For the Irish language course the recordings of native speakers were taken offline in 2023. The AI replacements are nonsensical.

This story is about AI generated courses, not voices, but my post was still (accidentally) on-topic: when they previously used AI to increase volume of content, they were ok with quality being thrown out the window.

The AI generated courses might be low quality, and the original (English) courses might also go downhill because the type of exercises they produce may now be restricted to the type of things that their AI is able to reorganise for other languages. E.g. it might go further in the direction of vocabulary memorisation.

Isn't it that they call it "español" in Latin America, and "castellano" in Spain?

Unfortunately, the voices are really bad.

It's a pity they don't also make available the old courses, with audio from native speakers.

The same way that you implement a language that does look like Common Lisp: macros, specifically reader macros.

FSF's Zoe Kooyman and Krzysztof Siewicz will give a presentation on Sunday 2nd of Feb:

"FSF's criteria for free machine learning applications"

https://ancillary-proxy.atarimworker.io?url=https%3A%2F%2Ffosdem.org%2F2025%2Fschedu...

It'll be streamed. Well worth tuning in for. A recording should be online soon after.

That itty-bitty asterisk is carrying a lot of weight.

I wonder if any betting sites have set up a pool for this project?

FYI, it would be wonderful if true. I just do not believe it, being of the generation that knew "fusion is only 30 years away and always will be".

This is the definition of fascism according to Mussolini:

“Fascism should more appropriately be called Corporatism because it is a merger of state and corporate power."

A large number of Very Smart Committed People Motivated To Do So have been trying to do exactly that for more than half a century, without success. And now my electric bills have an essentially permanent upcharge to pay for the failed nuc plants here in Ohio.

I agree mostly with the comment you are responding to, and believe you are missing the critical point of their comment. They were a bit loose with language "you cand as you see fit", which you note. That said, their comment is directly on point with respect to the fact that increasingly, access to acquired media can be revoked remotely, which I consider to be a major loss for non-owners.

Way back in the days of physical media including print, vinyl, magnetic and optical media, the buyer purchased durable license to content and ownership of the physical media. The license included the ability to play it back (or read it) as many times as they wished, and to sell the the media to another buyer. That license did not include public performance or duplication.

The arrival of digital (but still physical) media made enforcement of the non-copy provision of the law more difficult to enforce.

The arrival of The Internet has produced a new regime in which ownership of a perpetual license to read/play has diminished significantly (too much streaming, increasingly hard to buy media), and the impossibility of resale. My favorite commercial lie: "Own it on digital!"

I am very much concerned that conceptually, "ownership" is being hoarded by owners and cartels, restricting commoners to "rentership". This connects with right-to-repair, and to the new hotness of software as a service.

During the last 12 or so years, the company I work for has used and become dependent on Atlassian Confluence and JIRA, both of which have converted to Cloud service only (we are not large enough for the "enterprise" license). The result is that our family jewels are stored on someone else's computer out on the internet. We are hoping that Atlassian is secure. More than every single other provider of cloud services in the world. They all get hacked, and now our stuff is out there to be grabbed by the first asshole creative enough to get past Atlassian security.

Bottom line: The original conception of ownership of media and license to content is increasingly under control of someone else, such that continued access to content is no longer a trustworthy assumption. Rented rather than owned.

Thanks for the details.

Sounds solvable. Not simple, but sounds like they'll be able to solve it, unless they're trying not to.

Maybe new lists could be downloaded per-domain. If I view one page on a domain, I'll probably view others in the same session. And energy use, there are probably ways to make the plug-ins more efficient - in their own code and by improving the functionality the browser makes available.

For the privacy problem of ad-blockers needing access to all of every webpage you view, this could be fixed by plug-ins being reviewed and verified. Mozilla does something like this.

So, the postponed the disabling of Manifest V2, but can the problems faced by the ad-blocker projects be fixed with some extra time?

I.e. Is this an actual solution? I presume ad-blocking is a bit of a cat-and-mouse, so auto-update filter lists sound crucial for ad-blockers to function. If Chrome blocks that, then they're not allowing useful ad-blockers.

Ad-blockers are the canary in the coal mine of the open web.

US federal law only requires 1.5x hourly wage for hours >40. It's up to each establishment whether they want to pay extra based on shift, and typically weird hours are just part of being a restaurant worker.

Federal law is min. $7.25/hour, and here in NC only $2.13/hour for tipped employees.

I said the same thing tomorrow but the mods deleted it. They've picked their side.

The "robot holding a shotgun" was a plot device. We can't wrap our brains around billions of IoT devices self-organising, so he told that story through the representation of various characters.

That's the Terminator series of films to me. May there be many more!