BodyCulture

It is much more important to discuss the fact that the models still produce wrong results.

This type of product would not have been published in a world where we assumed that computers must produce reliably correct results.

The distributed training collective is a good idea, let’s discuss details:

A) How to prevent malicious injection of bad training data?

B) How to handle copyright violations, will participants be responsible and will they have to pay the creators?

C) Can this create an income stream for content creators and how to avoid abuse, eg feeding with AI content?

So many more, but let’s focus on these before we break things fast because we didn’t think about them.

Is Signal still a trusted company in the industry? They are based in the USA.

Code executed by user can do what user can do. What is the point you are trying to make?

Yes, of course there are still users out there that execute all software under the same account.

Inability to use the tools at hand to protect your data is not a technical issue.

It’s basically the official business model of the US now, have you spent the last two years on a yacht without contact to the world?

It would be great if you hacked your website to be better usable on a smaller mobile device! I need to zoom in to read, but then I also need to scroll sideways because of fixed layout.

You can do both, achieve a cool retro style and publish in a more modern format that supports multiple output devices.

The next class of problems to solve for future generations is how to survive on a hot planet and how to recover from the damage done by reckless lunatic profiteering assholes.

This will make it much easier to trap people into installing it without any isolation just to their primary user account or even make it run with admin privileges by default.

With the recent security related events we already see that it seems to be wide spread practice that many operators avoid the hassle of isolating their agents.

It is just an incredible cultural experience to see how quick stupidity gets accepted as a behavioural default.

Very good point, thanks for confirming that! What is the reasoning behind the Go design choices you described?

Everything would be perfect if only you were CEO of both companies!

Let’s go there and destroy everything! Will be lots of fun and make us a lot of money!

What is the state of the art regarding open source collaborative desktop publishing? Do exist any usable Wiki to Scribus workflows?

Most important piece of information is in the linked Frontiers article:

   However, the overall capability of the chatbot to fully meet user needs received a lower average score (3.1/5.0), highlighting the need for further improvements.

Also there is still the problem of hallucinations, as we see in the „Evaluation“ paragraph:

   Live traffic evaluations are essential for monitoring system behavior, identifying potential issues like hallucinations in production, and understanding performance on diverse live queries.

This are quite devastating results. This is a system for scientific research on medicines and mediocrity and hallucinations will kill people.

Would be interesting to know how much money was flushed down the toilet with these experts.

This is not a problem at all as most Apple computers come with plenty of RAM and lots of disk space! We are so lucky that Apple engineers always think so differently into the future!

This is not a question of ideology or personal beliefs. It’s just a matter of fact that AI systems, especially LLMs, are not capable of producing reliable results.

This is a fundamental issue and for many uses we still need the old school way of computing that produces reliable results. It doesn’t happen accidentally, you know, a lot of resources went into this. AI just isn’t there yet and probably won’t be there anytime soon, no matter how much compute we throw at it.

Funny how they are publishing free advertising for X, fb, yt etc. on their website. This is telling more about the situation than any speaker at the conference will be able to do.

From my pov and after using AI a lot to better understand its usefulness I see this discussion as totally misleading and irrelevant.

The elephant in the room is that even the best frontier models still produce errors.

This is a fundamental issue and should be discussed instead of some esoteric pseudo-scientific nonsense.

If the current systems can not produce reliable results they are useless. The promise of actual usefulness needs to be fulfilled in a very near future.

Why are your queries slow?

Is your comment referring to this project specifically?

Because the docs say:

  PgQue avoids that whole class of problems. It uses snapshot-based batching and TRUNCATE-based table rotation instead of per-row deletion.

Would be great if you could specify if you had problems with the exact implementation linked by op or if you did write about a different thing, thanks!

Be careful with any serious project, this software most certainly will crash and destroy your work. It crashes since many years and developers do not seem to care or are not able to understand how important stability for media creation software really is. Especially small and independent artists should absolutely avoid any software that introduces additional risk of project failure as one such crash scenario at an advanced project state has a high potential of total destruction.

Choose wisely! Resolve is available for very little money and not only a much safer choice, but you will also learn to use an industry standard tool and might be able to monetise that skill one day.

Kdenlive is a hobbiest project and is probably still ok for occasionally splitting a downloaded YouTube video or converting your OBS recordings, but never should you remotely think about using it for a project where you need to rely on your tools.

The developers are not warning you enough, instead still trying to market this software as kind of a serious competitor to pro software, so I do that as a service for the aspiring video editor, taking your downvotes proudly as the price honest people have to pay.

Yes, obviously I write from experience.

What are you using instead of pandas? Thanks!

No.

Can we study this second pipeline? Is it open so we can understand how it works? Did not find any hints about it in the article, unfortunately.

Thank you very much for sharing your research results!

I really appreciate your work and even more that you took time and risk exposing your findings, I wish more people did this.

Distribution of the content as static html or in any other format is a very tiny aspect of managing content and mostly a solved problem for any CMS nowadays. Focusing on that minimal aspect seems grotesque as there are much bigger challenges in making potentially large amounts of content actually manageable by a potentially very heterogeneous group of content creators with varying skills, responsibilities and relationships.

We need more aggressive laws to prevent privacy destroying platforms. Every person who creates a website or platform that advertises any kind of private communication but does not fully encrypt user data must go to jail. This cancer needs to be stopped.

Seems like it still has no official support for any kind of disk encryption, so you are on your own if you fiddle that in somehow and things may break. Such a beautiful, peaceful world where disk encryption is not needed!

You underestimate the value of this piece of information taken at different times. It can be enough to know in which country a person was yesterday or is today.

Thank you very much!

Wound be great if you posted the URL to the relevant documentation for this… I guess there must be some docs about these delicate details? Thank you very much!