Parienve @Parienve@social.cleverlibre.org
- alt
- Parienve@mas.to
- pronouns/pronomines
- en: they/them/themself misc: ¯\\\_(ツ)\_/¯
en: Mostly tech, but not entirely. Privacy is a human right.
ia: Principalmente technologia, ma non in toto. Privacitate es un derecto human.
Joined Feb 2020
Parienve
boosted
Parienve
boosted
Reaching the Unix Philosophy's Logical Extreme with Webassembly
https://xeiaso.net/talks/unix-philosophy-logical-extreme-wasm
Discussions: https://discu.eu/q/https://xeiaso.net/talks/unix-philosophy-logical-extreme-wasm
Parienve
boosted
jank development update - Object model results
https://jank-lang.org/blog/2023-08-26-object-model/
Discussions: https://discu.eu/q/https://jank-lang.org/blog/2023-08-26-object-model/
Parienve
boosted
"No serious scholar or scientist in their right mind would want LLMs to produce their texts; and hence, also no student pursuing an academic education would want to do so." - Iris van Rooij
A short, but very good blog post by @Iris - https://irisvanrooijcogsci.com/2022/12/29/against-automated-plagiarism/
I very much agree with Professor van Rooij's argument in her blog. How can we, at the same time, teach 1) that manual plagiarism is wrong but 2) automated plagiarism is okay?
Parienve
boosted
I have been testing @organicmaps in hiking this week besides the paper map. It is working smoothly. At least for me, it found existing hiking routes and wickets. Advantage is it is showing estimated time and distance to destination and for route planning also the elevation profile.
Parienve
boosted
Recommendation systems are useful, but sometimes we just want to discover new things by ourselves. In this article I present 6 alternatives to "the algorithm".
https://blog.ddavo.me/posts/death-to-the-algorithm
#algorithm #youtube #radio #ethics #blog #recommendersystems #AI #datagovernance
@SecureOwl
This metric dilutes the influence of 1-star and 5-star ratings. I wonder, how does Amazon benefit from this?
@urusan
I believe it is possible in principle to construct a valid proof whose construction relies on at least one step that uses AI, and that this would involve a secondary proof of the validity of the AI-dependent step.
I voted no, however, because I conjecture that, for any proof with an AI-dependent step that is justified by a secondary proof, there exists an alternate formulation of the overall proof without the AI-dependent step.
Parienve
boosted
The #Forgejo summer update was published ✨ It is a high level overview of the project activities.
Forgejo v1.20, was published and was 100% built with Forgejo Actions. Codeberg was upgraded 10 August and hit an unexpected issue. It was quickly resolved and the fix will be integrated in the next release. Meanwhile development continued on the code, the website, the documentaion etc. All signs of a healthy project that needs your help to keep going in the long run. 🙏
Parienve
boosted
@reginagrogan
Me: Bank, do you support 2FA?
Bank: no, what is that?
Me: Two-factor authentication. Like using an app or a yubikey.
Bank: ah, no, we just added SMS auth
Me: but it's well-documented that SMS is not as secure, aren't you a bank?
Bank: ...
Parienve
boosted
It occurred to me today that the social value of the open-source work I do in my free time has probably been an order of magnitude more useful to the world than everything I've ever done as a paid employee. Needing to seek a wage almost certainly makes me a less productive member of society than I would otherwise be.
bad musical notation, even worse pun
@gsllc
If the second staff had a treble clef, it would spell GEEK, but somebody decided to use a bass clef without moving the notes. You might say that they flouted proper notation.
Alt text
@gsllc
A t-shirt with two musical staves, and underneath, the text "IT'S A BAND THING". The top staff has a treble clef with musical notes and a letter that spell BAND. The bottom staff has a bass clef with musical notes and a letter that spell BGGK.
@gsllc
I'd probably double down on the irony and add Essos somewhere to the west of Middle Earth.
Parienve
boosted
im BEGGING YOU, please for the love of god, open-source projects, HAVE A FORUM STOP USING DISCORD SERVERS
i cant get support for a project, because there is no documentation and the discord invite is dead!
AI ethics poll
@urusan
The "other four" are:
1. "Never" + "No higher standard" -> "Humans are never ethical."
Bleak, indeed. You'd probably have to either reject the entire concept of ethics as ill-formed, or accept a standard for ethical behavior so high that, in practice, no human ever reaches it at any time.
2. "Specific internal rules" + "No higher standard" -> "Humans are ethical only when they follow specific internal rules."
Some people would agree with this.
3. "Non-profit" + "No higher standards" -> "Humans are only ethical when they pursue non-profit purposes."
Some people would agree with this.
4. "Specific internal rules" + "Non-profit" + "No higher standards" -> "Humans are only ethical when following specific internal rules for non-profit purposes."
The more ethically-demanding union of the previous two.
@Linux_in_a_Bit
In #Fedilab at least, you can go to their profile, long press the "Messages" tab (the lower one), and uncheck "Show boosts" and "Show replies".
AI ethics poll
@urusan
I think this poll would be better if it allowed multiple choices. Out of the 16 combinations, I count 9 logically consistent positions.
- alt
- Parienve@mas.to
- pronouns/pronomines
- en: they/them/themself misc: ¯\\\_(ツ)\_/¯
en: Mostly tech, but not entirely. Privacy is a human right.
ia: Principalmente technologia, ma non in toto. Privacitate es un derecto human.
Joined Feb 2020
