I never understood why there is no interactive Help program like there was in the "old days" when CHM files on Windows 95/98/XP were a thing. These CHM files and the interactiveness are heavily underrated, and they were some really good documentation, especially the ones from IDEs and compiler suites.
Today I wish there was something like this but made for tutorials and wizards. If someone presses "Help" they should not have to go online on your website just to literally never find any help for their problems.
We are in the golden age of LLMs, yet nobody uses LLMs to explore and discover locally hosted knowledge bases ... which are in my opinion the single most useful use case of them. You could build such a great UX with it.
For example, I'm selfhosting a lot of archived wikis via a kiwix server. Devdocs, wikipedia, dev and cyber related wikis. Having an LLM assistant running on those locally was probably the best improvement for my learning experience. And the workflow is integrated into my custom New Tab page, it's literally a search field on my homepage of the browser, so it's always accessible.
The real question is why Anthropic was able to use DMCA takedown requests "in good faith" against the Claude leaks when their own CTO claimed it is a 100% slopcoded codebase, and they themselves argue that all LLM generated code is transformed enough to not be copyrightable. Which they have to state without being able to turn back because they violated millions of book and software licenses during training.
You can profit from getting away with lying to judges.
A judge isn't involved, anyway. The leaker would have to take you to court and then prove that your request was in bad faith and that they didn't infringe copyright.
Competent programmers understand how to tell the computer what needs to happen. Really good programmers understand how the computer executed the code, and take advantage of it - they know about speculative execution and cache prefetching. Competent lawyers know what the law says. Really good lawyers understand how the law is executed, and take advantage of it - they know when it won't be enforced.
Talking about ICQ with zero screenshots is like talking about Hamachi without talking about LAN parties and how games were played at the time.
Pretty much all articles are just slop-text. Not even talking about alternatives or what has been done in the meantime. For example, ICQ led to AIM and Trillian, which led to pidgin/libpurple, then to jabber/xmpp etc.
On our LAN parties at the time, everyone started to use Trillian because of their support for the Bonjour protocol which used multicast DNS locally. That way you could easily "group teams" together, because the hostnames / IPs were also visible and you could add them then to the Hamachi groups where the games happened.
Hamachi was an amazing tool before they got bought, it made networking so easy for everyone; and you didn't even need to know what an IP was.
Shameless drop: My own agentic environment is also using a summarizer to sum up agent histories when they overflow in their context windows. Additionally, all short lived agents are based on requirements (WIP) as a centralized point for code, unit tests, and architecture.
(Everything is tailored to Go as a language)
Works pretty good so far, the user only interacts with the planner. I'm working atm on the requirements to have a spec driven workflow. Web UI is the most polished atm because of ability to have agent tabs on the side for better overview.
...because Alzheimer is a dormant side effect of a virus, not of a messenger chemical. But that doesn't go well in studies and "self populism" of what funded research wanted to hear.
If you study effects and not causes due to lack of measurements for reproducibility in any field of research, that's what comes out.
Also check out how the new and promising correlation started by observing the Wales eligibility for mandatory shingles vaccination during an outbreak and the effect on that test group when it comes to alzheimer or dementia in their old age.
Note that shingles (herpes zoster) virus is a dormant virus for decades, and it's not really treated because of that.
Also note that this was only discovered because people died and their data set was publicized because of that, which I hope that can happen in an anonymous way due to it being invaluable for medical research.
That's why I was emphasizing "one way" and "can begin".
Clearly viral infection may be a distal factor but it is not the proximate cause of Alzheimer's... which is quite conclusively due to protein aggregates.
Yes, exactly. Whether it APOE mutation, traumatic injury, metabolic dysfunction, Herpes infection, or some other inflammation that upregulates kinases, you end up with hyperphosphorylated tau which in turn forms the neurofibrillary tangles that destroy nutrient transport in neurons and eventually kills them.
"Neuron death by protein aggregates" is the best way to define Alzheimer's. Anything more specific refers to only one of those aforementioned causes; anything less is just "brain don't work no more."
These clickbait headlines are so frustrating, especially since the article itself explains the tau mechanism and all the progress that has been made in understanding the disease.
It very much has been. Which is why it's tiring to see these articles (and HN comments) portraying it otherwise. Search HN for Alzheimer's and you will find lots of updates on the research and lots of the same uninformed comment threads.
... which kind of points to the indicator that "Alzheimer != Alzheimer", implying that too many diseases with the same side effects are categorized together?
A lot of virological and parasitical components have historically been wrongly associated with genetic markers, too. Toxoplasmosis parasite comes to mind.
I am not a researcher in this field. I have read a lot about APOE4, and nearly everything I have seen about APOE4 makes it out to be the clearest of all predictors for Alzheimer's. I get that you're upset that I posted like I did, but I am not trolling. The point was that genetic causes are not an exceptional predictor for Alzheimer's. I had a hard time seeing how the best established predictor for Alzheimer's validates there being too many things labeled as Alzheimer's. If I had posted with a focus on some edge case, then it probably would have made more sense to me, but as I read it, it really seemed like a stretch.
Dropping this thread now. I am fine with being wrong, but my brief posting was not trolling.
In my opinion current research should focus on revisiting older concepts to figure out if they can be applied to transformers.
Transformers are superior "database" encodings as the hype about LLMs points out, but there have been promising ML models that were focusing on memory parts for their niche use cases, which could be promising concepts if we could make them work with attention matrixes and/or use the frequency projection idea on their neuron weights.
The way RNNs evolved to LSTMs, GRUs, and eventually DNCs was pretty interesting to me. In my own implementations and use cases I wasn't able to reproduce Deepmind's claims in the DNC memory related parts. Back at the time the "seeking heads" idea of attention matrixes wasn't there yet, maybe there's a way to build better read/write/access/etc gates now.
> I'd be genuinely curious what you could switch to that still has MV2 because, AFAIK, Firefox is the last holdout.
My last hope is ladybird right now, I don't use Firefox or Chrome as my main browsers anymore, and use them only within temporary sandboxes. Without history, without cookies, without logins for the most part.
For the most part, it doesn't. It's not a consumer ready browser, but a pretty nice little rendering engine. If you use ladybird as bindings, it's a bit unstable right now because they are refactoring a lot of parts in the codebase.
I built my own tools on top of it, mostly to use internet websites and selfhosted kiwix archives with my local agentic env.
I guess what I am saying is that I don't have a primary browser anymore. Not a browser where I just can trust it that it doesn't do shit with my data. Being able to selfhost kiwix is a superb internet experience if you build your own search dashboard for it, I can fully recommend it.
Have to merge my things upstream with ZIMdex when I have the time (probably around June).
It seems to me that --unless you really, strictly compartimentalize your browser usage--, using multiple browsers will only supply your data to more parties.
Oof indeed. Now I know that Kling is indeed open towards some alt right positions, but I really wouldn't call him a fascist for that. Conservative probably, but conflating conservative positions with fascism is probably not helpful in the fight against the real fascists.
But also oof to .. some other items there from the blog. Apparently rsync is now banned from the list of acceptable software, because they do not ban LLM's completely?
Isn't this blog post more evidence that drewdevault became an extreme leftist?
I mean he's basically going off a checklist of leftist stereotypes here and trying to check as many of them as possible.
Meanwhile the other guy he's criticising is literally just a standard right-wing conservative, not far right, not alt right, just the regular kind. The far right I've seen is basically beyond the idea of being merely anti-immigration, they demand ICE style mass deportations immediately and in every country.
If both of them met in a bar through sheer coincidence, I'd expect drewdevault to start the fight.
Sorry, but painting these people as "standard right-wing" is just evidence for the shifting of the Overton window further to the right. White replacement theory and expressing support for an alt-right ideologue who manipulated people with bad faith, dishonest and downright monstrous arguments is not "standard right-wing".
Charlie Kirk was for mass deportation. He didn't even hide it. He said it openly. How do you come off saying that these people aren't far-right or alt-right when they are unabashedly so?
Well, the overtone window certainly changed, but ... I judge a bit different here.
"expressing support for an alt-right ideologue"
This is what Kling actually said:
"RIP Charlie Kirk
I hope many more debate nerds carry on his quest to engage young people with words, not fists."
I also support fighting with words, not fists. I do not support his ideology at all and would have loved to debate him openly, but the concept of murdering someone for having the wrong opinion is disturbing to me, so I agree with Kling here.
And about "white replacement"
"'White males are actively discriminated against in tech.
It’s an open secret of Silicon Valley.'
One of the last meetings I attended before leaving Apple (in 2017) was management asking us to “keep the corporate diversity targets in mind” when interviewing potential new hires.
The phrasing was careful, but the implication was pretty clear.
I knew in my heart this wasn’t wholesome, but I was too scared to rock the boat at the time."
He said whites were discriminated for being white. Not replaced. That is not really the same to me.
>White replacement theory and expressing support for an alt-right ideologue who manipulated people with bad faith, dishonest and downright monstrous arguments is not "standard right-wing".
It is now. That's what the shifting of the Overton Window and normalization of right-wing ideology does. These aren't fringe beliefs anymore, they're commonly held, mainstream right-wing views. They're policy within the US government. Charlie Kirk was treated as a martyr and a hero by the administration. He was treated with more dignity and respect than war veterans. The DHS posts memes about mass deportation.
The "far right" and "alt-right" no longer exist. Those labels are no longer useful and no longer describe reality.
Today I wish there was something like this but made for tutorials and wizards. If someone presses "Help" they should not have to go online on your website just to literally never find any help for their problems.
We are in the golden age of LLMs, yet nobody uses LLMs to explore and discover locally hosted knowledge bases ... which are in my opinion the single most useful use case of them. You could build such a great UX with it.
For example, I'm selfhosting a lot of archived wikis via a kiwix server. Devdocs, wikipedia, dev and cyber related wikis. Having an LLM assistant running on those locally was probably the best improvement for my learning experience. And the workflow is integrated into my custom New Tab page, it's literally a search field on my homepage of the browser, so it's always accessible.
reply