a235

I gave a go with a less popular language. This is a clone of Duolingo from the first glance, and does not place you into a correct level after examination.

Funny vibe code glitche in how an excersirce gives away the answer in a transaction of each question.

Maintainer or curl gave recently a talk on AI slop in security reports, showing this and other examples:

https://youtu.be/6n2eDcRjSsk?si=p5ay52dOhJcgQtxo -- AI slop attacks on the curl project - Daniel Stenberg. Keynote at the FrOSCon 2025 conference, August 16, in Bonn Germany by Daniel Stenberg.

Plus, linked above, his blogpost on the same subject https://daniel.haxx.se/blog/2025/08/18/ai-slop-attacks-on-th...

How does GraphScope graph backend compares to the previous one built by Alibaba - Euler? https://github.com/alibaba/euler/wiki

Done, now I'm waiting for my pre-order to arrive.

These type of experimental interfaces what would make me to consider ordering one! Would love to try now

Wow, another "God-of-the-gaps" fallacy paper. Mixing all the things: math theory, algorithmics, evolution, genetic optimization algorithm in its simplest math model. All to make the conclusion that the authors could not understand those things all together.

On Goldbach’s conjecture: "Hypothesizing that the new axiom is true requires faith; faith in its consistency within the formal system"

And concluding: "However, not all possible phenomes can be produced by genetic algorithm. So it is a matter of faith to believe that every existing phenome [..] could be the result of material things interacting with one another."

FB published a bit on selection of the fact checkers - those are independent journalist organisations.

I'd say contrary - that would be so wrong for FB to verify facts in-house. This way they are more transparent

The sea shepherds are the key in seeding the legal debate - they are documenting and exposing whale hunting as it is.

Without this documentation and tracking, the whalers will continue to hide from anyone's view.

I would argue that the cause of the "Fake news" epidemic is partially due to the social media. Specifically, Information Overload is to blame.

A few decades back we had and now loosing: - Traceable reputation -- which is now lost, and much harder to learn in our uncontrolled, random and sparse feed interactions. - Focus on topic -- social crises were build up, considered, reflected on. Nowadays, we moved into the realm of emotional reactions. Noone is expected to have mental capacity to consume as many pieces of disjoint information as we do today on social media. In an average feed, a president eats a kid, and a cat does a funny trick one after another.

The prime incentives in the current social system is the instant gratification. This new system is easy to abuse, as we are moving further away from analytical to emotional consumption.

Sadly, all ocean studies are underfunded due to a general lack of public interest. "You don't see - you don't care", and pictures like these are the first steps to draw people's attention. Thank you!

Also sad from the AI perspective, this was a great, public human-labeled dataset.

On engine difference, which is critical but has not been pointed out: - Buran had no launching engines, whole Energia rocket has been lost (apart of boosters?) - Shuttle did, thus the most expensive part has been returned back This resulted in a fact that STS launches were expected to be much cheaper [citation needed]

Sound can be recovered partially from visual observations of any vibrations (to some degree, this is your how CD works). So technically, you can 'listen' to bacteria with a compound phase contrast microscopes, which are used to observe living organisms.

With DNA its trickier. First of all, it's too small to be seen using compound (optic) microscopes. While an electron microscope scans a specimen, so either you have to do its with a frequency way higher than any vibrations, or 'focus' on a single place to 'listen' to it (pico stethoscope). There are few further complications, such as you may burn a hole though the specimen, not forgetting that it also requires special treatment to be observed in an electron microscope.

Worth mentioning: Petrel, one of the first frameworks to appear ( https://github.com/AirSage/Petrel ). Worked well for us, few tricks inside, but importantly our code was concise and clear

Ukrainian here.

Ukraine gave up on its nukes in nineties in exchange for military guarantees from US, NATO and Russia. No warheads, no enriched uranium (has been given away by the current president in exchange of nice handshakes with US at the beginning of his term).

Russian leaders are trying to restore its Soviet era, where Ukraine used to be the second most populated country after Russia itself. Clearly, this is a big deal for them. Location wise is not that important, all Baltic countries are way closer to Moscow.

Support this statement! Design that sacrifices content for sake of nice looking white-spaces can be used only on personal pages that nobody reads.

I suspect it would be extremely challenging to observe a meteor from a rain forest, and practically infeasible to find it after. Unless we'll rely on satellite observations

You right, but I still see this map as a catalogue of found meteors, rather than an informative visualisation. To make your observations clear, you need to normalise this heatmap by the population density.

sorry, but this is another population density map

It's not google. This one is based on http://cartodb.com/, however the argument is still valid

wow, the title reflects how epic is the development of journalism standards.

oh, you'r right, it's my mistake

The actual read/write speed should be awfully slow. His screenshot estimates that 322mb will be transferring 18hours!

great starting point, I wish to have a tool that will show me internals of an unknown app.

RISC OS version 5.17

RC5 is actually a released version, so the link above contains the most recent code. More links can be found on he forum: http://www.raspberrypi.org/phpBB3/viewtopic.php?f=55&t=2...

for those that would like to play, raspberrypi.org has a mirror with the latest-to-release RC5: http://downloads.raspberrypi.org/images/riscos/riscos-2012-1...

Whoa, modules with dozen of classes and 2450 lines of code! They definitely don't rely on file-navigation.

Project is great, but I've got a feeling that many smallish components have been already implemented, but here they were rewritten from scratch.

This protocol seems to miss a secure routing of messages inside of its network, insead offering point-to-point interaction. However, this will be required in order to build indeed a decentralised social network. Something that was developed in Tonika project: http://pdos.csail.mit.edu/~petar/5ttt.org/

I've met Ben last week, he is an awesome person to work with. He has a beard, a great passion to work and a neat information management platform polished in tiny details. Strongly recommend to apply until I haven't made my mind, really tempted regardless being a bit away from his needs.

Really neat solution, useful for straightforward things. However, it's quite far from telling us the full truth. Python won't release memory instantly and there will be some free memory that can be reused but won't ever be reused (int arrays?)