothing in gpl forces code to stay online. As code owner you can always make a closed fork.
(That’s why software on gpl but with CLAs that sign over contributor’s rights, or software that doesn’t accept contributions, is fake foss. But that’s a different topic)

I’d then say Telegram have a free “foss” branch and a paid closed branch based on it, which is a common business model for “foss” software companies.
With the difference that Telegram obfuscates that by making the monetized project publicly separate, and would now potentially be hiding that fact by censoring discussion?

deleted by creator

Not irrelevant at all. For one it’s something you should declare at a minimum, not doing so seems dishonest to me, especially on a foss community.

I checked and it’s not just my impression, the “baseline ai detector” pangram also flags you reliably in all parts I tried for example (even more than me, I only get the impression from some paragraphs). So I would like to know why you use very ai-like phrases and formatting if this isn’ tai.

Also, sounding like ai is a valid explanation for a lot of the treatment you described. You call yourself “opensource-compliance-gh”, and revive a 6 year old github issue, by posting a lengthy text about legal stuff and morals that sounds like ai.

And, if this is not ai, then sounding like it is a massive sign of ai psychosis, which would mean any facts you reference are suspect by default. “AI voice” is a sign of untrustworthiness, and for good reason.

As for the actual post, have you actually made sure you understand the legal situation? Have you tried finding out from the devs if there is a special licensing going on? Have you checked for CLAs (do you even know what that is)?

PS: “Irrelevant. Stick to the topic.” is how you talk to an ai, not a person. You don’t command prople like that, it’s at best extremely unfriendly.

Did you write this using ai?

A second nas. Both my nas’ have raid. They also have revisions, and one is off except for when running backups.
A lot of stuff has a third copy on the server I run stuff on.
I may some day take your approach and make the backup nas smaller, then exclude media I have in torrents. Though restoring 10 TB over i2p may take a literal year in case of failure.

PS: the backup raid is btrfs raid :)

Yeah, it’s actually a great setup. I’d hope to achieve the same one day by moving all my media to torrents/making torrents, but my personal data alone is still hundreds of GB. I have no idea how you did it.
My music is 3 of GB, and that is just my core playlists often in bad (youtube) bitrate. 13GB for music videos.
My jellyfin db itself is like a GB.
I still have to make lower-res versions of my scanned documents, but I doubt that’ll ever leave the GB, or even tens of GBs (yay bureaucracy).

There’s people that have 200TB Nas’, and people talking about the 10€/month/5TB Hetzner tier. And then there’s op backing up 1GB across 5 cloud provider’s free tiers.

What you got in there? Your light novel library?

op on their recent shopping trip

I’ve used sayboard for a long time and that works well enough I’d say. Offline and with many languages.

Does this one do anything better op? Like punctuation or smthn?