Jonathan Lamothe

Programmer and free/libre software enthusiast. This is my *primary* account on the Fediverse. See my website for an authoritative list of accounts.
[jlamothe.net](https://jlamothe.net )

Pronouns: he/him

Last Notes

2026-10-28 13:03:58 GMT

- reply

@[url=https://mastodon.scot/users/carusb]ChrisR[/url] @[url=https://mamot.fr/users/pluralistic]Cory Doctorow[/url] I'm quite certain that the things we're currently calling AI are often used as a means of dodging said accountability. It's the modern-day "the devil made me do it."

2026-10-26 12:29:40 GMT

- reply

@[url=https://cosocial.ca/users/evan]Evan Prodromou 🇨🇦🇺🇸🇬🇷🇵🇸[/url] I guess my problem here is that I can't trust the LLM not to "hallucinate" the answer. I would feel the need to double check them, negating any benefit I might get because I'd have to do the same amount of work anyway. I can perhaps see some limited utility in using the LLM as a sort of search though, e.g.: "where in this codebase can I find the logic that handles $thing?"
The one point where I [em]really[/em] get hung up is this: if machines can in fact [em]think[/em]—which I doubt for reasons that are outside the scope of this post—would that not make it morally wrong for us to try to bend them to our will? What makes that different from slavery? I bring this up because it seems to be the argument supporting your position that LLM output doesn't constitute a derived work of its training data.

2026-10-26 11:55:37 GMT

- reply

@[url=https://cosocial.ca/users/evan]Evan Prodromou 🇨🇦🇺🇸🇬🇷🇵🇸[/url] This is where I'm going to have to respectfully disagree.
If I'm going to ship code, I need to be able to be able to answer for what that code does. The necessary level of understanding [em]can[/em] be achieved with LLM-generated code, but it involves [em]studging[/em] its output to be sure that every line—every statement—is understood. To me, it seems that the amount of work necessary to achieve that level of understanding negates any boost in prodictivity I may get from using the LLM in the first place, and it certainly doesn't justify the ethical cost.
I've been working on a fork of a large FOSS project lately. For me, the bottleneck hasn't been the actual writing of the code. More than half of my time has been spent [em]studying[/em] the existing code to understand how it works so I can figure out the best way to make my edits. An LLM can't just magically give me that understanding.
Also, when it comes the actual [em]writing[/em] of code, it's easier for me to express what I want in the unambiguous syntax of a programming language rather than the crude tool that is natural language. The latter I only use for communicating with other humans, and only because I lack a better tool for doing so. I'll freely admit that that last point is not universally true for everyone.
Also, I'll concede that the use of third-party libraries muddies the waters in terms of understanding how the code works. This is why I choose not to crucify those who [em]do[/em] use LLMs to write code. I just choose not to participate.

2026-10-03 17:00:02 GMT

- reply

@[url=https://fosstodon.org/users/circfruit]circfruit[/url] @[url=https://pouet.chapril.org/users/nicotentin]Nicolas Vivier[/url] @[url=https://mastodon.online/users/PistonPin]PistonPin[/url] Yeah, the tricky bit is that for routing, the protocol can't be aware of anything beyond the next hop by design.

2026-10-01 19:59:23 GMT

- reply

@[url=https://mastodon.online/users/PistonPin]PistonPin[/url] I still want to get a LoRa radio. So far I've had to restrict myself to internet-based connections.

2026-06-05 20:00:34 GMT

- reply

@[url=https://fedicy.us.to/cy]Cy[/url] @[url=https://drupal.community/users/mikemccaffrey]Mike McCaffrey[/url] As someone who is a harsh critic of AI, and soneone who also suffers from ADHD, I get this. I don't like announcing to my coworkers that I have ADHD, and I fear I'd start annoying people with my repeated questions. People who don't have ADHD simply can not really comprehend the sheer number of times I need to process certain pieces of information before I successfully internalize them.
This is why I prefer things in written form, because I can re-read them as many times as necessary without having to involve anyone else.

2025-12-14 01:55:40 GMT

- reply

@[url=https://mastodon.social/users/wesdym]WesDym[/url] It's not just about the data.

2025-10-21 17:58:36 GMT

It has come to my attention that my favourite jacket is falling apart. I would like to repair it for two reasons:
[ol]
[li]I love this jacket.
[li]I can't afford a new one right now.
[/ol]
I have no opposition to #[url=https://social.jlamothe.net/search?tag=VisibleMending]VisibleMending[/url].
Suggestions?
[url=https://social.jlamothe.net/photos/me/image/866062757168f7c958eb472937160083][img=https://social.jlamothe.net/photo/866062757168f7c958eb472937160083-1.jpeg]Close up of the sleeve of an olive green jacket. The material is frayed and torn just below the seam in the cuff.[/img][/url]
#[url=https://social.jlamothe.net/search?tag=AskFedi]AskFedi[/url] #[url=https://social.jlamothe.net/search?tag=ClothingRepair]ClothingRepair[/url]
https://social.jlamothe.net/photo/866062757168f7c958eb472937160083-0.jpeg

2025-04-25 20:41:53 GMT

- reply

@[url=https://fosstodon.org/users/mykhaylo]🇺🇦 Myke[/url] True enough, though I still want to know how to do it.
Plus, for larger projects it's sometimes easier to separate things out for clarity.

2025-02-05 21:35:41 GMT

- reply

@[url=https://ottawa.place/users/stephanie]Stéphanie[/url] For crying out loud. How long has Unicode been a standard now? At this point, that's just embarrassing for them.

2024-11-02 13:25:10 GMT

- reply

@[url=https://mamot.fr/users/pluralistic]Cory Doctorow[/url] Well said (written?)
This is the exact reason I'm not interested in Bluesky (though I might use Friendica's bridge). I've made that mistake before, and I'm not about to make it again.