"With AI, I can replace 20 software engineers with 1 'prompt engineer'"

A few months later: "plain text editor that was rewritten by AI to be more than that with RCE vulnerability".

(but congratulations to Microsoft for managing to put a remote execution vuln in something that should never have anything 'remote', like notepad)

@RuiSeabra Marques Mendes já disse que não tem espinha, Gouveia e Melo deixou a resposta para outro dia, pelos vistos ainda vai ter de decidir... 🙄

@RuiSeabra
Cotrim até a perder acha que fascismo ou socialismo é igual e não endossa ninguém

Yet another free software web git interface going behind a paywall thanks to all the "AI bots" trying to kill the small web. Just great.

Coisas sobre a Greve Geral que ainda nem toda a gente sabe:

1. A greve não é só para funcionários públicos, é para todos;

2. Aderir à greve é tão simples como não ir trabalhar: não é preciso mais nada, nem avisar a chefia nem nada. Tudo é automático;

3. A única "repercussão negativa" de fazer greve é que o trabalhador não recebe a remuneração desse dia;

4. Fazer greve não é uma "coisa da política", é uma defesa dos direitos dos trabalhadores, que estão em causa neste pacote laboral (fica mais fácil ser despedido, mesmo sem razão; horas de trabalho de graça; etc.);

5. A greve é mesmo para todos: não é preciso um contrato de trabalho, e nem é preciso ser trabalhador - reformados, estudantes, todos podem fazer greve. Uma forma de o fazer é minimizando dar trabalho aos outros, a não ser em casos de urgência.

#GreveGeral

The EU "19.11.2025 COM(2025) 837 final 2025/0360 (COD): Digital Omnibus" Regulation Proposal is out, and it it we have:

• "proposed simplification measures" watering down personal data protection
• There are two terms, "open format" and "formal open standard" with different definitions - none "open enough"

I'm sure there is a lot more to digest from it, so here you go:

https://digital-strategy.ec.europa.eu/en/library/digital-omnibus-regulation-proposal

@lightweight

Portuguese insurance company Tranquilidade migrates desktops to GNU/Linux, reported a costs cut of 80%.

Here's a link to a presentation about it, there is a study case paper about it published somewhere, but probably in Portuguese.

https://www.slideshare.net/slideshow/desktop-linux-na-tranquilidade-opentech-2014/36740025

Maybe @gvlx knows where the paper can be found (or other migration stories with figures).

@RuiSeabra @d3 pensava que ia morrer nessa colina, mas perdi a batalha - e a palavra já está não só no léxico mas também no dicionário:
https://dicionario.priberam.org/encripta%C3%A7%C3%A3o

@RuiSeabra oh pá, sim, mas...
https://ciberlandia.pt/@marado/114837332530345013

A look at the issues list of a project, and you can easilly see how misguided the whole thing is:

https://github.com/eu-digital-identity-wallet/av-app-android-wallet-ui/issues

#EUDIWallet

@vdd @RuiSeabra and yet, I think it is the first time Microsoft admits it, which (to me) is newsworthy.

Acho que não é a primeira vez que falo desta música numa #musiquinta mas se o tema é #anime é difícil que a escolha não seja esta:

#CowboyBebop

https://youtu.be/EL-D9LrFJd4

@RuiSeabra

Eh pá, isto não existe...

"Fizemos duas provas e só foi preciso anular perguntas numa delas? Great Success!"

A #WikiConPT terminou com a inauguração do #WikiLovesEarth 2025, um concurso fotográfico para o qual podem participar durante todo o mês de Junho!

https://wikilovesearth.wikimedia.pt/

O Dia da Liberdade Documental celebra-se já na próxima semana, e em Portugal há conversa pelo Porto!

#documentfreedomday2025 #DocumentFreedomDay #ECTL

Faz hoje #5anos que a #INCM começou a distribuir gratuitamente livros em #PDF:

https://imprensanacional.pt/livros-em-pdf/

@RuiSeabra several times - but I don't think I managed to be as happy about it as you seem to be... looking forward to read about it!

@RuiSeabra welcome to the club 😞

It has been interesting to see so many reflecting at the same time about the current state and the possible future of web browsers - even if it was fired up by some bad news from @mozillaofficial.

Still, it is important to acknowledge what is and has been going on - how a web browser currently became a lot more than just a piece of software implementing a number of open standards, how they are all basically based over one of two existing browser agents (Gecko and Webkit) with the emergence of two upcoming ones (Servo and LibWeb); the difficulties of creating or maintaining a new web browser, or how dependent are all firefox forks (or older, harder derivatives like SeaMonkey) of Mozilla (both Firefox and Gecko).

I'm not trying to get to any point with this text, but rather to highlight that now is the time to have a few conversations around open standards, and in particular web standards. And as it happens, the "annual day" to have discussions of that sort is around the corner: @dff's Document Freedom Day 2025 is already in the 26th of this month. So why aren't there more events on the map? How is your community, user group, etc. going to mark and celebrate this day?

Organize something, and put it on the events map, right this way: https://digitalfreedoms.org/en/dfd

So, #Mozilla wrote something up to "explain" their #Firefox move. Read it in full here:
https://connect.mozilla.org/t5/discussions/information-about-the-new-terms-of-use-and-updated-privacy/m-p/87922/highlight/true#M33705

TLDR? It's bull. Let's dive in?

They focus basically in saying three things:
1) Mozilla needs a license to your data to do things like sending your words to a search engine;
2) there's #AI stuff that needs ToS and Priv. Policy;
3) the acceptable use policy was already in place at Mozilla for the website, etc..

Why bull?
1) if what I type on Firefox (the browser) was sent to Mozilla so they could send it to the search engine, then they'd need access to it. It's not how a browser works or how a privacy-caring browser should ever work. Since what goes on between my Firefox textfields and the browsers I submit to is between us and then, Mozilla should have nothing to do with that data, with or without license.
2) They talk about two kinds of AI: AI services you can access to from Firefox, and AI offline and private abilities within Firefox. For the first, each of those services - if we choose to use them - can and should have their own ToS (distinct from Firefox or any other software used to reach them). AI services running locally should never share anything with anyone, so there's no licenses Mozilla should get.
3) Keep your website's ToS to your website, I don't need it on Firefox - and Mozilla neither.