Reminder of how much we like to criminalize being poor.

Skipping a $3.50 transit fare results in a $425 fine while skipping a $20 parking pass lands you a $70 fine that rich people complained to high heaven about. What justification, other than hating poor people, is there for this not scaling?

Make parking fines $2000 and then let's talk about the fictional "war on cars"

#Toronto
#FareIsFair

I've finished adding subtitles to my 4th DebConf presentation, and with that I wrote a blog post with a summary of everything.

There's some background into curl and wcurl, including the draft wcurl logo and the recording of the Debian curl maintainers BoF, if you want to have a peak at the maintenance of curl on Debian.

DebConf24 was fun!: Security, curl, wcurl, Debian's quality:
samueloph.dev/blog/debconf24-w…

Others have said this, but the Internet Archive's appeals-court loss to Big Publishing is a disaster for everyone but the cartel of companies and a tiny number of A list authors.

The publishers will tolerate libraries only as long as they can control everything about how books can be loaned. If public libraries were being invented today, the cartel would make their core functions illegal.

theverge.com/2024/9/4/24235958…

reshared this

The August 2024 report from the Reproducible Builds project has been published.

"In Android news, the IzzyOnDroid project added 49 new rebuilder recipes and now features 256 total reproducible applications representing 21% of the total offerings in the repository."

Featured are Holger Levsen's DebConf24 talk (and the related LWN article), PostgreSQL removing intermediate build artifacts from tarballs, various improvements to our website, and more.

reproducible-builds.org/report…

#ReproducibleBuilds #IzzyOnDroid

Welcome to the RB family, YTDLnis 🥳

apt.izzysoft.de/packages/com.d…

Was quite a ride to get there (github.com/deniscerri/ytdlnis/…), but thanks to joint efforts of the app team and the IoD team, we got it tackled! :awesome:

YTDLnis lets you download audio/video files from more than 1000 websites, and offers a lot of features.

#reproducibleBuilds #IzzyOnDroid

I am pretty much against biological essentialism and I do my best to ensure my work reflects that.

Unfortunately that is relevant to this review of representation in the 2024 D&D player's handbook which covers quite the range of topics.

It's spicy stuff! 😋 And quite refreshing to hear in honesty.

#5e #Dnd #wyrmworksPublishing

wyrmworkspublishing.com/wotcs-…

This entry was edited (1 year ago)

Firefox will reconsider supporting JPEG XL if they get a Rust implementation:

github.com/mozilla/standards-p…

This is a very good news for web standards:

mastodon.social/@kornel/113078…

and will fix a blocker that is hurting adoption of JPEG XL.

The reference implementation has unfortunately been written in C++ just as browser vendors started looking into migrating away from C++ for security reasons, and saw the C++ codec primarily as a big new attack surface.

Cybersecurity course: 𝗢𝗻𝗹𝗶𝗻𝗲, 𝗵𝗮𝗻𝗱𝘀-𝗼𝗻, 𝗽𝗿𝗮𝗰𝘁𝗶𝗰𝗮𝗹, 𝗮𝗻𝗱 𝗳𝗿𝗲𝗲!
Czech Technical Univeristy's "Introduction to Security" class opens online for free! 14 weeks of deep attacking and defending. Join us and register for free. Starting on Sep 26th. 
cybersecurity.bsy.fel.cvut.cz/
#cybersec #infosec #blueteam #redteam #education #security

Does anybody still remember the scandalous headlines about how AI companies are killing themselves because AI-generated data appears on the web, and training AI on its own data destroys model performance?

Now it turns out that *basically none of this is true*.

This problem happens only when the new model is trained exclusively with the synthetic data, and when the old, human-made data is removed. Nobody actually does this in practice. What people do instead is adding synthetic data to the pool they already have, and then the problem does not occur.

There's more (including a paper) at nitter.poast.org/RylanSchaeffe…

in reply to miki

As I understand it, it's about ratios. Obviously they would still have the human data there, the trick is when the synthetic data, since there is so much of it and it can grow at a rate far faster than human written, approaches or eclipses the volume of human data. Right now, it likely won't happen. A few years down the line in one possible future, when everyone uses AIs on their devices to write social media posts, when article writers have been fired and replaced with LLMs etc? Who's to say.
This entry was edited (1 year ago)

August seems like it just flew by! Catch up on what the Thunderbird Desktop Team has been working on in our latest Developer's Digest. In it, we have:

more Rust! 🦀
updates on Exchange and the global database 🌐
news about the upcoming Calendar rebuild. 🗓️

#Thunderbird #OpenSource #Development

blog.thunderbird.net/2024/09/t…

Aleca reshared this.

Heute ist #WSHD2024!
Der 4. September ist dem Bewusstsein für die Bedeutung sexueller Gesundheit gewidmet. Das Motto für 2024: Positive Beziehungen.

Die World Association for Sexual Health hat den Tag 2010 ins Leben gerufen.

Im Fokus der Arbeit am RKI stehen sexuell übertragbare Infektionen und deren Prävention, doch sexuelle Gesundheit geht weit darüber hinaus. Sie umfasst ebenso Aspekte wie Lebensqualität, Wohlbefinden, Rechte und zwischenmenschliche Beziehungen.

🔗 rki.de/sti

This entry was edited (1 year ago)
in reply to Robert Koch-Institut

@aquasscum
@DenisCOVIDinfoguy

Wie wäre es, wenn ihr hier auf Mastodon #LongCovid und #corona etwas ernster nehmen würdet!

Einmal pro Woche ein Tröt zum Abwassermonitoring wäre schön! Welche Varianten breiten sich gerade aus?

Eure Sprachlosigkeit erzeugt den verheerend falschen Eindruck, Corona sei kein Problem mehr...

Wie viele Infizierte entwickeln nochmal Long Covid? Wer hat ein erhöhtes Risiko Long Covid zu entwickeln?

Schweigen von euch. 🤨😷😬

This entry was edited (1 year ago)